***************************************** Setting OMP_NUM_THREADS environment variable for each process to be 1 in default, to avoid your system being overloaded, please further tune the variable for optimal performance in your application as needed. ***************************************** ================ Configuration ================ mode : DDP x8 (torchrun, NCCL) cuda device : 0 (NVIDIA H200) d_model : 384 n_layers : 12 q_heads/kv_heads: 6 / 2 (G=3) head_dim : 64 ffn_hidden : 1024 (SwiGLU) context_length : 4096 rope_theta : 500000 (NeoX half-split) vocab_size : 50304 weight_tying : OFF B x T : <= 6 x 4096 (micro-batch set per stage) total tokens : 14,000,000,000 (14,000,062,464 in 26703 steps) batch schedule : by tokens consumed, world_size=8; LR x sqrt(batch / 524,288) from tokens target batch micro x accum lr x steps 0 524,288 524,288 4 x 4 1.0000 26703 optimizer : MUON (hidden 2D) + AdamW (embed/head/1D) max_lr / min_lr : 0.001 / 0.0001 (AdamW group) muon lr / min : 0.001 / 0.0001 (adjust_lr_fn=match_rms_adamw) muon momentum : 0.95 | NS steps 5 warmup / steps : 500 / 26703 WSD stable_steps: 25935 (decay tail=268) target losses : [2.6906, 2.55404, 2.5054, 2.48068, 2.45006, 2.43759] (ema beta 0.99, checkpointed on crossing, never pruned) interval ckpts : every 1336 steps (keep newest 20) precision : bf16-params seed / determ. : 1234 / strict rms+rope : fused | compile False | act-ckpt False | loss_chunks 1 =============================================== Initializing model on CUDA device 0... /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) Number of parameters: 57518976 /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) /home/operator/run_oct16/train_57M_dynamic.py:1788: FutureWarning: `broadcast_buffers` is deprecated. Use `forward_sync_buffers` instead. IMPORTANT: unlike `broadcast_buffers=False`, `forward_sync_buffers=False` still syncs buffers at init. If you rely on buffers NOT being synced at init (e.g. rank-local buffers), keep using `broadcast_buffers=False` until `init_sync_buffers` is available. model = DDP(raw_model, device_ids=[ddp_local_rank], broadcast_buffers=False) muon params: 18874368 | adamw decayed: 38633472, non-decayed: 11136 [ddp] static footprint (per GPU: params + grads + optimizer state) parameters : 0.14 GiB gradients : 0.14 GiB fp32 master weights : 0.14 GiB fp32 grad landing bufs : 0.14 GiB Muon momentum : 0.07 GiB Adam exp_avg + exp_avg_sq: 0.29 GiB static total : 0.93 GiB found 240 shards for split train found 1 shards for split val Starting training... [val] step 0 | val_loss 10.890124 [batch] step 0 | 0 tokens | batch 524,288 tokens/step (4 x 4 micro-steps per rank) | LR x 1.0000 (sqrt of batch / 524,288) step 0 | train 10.884307 | val 10.890124 | tok/s 712528.95 | norm 8.6505 | dt 735.81 ms | lr 2.0000e-06 | mlr 2.0000e-06 | mem 13.37/17.79 GiB step 1 | train 10.882978 | val 10.890124 | tok/s 1727537.13 | norm 8.8720 | dt 303.49 ms | lr 4.0000e-06 | mlr 4.0000e-06 | mem 13.73/16.12 GiB step 2 | train 10.882508 | val 10.890124 | tok/s 1925246.11 | norm 8.7588 | dt 272.32 ms | lr 6.0000e-06 | mlr 6.0000e-06 | mem 13.73/16.12 GiB step 3 | train 10.878126 | val 10.890124 | tok/s 2569546.39 | norm 8.9162 | dt 204.04 ms | lr 8.0000e-06 | mlr 8.0000e-06 | mem 13.73/16.12 GiB step 4 | train 10.878796 | val 10.890124 | tok/s 2623868.05 | norm 9.7414 | dt 199.81 ms | lr 1.0000e-05 | mlr 1.0000e-05 | mem 13.73/16.12 GiB step 5 | train 10.871430 | val 10.890124 | tok/s 2613743.98 | norm 8.5752 | dt 200.59 ms | lr 1.2000e-05 | mlr 1.2000e-05 | mem 13.73/16.12 GiB step 6 | train 10.868461 | val 10.890124 | tok/s 2608558.74 | norm 8.6891 | dt 200.99 ms | lr 1.4000e-05 | mlr 1.4000e-05 | mem 13.73/16.12 GiB step 7 | train 10.863358 | val 10.890124 | tok/s 2587269.59 | norm 8.5551 | dt 202.64 ms | lr 1.6000e-05 | mlr 1.6000e-05 | mem 13.73/16.12 GiB step 8 | train 10.843767 | val 10.890124 | tok/s 2625310.94 | norm 9.0113 | dt 199.71 ms | lr 1.8000e-05 | mlr 1.8000e-05 | mem 13.73/16.12 GiB step 9 | train 10.837104 | val 10.890124 | tok/s 2623536.03 | norm 8.5108 | dt 199.84 ms | lr 2.0000e-05 | mlr 2.0000e-05 | mem 13.73/16.12 GiB step 10 | train 10.816265 | val 10.890124 | tok/s 2625432.61 | norm 9.2901 | dt 199.70 ms | lr 2.2000e-05 | mlr 2.2000e-05 | mem 13.73/16.12 GiB step 11 | train 10.804220 | val 10.890124 | tok/s 2524604.50 | norm 8.9428 | dt 207.67 ms | lr 2.4000e-05 | mlr 2.4000e-05 | mem 13.73/16.12 GiB step 12 | train 10.789590 | val 10.890124 | tok/s 2626487.77 | norm 8.3072 | dt 199.62 ms | lr 2.6000e-05 | mlr 2.6000e-05 | mem 13.73/16.12 GiB step 13 | train 10.768749 | val 10.890124 | tok/s 2630312.52 | norm 8.3749 | dt 199.33 ms | lr 2.8000e-05 | mlr 2.8000e-05 | mem 13.73/16.12 GiB step 14 | train 10.743412 | val 10.890124 | tok/s 2479862.47 | norm 8.7304 | dt 211.42 ms | lr 3.0000e-05 | mlr 3.0000e-05 | mem 13.73/16.12 GiB step 15 | train 10.726479 | val 10.890124 | tok/s 2620049.51 | norm 8.1716 | dt 200.11 ms | lr 3.2000e-05 | mlr 3.2000e-05 | mem 13.73/16.12 GiB step 16 | train 10.711257 | val 10.890124 | tok/s 2616882.35 | norm 8.1185 | dt 200.35 ms | lr 3.4000e-05 | mlr 3.4000e-05 | mem 13.73/16.12 GiB step 17 | train 10.676197 | val 10.890124 | tok/s 2627624.79 | norm 8.2753 | dt 199.53 ms | lr 3.6000e-05 | mlr 3.6000e-05 | mem 13.73/16.12 GiB step 18 | train 10.647754 | val 10.890124 | tok/s 2616978.26 | norm 8.4019 | dt 200.34 ms | lr 3.8000e-05 | mlr 3.8000e-05 | mem 13.73/16.12 GiB step 19 | train 10.621634 | val 10.890124 | tok/s 2471951.63 | norm 8.2809 | dt 212.09 ms | lr 4.0000e-05 | mlr 4.0000e-05 | mem 13.73/16.12 GiB step 20 | train 10.600396 | val 10.890124 | tok/s 2563452.69 | norm 7.8947 | dt 204.52 ms | lr 4.2000e-05 | mlr 4.2000e-05 | mem 13.73/16.12 GiB step 21 | train 10.569905 | val 10.890124 | tok/s 2618054.53 | norm 7.7307 | dt 200.26 ms | lr 4.4000e-05 | mlr 4.4000e-05 | mem 13.73/16.12 GiB step 22 | train 10.538155 | val 10.890124 | tok/s 2620326.87 | norm 7.7783 | dt 200.08 ms | lr 4.6000e-05 | mlr 4.6000e-05 | mem 13.73/16.12 GiB step 23 | train 10.520214 | val 10.890124 | tok/s 2621258.88 | norm 7.4234 | dt 200.01 ms | lr 4.8000e-05 | mlr 4.8000e-05 | mem 13.73/16.12 GiB step 24 | train 10.476865 | val 10.890124 | tok/s 2623455.33 | norm 7.3799 | dt 199.85 ms | lr 5.0000e-05 | mlr 5.0000e-05 | mem 13.73/16.12 GiB step 25 | train 10.445904 | val 10.890124 | tok/s 2629825.78 | norm 7.2532 | dt 199.36 ms | lr 5.2000e-05 | mlr 5.2000e-05 | mem 13.73/16.12 GiB step 26 | train 10.402387 | val 10.890124 | tok/s 2628660.00 | norm 7.4086 | dt 199.45 ms | lr 5.4000e-05 | mlr 5.4000e-05 | mem 13.73/16.12 GiB step 27 | train 10.370537 | val 10.890124 | tok/s 2611162.40 | norm 6.9546 | dt 200.79 ms | lr 5.6000e-05 | mlr 5.6000e-05 | mem 13.73/16.12 GiB step 28 | train 10.339090 | val 10.890124 | tok/s 2626306.53 | norm 6.7543 | dt 199.63 ms | lr 5.8000e-05 | mlr 5.8000e-05 | mem 13.73/16.12 GiB step 29 | train 10.303331 | val 10.890124 | tok/s 2627057.40 | norm 6.6173 | dt 199.57 ms | lr 6.0000e-05 | mlr 6.0000e-05 | mem 13.73/16.12 GiB step 30 | train 10.266535 | val 10.890124 | tok/s 2634547.60 | norm 6.3741 | dt 199.00 ms | lr 6.2000e-05 | mlr 6.2000e-05 | mem 13.73/16.12 GiB step 31 | train 10.229484 | val 10.890124 | tok/s 2628894.81 | norm 6.1840 | dt 199.43 ms | lr 6.4000e-05 | mlr 6.4000e-05 | mem 13.73/16.12 GiB step 32 | train 10.187298 | val 10.890124 | tok/s 2633477.84 | norm 6.0045 | dt 199.09 ms | lr 6.6000e-05 | mlr 6.6000e-05 | mem 13.73/16.12 GiB step 33 | train 10.144039 | val 10.890124 | tok/s 2587082.98 | norm 5.8420 | dt 202.66 ms | lr 6.8000e-05 | mlr 6.8000e-05 | mem 13.73/16.12 GiB step 34 | train 10.098894 | val 10.890124 | tok/s 2621371.36 | norm 6.1463 | dt 200.01 ms | lr 7.0000e-05 | mlr 7.0000e-05 | mem 13.73/16.12 GiB step 35 | train 10.065420 | val 10.890124 | tok/s 2617334.37 | norm 5.5384 | dt 200.31 ms | lr 7.2000e-05 | mlr 7.2000e-05 | mem 13.73/16.12 GiB step 36 | train 10.036838 | val 10.890124 | tok/s 2625135.76 | norm 5.2102 | dt 199.72 ms | lr 7.4000e-05 | mlr 7.4000e-05 | mem 13.73/16.12 GiB step 37 | train 9.998035 | val 10.890124 | tok/s 2625592.06 | norm 5.1245 | dt 199.68 ms | lr 7.6000e-05 | mlr 7.6000e-05 | mem 13.73/16.12 GiB step 38 | train 9.940694 | val 10.890124 | tok/s 2622406.67 | norm 5.1103 | dt 199.93 ms | lr 7.8000e-05 | mlr 7.8000e-05 | mem 13.73/16.12 GiB step 39 | train 9.893492 | val 10.890124 | tok/s 2617677.31 | norm 5.2119 | dt 200.29 ms | lr 8.0000e-05 | mlr 8.0000e-05 | mem 13.73/16.12 GiB step 40 | train 9.867300 | val 10.890124 | tok/s 2611466.14 | norm 4.6559 | dt 200.76 ms | lr 8.2000e-05 | mlr 8.2000e-05 | mem 13.73/16.12 GiB step 41 | train 9.815537 | val 10.890124 | tok/s 2617265.45 | norm 4.6260 | dt 200.32 ms | lr 8.4000e-05 | mlr 8.4000e-05 | mem 13.73/16.12 GiB step 42 | train 9.778620 | val 10.890124 | tok/s 2626900.24 | norm 4.4274 | dt 199.58 ms | lr 8.6000e-05 | mlr 8.6000e-05 | mem 13.73/16.12 GiB step 43 | train 9.737126 | val 10.890124 | tok/s 2619446.03 | norm 4.2462 | dt 200.15 ms | lr 8.8000e-05 | mlr 8.8000e-05 | mem 13.73/16.12 GiB step 44 | train 9.689110 | val 10.890124 | tok/s 2625589.13 | norm 4.1583 | dt 199.68 ms | lr 9.0000e-05 | mlr 9.0000e-05 | mem 13.73/16.12 GiB step 45 | train 9.665865 | val 10.890124 | tok/s 2627875.06 | norm 3.9397 | dt 199.51 ms | lr 9.2000e-05 | mlr 9.2000e-05 | mem 13.73/16.12 GiB step 46 | train 9.593012 | val 10.890124 | tok/s 2622158.96 | norm 3.8442 | dt 199.95 ms | lr 9.4000e-05 | mlr 9.4000e-05 | mem 13.73/16.12 GiB step 47 | train 9.550329 | val 10.890124 | tok/s 2624575.23 | norm 3.6498 | dt 199.76 ms | lr 9.6000e-05 | mlr 9.6000e-05 | mem 13.73/16.12 GiB step 48 | train 9.501406 | val 10.890124 | tok/s 2620404.64 | norm 3.5209 | dt 200.08 ms | lr 9.8000e-05 | mlr 9.8000e-05 | mem 13.73/16.12 GiB step 49 | train 9.453571 | val 10.890124 | tok/s 2622734.50 | norm 3.4135 | dt 199.90 ms | lr 1.0000e-04 | mlr 1.0000e-04 | mem 13.73/16.12 GiB step 50 | train 9.399961 | val 10.890124 | tok/s 2631560.44 | norm 3.2176 | dt 199.23 ms | lr 1.0200e-04 | mlr 1.0200e-04 | mem 13.73/16.12 GiB step 51 | train 9.350728 | val 10.890124 | tok/s 2628144.42 | norm 3.1442 | dt 199.49 ms | lr 1.0400e-04 | mlr 1.0400e-04 | mem 13.73/16.12 GiB step 52 | train 9.318025 | val 10.890124 | tok/s 2627591.26 | norm 3.0016 | dt 199.53 ms | lr 1.0600e-04 | mlr 1.0600e-04 | mem 13.73/16.12 GiB step 53 | train 9.239517 | val 10.890124 | tok/s 2618690.61 | norm 2.8843 | dt 200.21 ms | lr 1.0800e-04 | mlr 1.0800e-04 | mem 13.73/16.12 GiB step 54 | train 9.209435 | val 10.890124 | tok/s 2618936.32 | norm 2.7251 | dt 200.19 ms | lr 1.1000e-04 | mlr 1.1000e-04 | mem 13.73/16.12 GiB step 55 | train 9.129656 | val 10.890124 | tok/s 2622113.98 | norm 2.6933 | dt 199.95 ms | lr 1.1200e-04 | mlr 1.1200e-04 | mem 13.73/16.12 GiB step 56 | train 9.092917 | val 10.890124 | tok/s 2625818.63 | norm 2.5358 | dt 199.67 ms | lr 1.1400e-04 | mlr 1.1400e-04 | mem 13.73/16.12 GiB step 57 | train 9.032755 | val 10.890124 | tok/s 2625863.20 | norm 2.4139 | dt 199.66 ms | lr 1.1600e-04 | mlr 1.1600e-04 | mem 13.73/16.12 GiB step 58 | train 9.003803 | val 10.890124 | tok/s 2622379.66 | norm 2.3591 | dt 199.93 ms | lr 1.1800e-04 | mlr 1.1800e-04 | mem 13.73/16.12 GiB step 59 | train 8.931199 | val 10.890124 | tok/s 2621916.19 | norm 2.2874 | dt 199.96 ms | lr 1.2000e-04 | mlr 1.2000e-04 | mem 13.73/16.12 GiB step 60 | train 8.869898 | val 10.890124 | tok/s 2623306.96 | norm 2.2297 | dt 199.86 ms | lr 1.2200e-04 | mlr 1.2200e-04 | mem 13.73/16.12 GiB step 61 | train 8.841613 | val 10.890124 | tok/s 2613551.04 | norm 2.1677 | dt 200.60 ms | lr 1.2400e-04 | mlr 1.2400e-04 | mem 13.73/16.12 GiB step 62 | train 8.785708 | val 10.890124 | tok/s 2619314.31 | norm 2.1582 | dt 200.16 ms | lr 1.2600e-04 | mlr 1.2600e-04 | mem 13.73/16.12 GiB step 63 | train 8.727670 | val 10.890124 | tok/s 2613294.90 | norm 2.1596 | dt 200.62 ms | lr 1.2800e-04 | mlr 1.2800e-04 | mem 13.73/16.12 GiB step 64 | train 8.675591 | val 10.890124 | tok/s 2617737.68 | norm 2.2228 | dt 200.28 ms | lr 1.3000e-04 | mlr 1.3000e-04 | mem 13.73/16.12 GiB step 65 | train 8.650561 | val 10.890124 | tok/s 2618278.29 | norm 2.1038 | dt 200.24 ms | lr 1.3200e-04 | mlr 1.3200e-04 | mem 13.73/16.12 GiB step 66 | train 8.564448 | val 10.890124 | tok/s 2618072.06 | norm 2.1432 | dt 200.26 ms | lr 1.3400e-04 | mlr 1.3400e-04 | mem 13.73/16.12 GiB step 67 | train 8.546173 | val 10.890124 | tok/s 2608329.57 | norm 2.0992 | dt 201.01 ms | lr 1.3600e-04 | mlr 1.3600e-04 | mem 13.73/16.12 GiB step 68 | train 8.458350 | val 10.890124 | tok/s 2618694.42 | norm 2.0775 | dt 200.21 ms | lr 1.3800e-04 | mlr 1.3800e-04 | mem 13.73/16.12 GiB step 69 | train 8.389072 | val 10.890124 | tok/s 2614346.99 | norm 2.1134 | dt 200.54 ms | lr 1.4000e-04 | mlr 1.4000e-04 | mem 13.73/16.12 GiB step 70 | train 8.345741 | val 10.890124 | tok/s 2618961.22 | norm 1.9973 | dt 200.19 ms | lr 1.4200e-04 | mlr 1.4200e-04 | mem 13.73/16.12 GiB step 71 | train 8.343431 | val 10.890124 | tok/s 2617008.31 | norm 1.9293 | dt 200.34 ms | lr 1.4400e-04 | mlr 1.4400e-04 | mem 13.73/16.12 GiB step 72 | train 8.291071 | val 10.890124 | tok/s 2617155.87 | norm 1.8710 | dt 200.33 ms | lr 1.4600e-04 | mlr 1.4600e-04 | mem 13.73/16.12 GiB step 73 | train 8.260945 | val 10.890124 | tok/s 2618976.45 | norm 1.9632 | dt 200.19 ms | lr 1.4800e-04 | mlr 1.4800e-04 | mem 13.73/16.12 GiB step 74 | train 8.190550 | val 10.890124 | tok/s 2618952.17 | norm 1.8819 | dt 200.19 ms | lr 1.5000e-04 | mlr 1.5000e-04 | mem 13.73/16.12 GiB step 75 | train 8.138272 | val 10.890124 | tok/s 2622972.59 | norm 1.7609 | dt 199.88 ms | lr 1.5200e-04 | mlr 1.5200e-04 | mem 13.73/16.12 GiB step 76 | train 8.058064 | val 10.890124 | tok/s 2622717.36 | norm 1.7330 | dt 199.90 ms | lr 1.5400e-04 | mlr 1.5400e-04 | mem 13.73/16.12 GiB step 77 | train 8.076279 | val 10.890124 | tok/s 2609801.42 | norm 1.6143 | dt 200.89 ms | lr 1.5600e-04 | mlr 1.5600e-04 | mem 13.73/16.12 GiB step 78 | train 8.008174 | val 10.890124 | tok/s 2622233.42 | norm 1.6270 | dt 199.94 ms | lr 1.5800e-04 | mlr 1.5800e-04 | mem 13.73/16.12 GiB step 79 | train 7.935701 | val 10.890124 | tok/s 2620957.32 | norm 1.5510 | dt 200.04 ms | lr 1.6000e-04 | mlr 1.6000e-04 | mem 13.73/16.12 GiB step 80 | train 7.878295 | val 10.890124 | tok/s 2625464.51 | norm 1.6492 | dt 199.69 ms | lr 1.6200e-04 | mlr 1.6200e-04 | mem 13.73/16.12 GiB step 81 | train 7.887467 | val 10.890124 | tok/s 2626100.07 | norm 1.6970 | dt 199.65 ms | lr 1.6400e-04 | mlr 1.6400e-04 | mem 13.73/16.12 GiB step 82 | train 7.773085 | val 10.890124 | tok/s 2533282.82 | norm 1.5644 | dt 206.96 ms | lr 1.6600e-04 | mlr 1.6600e-04 | mem 13.73/16.12 GiB step 83 | train 7.775243 | val 10.890124 | tok/s 2621669.00 | norm 1.5831 | dt 199.98 ms | lr 1.6800e-04 | mlr 1.6800e-04 | mem 13.73/16.12 GiB step 84 | train 7.764616 | val 10.890124 | tok/s 2628013.09 | norm 1.5332 | dt 199.50 ms | lr 1.7000e-04 | mlr 1.7000e-04 | mem 13.73/16.12 GiB step 85 | train 7.652994 | val 10.890124 | tok/s 2614800.11 | norm 1.7092 | dt 200.51 ms | lr 1.7200e-04 | mlr 1.7200e-04 | mem 13.73/16.12 GiB step 86 | train 7.621388 | val 10.890124 | tok/s 2616775.77 | norm 1.6057 | dt 200.36 ms | lr 1.7400e-04 | mlr 1.7400e-04 | mem 13.73/16.12 GiB step 87 | train 7.604894 | val 10.890124 | tok/s 2624180.12 | norm 1.3607 | dt 199.79 ms | lr 1.7600e-04 | mlr 1.7600e-04 | mem 13.73/16.12 GiB step 88 | train 7.538146 | val 10.890124 | tok/s 2623583.69 | norm 1.4537 | dt 199.84 ms | lr 1.7800e-04 | mlr 1.7800e-04 | mem 13.73/16.12 GiB step 89 | train 7.650591 | val 10.890124 | tok/s 2622176.92 | norm 2.0019 | dt 199.94 ms | lr 1.8000e-04 | mlr 1.8000e-04 | mem 13.73/16.12 GiB step 90 | train 7.492693 | val 10.890124 | tok/s 2616096.92 | norm 1.9577 | dt 200.41 ms | lr 1.8200e-04 | mlr 1.8200e-04 | mem 13.73/16.12 GiB step 91 | train 7.457668 | val 10.890124 | tok/s 2617597.76 | norm 1.2333 | dt 200.29 ms | lr 1.8400e-04 | mlr 1.8400e-04 | mem 13.73/16.12 GiB step 92 | train 7.427184 | val 10.890124 | tok/s 2617160.05 | norm 1.3266 | dt 200.33 ms | lr 1.8600e-04 | mlr 1.8600e-04 | mem 13.73/16.12 GiB step 93 | train 7.424969 | val 10.890124 | tok/s 2620359.46 | norm 1.4070 | dt 200.08 ms | lr 1.8800e-04 | mlr 1.8800e-04 | mem 13.73/16.12 GiB step 94 | train 7.320952 | val 10.890124 | tok/s 2626715.27 | norm 1.3784 | dt 199.60 ms | lr 1.9000e-04 | mlr 1.9000e-04 | mem 13.73/16.12 GiB step 95 | train 7.332765 | val 10.890124 | tok/s 2616455.46 | norm 1.4033 | dt 200.38 ms | lr 1.9200e-04 | mlr 1.9200e-04 | mem 13.73/16.12 GiB step 96 | train 7.276710 | val 10.890124 | tok/s 2617385.20 | norm 1.3482 | dt 200.31 ms | lr 1.9400e-04 | mlr 1.9400e-04 | mem 13.73/16.12 GiB step 97 | train 7.264753 | val 10.890124 | tok/s 2625066.58 | norm 1.3534 | dt 199.72 ms | lr 1.9600e-04 | mlr 1.9600e-04 | mem 13.73/16.12 GiB step 98 | train 7.252795 | val 10.890124 | tok/s 2462588.73 | norm 1.4386 | dt 212.90 ms | lr 1.9800e-04 | mlr 1.9800e-04 | mem 13.73/16.12 GiB step 99 | train 7.233990 | val 10.890124 | tok/s 2542469.19 | norm 1.5351 | dt 206.21 ms | lr 2.0000e-04 | mlr 2.0000e-04 | mem 13.73/16.12 GiB step 100 | train 7.250553 | val 10.890124 | tok/s 2621161.45 | norm 1.2044 | dt 200.02 ms | lr 2.0200e-04 | mlr 2.0200e-04 | mem 13.73/16.12 GiB step 101 | train 7.095853 | val 10.890124 | tok/s 2622293.82 | norm 1.1549 | dt 199.93 ms | lr 2.0400e-04 | mlr 2.0400e-04 | mem 13.73/16.12 GiB step 102 | train 7.130420 | val 10.890124 | tok/s 2625127.04 | norm 1.0311 | dt 199.72 ms | lr 2.0600e-04 | mlr 2.0600e-04 | mem 13.73/16.12 GiB step 103 | train 7.127869 | val 10.890124 | tok/s 2625928.46 | norm 1.4623 | dt 199.66 ms | lr 2.0800e-04 | mlr 2.0800e-04 | mem 13.73/16.12 GiB step 104 | train 7.107324 | val 10.890124 | tok/s 2625648.64 | norm 1.3400 | dt 199.68 ms | lr 2.1000e-04 | mlr 2.1000e-04 | mem 13.73/16.12 GiB step 105 | train 7.098248 | val 10.890124 | tok/s 2615240.64 | norm 1.5608 | dt 200.47 ms | lr 2.1200e-04 | mlr 2.1200e-04 | mem 13.73/16.12 GiB step 106 | train 7.000329 | val 10.890124 | tok/s 2584256.58 | norm 1.5546 | dt 202.88 ms | lr 2.1400e-04 | mlr 2.1400e-04 | mem 13.73/16.12 GiB step 107 | train 6.982265 | val 10.890124 | tok/s 2616401.74 | norm 1.8075 | dt 200.39 ms | lr 2.1600e-04 | mlr 2.1600e-04 | mem 13.73/16.12 GiB step 108 | train 6.947188 | val 10.890124 | tok/s 2622429.47 | norm 1.3759 | dt 199.92 ms | lr 2.1800e-04 | mlr 2.1800e-04 | mem 13.73/16.12 GiB step 109 | train 6.945286 | val 10.890124 | tok/s 2626368.84 | norm 1.5908 | dt 199.62 ms | lr 2.2000e-04 | mlr 2.2000e-04 | mem 13.73/16.12 GiB step 110 | train 7.005113 | val 10.890124 | tok/s 2621094.46 | norm 1.3169 | dt 200.03 ms | lr 2.2200e-04 | mlr 2.2200e-04 | mem 13.73/16.12 GiB step 111 | train 6.925069 | val 10.890124 | tok/s 2623594.46 | norm 1.7664 | dt 199.84 ms | lr 2.2400e-04 | mlr 2.2400e-04 | mem 13.73/16.12 GiB step 112 | train 6.974796 | val 10.890124 | tok/s 2629126.18 | norm 1.4850 | dt 199.42 ms | lr 2.2600e-04 | mlr 2.2600e-04 | mem 13.73/16.12 GiB step 113 | train 6.881094 | val 10.890124 | tok/s 2617904.10 | norm 1.8413 | dt 200.27 ms | lr 2.2800e-04 | mlr 2.2800e-04 | mem 13.73/16.12 GiB step 114 | train 6.947341 | val 10.890124 | tok/s 2625866.00 | norm 1.6640 | dt 199.66 ms | lr 2.3000e-04 | mlr 2.3000e-04 | mem 13.73/16.12 GiB step 115 | train 6.886013 | val 10.890124 | tok/s 2618847.43 | norm 1.9031 | dt 200.20 ms | lr 2.3200e-04 | mlr 2.3200e-04 | mem 13.73/16.12 GiB step 116 | train 6.842851 | val 10.890124 | tok/s 2620967.61 | norm 1.8825 | dt 200.04 ms | lr 2.3400e-04 | mlr 2.3400e-04 | mem 13.73/16.12 GiB step 117 | train 6.967352 | val 10.890124 | tok/s 2625354.17 | norm 1.7995 | dt 199.70 ms | lr 2.3600e-04 | mlr 2.3600e-04 | mem 13.73/16.12 GiB step 118 | train 6.732530 | val 10.890124 | tok/s 2618481.63 | norm 1.7866 | dt 200.23 ms | lr 2.3800e-04 | mlr 2.3800e-04 | mem 13.73/16.12 GiB step 119 | train 6.775620 | val 10.890124 | tok/s 2623319.88 | norm 1.7904 | dt 199.86 ms | lr 2.4000e-04 | mlr 2.4000e-04 | mem 13.73/16.12 GiB step 120 | train 6.711155 | val 10.890124 | tok/s 2620634.90 | norm 1.5568 | dt 200.06 ms | lr 2.4200e-04 | mlr 2.4200e-04 | mem 13.73/16.12 GiB step 121 | train 6.645148 | val 10.890124 | tok/s 2622693.48 | norm 1.5866 | dt 199.90 ms | lr 2.4400e-04 | mlr 2.4400e-04 | mem 13.73/16.12 GiB step 122 | train 6.842697 | val 10.890124 | tok/s 2623326.47 | norm 1.6113 | dt 199.86 ms | lr 2.4600e-04 | mlr 2.4600e-04 | mem 13.73/16.12 GiB step 123 | train 6.713111 | val 10.890124 | tok/s 2626455.85 | norm 1.9546 | dt 199.62 ms | lr 2.4800e-04 | mlr 2.4800e-04 | mem 13.73/16.12 GiB step 124 | train 6.682314 | val 10.890124 | tok/s 2623100.20 | norm 1.3421 | dt 199.87 ms | lr 2.5000e-04 | mlr 2.5000e-04 | mem 13.73/16.12 GiB step 125 | train 6.657461 | val 10.890124 | tok/s 2614459.51 | norm 2.0774 | dt 200.53 ms | lr 2.5200e-04 | mlr 2.5200e-04 | mem 13.73/16.12 GiB step 126 | train 6.660189 | val 10.890124 | tok/s 2626481.14 | norm 1.2643 | dt 199.62 ms | lr 2.5400e-04 | mlr 2.5400e-04 | mem 13.73/16.12 GiB step 127 | train 6.794259 | val 10.890124 | tok/s 2628172.87 | norm 2.2034 | dt 199.49 ms | lr 2.5600e-04 | mlr 2.5600e-04 | mem 13.73/16.12 GiB step 128 | train 6.669499 | val 10.890124 | tok/s 2627034.52 | norm 1.2665 | dt 199.57 ms | lr 2.5800e-04 | mlr 2.5800e-04 | mem 13.73/16.12 GiB step 129 | train 6.655749 | val 10.890124 | tok/s 2628743.37 | norm 1.9536 | dt 199.44 ms | lr 2.6000e-04 | mlr 2.6000e-04 | mem 13.73/16.12 GiB step 130 | train 6.573931 | val 10.890124 | tok/s 2625612.89 | norm 1.2665 | dt 199.68 ms | lr 2.6200e-04 | mlr 2.6200e-04 | mem 13.73/16.12 GiB step 131 | train 6.637104 | val 10.890124 | tok/s 2623698.72 | norm 1.6776 | dt 199.83 ms | lr 2.6400e-04 | mlr 2.6400e-04 | mem 13.73/16.12 GiB step 132 | train 6.604269 | val 10.890124 | tok/s 2634088.82 | norm 1.2559 | dt 199.04 ms | lr 2.6600e-04 | mlr 2.6600e-04 | mem 13.73/16.12 GiB step 133 | train 6.596828 | val 10.890124 | tok/s 2625368.09 | norm 1.7100 | dt 199.70 ms | lr 2.6800e-04 | mlr 2.6800e-04 | mem 13.73/16.12 GiB step 134 | train 6.551306 | val 10.890124 | tok/s 2631540.67 | norm 1.5109 | dt 199.23 ms | lr 2.7000e-04 | mlr 2.7000e-04 | mem 13.73/16.12 GiB step 135 | train 6.594377 | val 10.890124 | tok/s 2625576.90 | norm 1.4653 | dt 199.68 ms | lr 2.7200e-04 | mlr 2.7200e-04 | mem 13.73/16.12 GiB step 136 | train 6.652698 | val 10.890124 | tok/s 2627690.37 | norm 1.6780 | dt 199.52 ms | lr 2.7400e-04 | mlr 2.7400e-04 | mem 13.73/16.12 GiB step 137 | train 6.542068 | val 10.890124 | tok/s 2632535.48 | norm 1.1670 | dt 199.16 ms | lr 2.7600e-04 | mlr 2.7600e-04 | mem 13.73/16.12 GiB step 138 | train 6.517333 | val 10.890124 | tok/s 2631412.55 | norm 1.2278 | dt 199.24 ms | lr 2.7800e-04 | mlr 2.7800e-04 | mem 13.73/16.12 GiB step 139 | train 6.421658 | val 10.890124 | tok/s 2623340.72 | norm 1.2747 | dt 199.86 ms | lr 2.8000e-04 | mlr 2.8000e-04 | mem 13.73/16.12 GiB step 140 | train 6.509473 | val 10.890124 | tok/s 2622798.50 | norm 1.2973 | dt 199.90 ms | lr 2.8200e-04 | mlr 2.8200e-04 | mem 13.73/16.12 GiB step 141 | train 6.428558 | val 10.890124 | tok/s 2626284.90 | norm 1.4457 | dt 199.63 ms | lr 2.8400e-04 | mlr 2.8400e-04 | mem 13.73/16.12 GiB step 142 | train 6.463293 | val 10.890124 | tok/s 2628026.45 | norm 1.6122 | dt 199.50 ms | lr 2.8600e-04 | mlr 2.8600e-04 | mem 13.73/16.12 GiB step 143 | train 6.426427 | val 10.890124 | tok/s 2629692.29 | norm 1.3460 | dt 199.37 ms | lr 2.8800e-04 | mlr 2.8800e-04 | mem 13.73/16.12 GiB step 144 | train 6.439695 | val 10.890124 | tok/s 2628557.00 | norm 1.8925 | dt 199.46 ms | lr 2.9000e-04 | mlr 2.9000e-04 | mem 13.73/16.12 GiB step 145 | train 6.471744 | val 10.890124 | tok/s 2621774.96 | norm 1.6747 | dt 199.97 ms | lr 2.9200e-04 | mlr 2.9200e-04 | mem 13.73/16.12 GiB step 146 | train 6.442837 | val 10.890124 | tok/s 2626084.48 | norm 2.2375 | dt 199.65 ms | lr 2.9400e-04 | mlr 2.9400e-04 | mem 13.73/16.12 GiB step 147 | train 6.407235 | val 10.890124 | tok/s 2627483.94 | norm 1.4127 | dt 199.54 ms | lr 2.9600e-04 | mlr 2.9600e-04 | mem 13.73/16.12 GiB step 148 | train 6.360006 | val 10.890124 | tok/s 2628292.33 | norm 2.3021 | dt 199.48 ms | lr 2.9800e-04 | mlr 2.9800e-04 | mem 13.73/16.12 GiB step 149 | train 6.417266 | val 10.890124 | tok/s 2627391.04 | norm 1.5549 | dt 199.55 ms | lr 3.0000e-04 | mlr 3.0000e-04 | mem 13.73/16.12 GiB step 150 | train 6.716017 | val 10.890124 | tok/s 2622632.62 | norm 2.3467 | dt 199.91 ms | lr 3.0200e-04 | mlr 3.0200e-04 | mem 13.73/16.12 GiB step 151 | train 6.466736 | val 10.890124 | tok/s 2622739.44 | norm 1.2291 | dt 199.90 ms | lr 3.0400e-04 | mlr 3.0400e-04 | mem 13.73/16.12 GiB step 152 | train 6.279354 | val 10.890124 | tok/s 2632852.57 | norm 1.6300 | dt 199.13 ms | lr 3.0600e-04 | mlr 3.0600e-04 | mem 13.73/16.12 GiB step 153 | train 6.323865 | val 10.890124 | tok/s 2625606.25 | norm 1.2568 | dt 199.68 ms | lr 3.0800e-04 | mlr 3.0800e-04 | mem 13.73/16.12 GiB step 154 | train 6.300718 | val 10.890124 | tok/s 2624347.75 | norm 1.7707 | dt 199.78 ms | lr 3.1000e-04 | mlr 3.1000e-04 | mem 13.73/16.12 GiB step 155 | train 6.302104 | val 10.890124 | tok/s 2623090.24 | norm 1.5645 | dt 199.87 ms | lr 3.1200e-04 | mlr 3.1200e-04 | mem 13.73/16.12 GiB step 156 | train 6.272314 | val 10.890124 | tok/s 2625850.55 | norm 1.7648 | dt 199.66 ms | lr 3.1400e-04 | mlr 3.1400e-04 | mem 13.73/16.12 GiB step 157 | train 6.115891 | val 10.890124 | tok/s 2627306.38 | norm 1.7056 | dt 199.55 ms | lr 3.1600e-04 | mlr 3.1600e-04 | mem 13.73/16.12 GiB step 158 | train 6.231004 | val 10.890124 | tok/s 2622897.26 | norm 1.7549 | dt 199.89 ms | lr 3.1800e-04 | mlr 3.1800e-04 | mem 13.73/16.12 GiB step 159 | train 6.292294 | val 10.890124 | tok/s 2625147.92 | norm 1.9287 | dt 199.72 ms | lr 3.2000e-04 | mlr 3.2000e-04 | mem 13.73/16.12 GiB step 160 | train 6.249069 | val 10.890124 | tok/s 2621874.84 | norm 1.7343 | dt 199.97 ms | lr 3.2200e-04 | mlr 3.2200e-04 | mem 13.73/16.12 GiB step 161 | train 6.246403 | val 10.890124 | tok/s 2623741.26 | norm 2.0770 | dt 199.82 ms | lr 3.2400e-04 | mlr 3.2400e-04 | mem 13.73/16.12 GiB step 162 | train 6.219712 | val 10.890124 | tok/s 2628073.47 | norm 1.7433 | dt 199.50 ms | lr 3.2600e-04 | mlr 3.2600e-04 | mem 13.73/16.12 GiB step 163 | train 6.262054 | val 10.890124 | tok/s 2626188.92 | norm 2.4484 | dt 199.64 ms | lr 3.2800e-04 | mlr 3.2800e-04 | mem 13.73/16.12 GiB step 164 | train 6.279904 | val 10.890124 | tok/s 2626342.84 | norm 2.0787 | dt 199.63 ms | lr 3.3000e-04 | mlr 3.3000e-04 | mem 13.73/16.12 GiB step 165 | train 6.198001 | val 10.890124 | tok/s 2612017.07 | norm 2.2316 | dt 200.72 ms | lr 3.3200e-04 | mlr 3.3200e-04 | mem 13.73/16.12 GiB step 166 | train 6.216063 | val 10.890124 | tok/s 2602142.06 | norm 1.8225 | dt 201.48 ms | lr 3.3400e-04 | mlr 3.3400e-04 | mem 13.73/16.12 GiB step 167 | train 6.019052 | val 10.890124 | tok/s 2633185.03 | norm 2.6026 | dt 199.11 ms | lr 3.3600e-04 | mlr 3.3600e-04 | mem 13.73/16.12 GiB step 168 | train 6.191161 | val 10.890124 | tok/s 2625154.73 | norm 1.6028 | dt 199.72 ms | lr 3.3800e-04 | mlr 3.3800e-04 | mem 13.73/16.12 GiB step 169 | train 6.241641 | val 10.890124 | tok/s 2627191.88 | norm 2.7061 | dt 199.56 ms | lr 3.4000e-04 | mlr 3.4000e-04 | mem 13.73/16.12 GiB step 170 | train 6.095707 | val 10.890124 | tok/s 2620072.79 | norm 1.5601 | dt 200.10 ms | lr 3.4200e-04 | mlr 3.4200e-04 | mem 13.73/16.12 GiB step 171 | train 6.218936 | val 10.890124 | tok/s 2621657.82 | norm 2.6651 | dt 199.98 ms | lr 3.4400e-04 | mlr 3.4400e-04 | mem 13.73/16.12 GiB step 172 | train 6.128318 | val 10.890124 | tok/s 2623373.64 | norm 1.9224 | dt 199.85 ms | lr 3.4600e-04 | mlr 3.4600e-04 | mem 13.73/16.12 GiB step 173 | train 6.090058 | val 10.890124 | tok/s 2629469.22 | norm 2.3104 | dt 199.39 ms | lr 3.4800e-04 | mlr 3.4800e-04 | mem 13.73/16.12 GiB step 174 | train 6.093761 | val 10.890124 | tok/s 2624171.58 | norm 1.7503 | dt 199.79 ms | lr 3.5000e-04 | mlr 3.5000e-04 | mem 13.73/16.12 GiB step 175 | train 6.208789 | val 10.890124 | tok/s 2625314.56 | norm 2.3003 | dt 199.70 ms | lr 3.5200e-04 | mlr 3.5200e-04 | mem 13.73/16.12 GiB step 176 | train 6.137531 | val 10.890124 | tok/s 2624989.14 | norm 2.1430 | dt 199.73 ms | lr 3.5400e-04 | mlr 3.5400e-04 | mem 13.73/16.12 GiB step 177 | train 6.071106 | val 10.890124 | tok/s 2629852.46 | norm 3.2150 | dt 199.36 ms | lr 3.5600e-04 | mlr 3.5600e-04 | mem 13.73/16.12 GiB step 178 | train 6.093849 | val 10.890124 | tok/s 2625015.03 | norm 2.9801 | dt 199.73 ms | lr 3.5800e-04 | mlr 3.5800e-04 | mem 13.73/16.12 GiB step 179 | train 6.031875 | val 10.890124 | tok/s 2628247.78 | norm 3.0021 | dt 199.48 ms | lr 3.6000e-04 | mlr 3.6000e-04 | mem 13.73/16.12 GiB step 180 | train 6.063037 | val 10.890124 | tok/s 2627551.63 | norm 2.5557 | dt 199.53 ms | lr 3.6200e-04 | mlr 3.6200e-04 | mem 13.73/16.12 GiB step 181 | train 6.034023 | val 10.890124 | tok/s 2625435.74 | norm 2.3965 | dt 199.70 ms | lr 3.6400e-04 | mlr 3.6400e-04 | mem 13.73/16.12 GiB step 182 | train 6.070764 | val 10.890124 | tok/s 2627668.88 | norm 2.8298 | dt 199.53 ms | lr 3.6600e-04 | mlr 3.6600e-04 | mem 13.73/16.12 GiB step 183 | train 6.092714 | val 10.890124 | tok/s 2627043.29 | norm 2.1460 | dt 199.57 ms | lr 3.6800e-04 | mlr 3.6800e-04 | mem 13.73/16.12 GiB step 184 | train 6.076263 | val 10.890124 | tok/s 2628460.46 | norm 2.8300 | dt 199.47 ms | lr 3.7000e-04 | mlr 3.7000e-04 | mem 13.73/16.12 GiB step 185 | train 6.045171 | val 10.890124 | tok/s 2625883.40 | norm 1.8217 | dt 199.66 ms | lr 3.7200e-04 | mlr 3.7200e-04 | mem 13.73/16.12 GiB step 186 | train 6.009628 | val 10.890124 | tok/s 2584081.53 | norm 2.6294 | dt 202.89 ms | lr 3.7400e-04 | mlr 3.7400e-04 | mem 13.73/16.12 GiB step 187 | train 5.999230 | val 10.890124 | tok/s 2629205.40 | norm 2.2152 | dt 199.41 ms | lr 3.7600e-04 | mlr 3.7600e-04 | mem 13.73/16.12 GiB step 188 | train 6.081181 | val 10.890124 | tok/s 2621660.84 | norm 2.9613 | dt 199.98 ms | lr 3.7800e-04 | mlr 3.7800e-04 | mem 13.73/16.12 GiB step 189 | train 6.026134 | val 10.890124 | tok/s 2628378.88 | norm 2.3919 | dt 199.47 ms | lr 3.8000e-04 | mlr 3.8000e-04 | mem 13.73/16.12 GiB step 190 | train 5.998135 | val 10.890124 | tok/s 2627260.83 | norm 3.7333 | dt 199.56 ms | lr 3.8200e-04 | mlr 3.8200e-04 | mem 13.73/16.12 GiB step 191 | train 5.992144 | val 10.890124 | tok/s 2625304.99 | norm 2.6757 | dt 199.71 ms | lr 3.8400e-04 | mlr 3.8400e-04 | mem 13.73/16.12 GiB step 192 | train 6.017836 | val 10.890124 | tok/s 2626342.61 | norm 3.6348 | dt 199.63 ms | lr 3.8600e-04 | mlr 3.8600e-04 | mem 13.73/16.12 GiB step 193 | train 6.100421 | val 10.890124 | tok/s 2625104.76 | norm 2.4505 | dt 199.72 ms | lr 3.8800e-04 | mlr 3.8800e-04 | mem 13.73/16.12 GiB step 194 | train 6.046615 | val 10.890124 | tok/s 2627964.43 | norm 3.3168 | dt 199.50 ms | lr 3.9000e-04 | mlr 3.9000e-04 | mem 13.73/16.12 GiB step 195 | train 5.954826 | val 10.890124 | tok/s 2623924.61 | norm 2.7958 | dt 199.81 ms | lr 3.9200e-04 | mlr 3.9200e-04 | mem 13.73/16.12 GiB step 196 | train 6.016832 | val 10.890124 | tok/s 2601536.36 | norm 2.8702 | dt 201.53 ms | lr 3.9400e-04 | mlr 3.9400e-04 | mem 13.73/16.12 GiB step 197 | train 5.972055 | val 10.890124 | tok/s 2620414.13 | norm 2.7414 | dt 200.08 ms | lr 3.9600e-04 | mlr 3.9600e-04 | mem 13.73/16.12 GiB step 198 | train 6.021826 | val 10.890124 | tok/s 2616007.36 | norm 2.5380 | dt 200.42 ms | lr 3.9800e-04 | mlr 3.9800e-04 | mem 13.73/16.12 GiB step 199 | train 5.953606 | val 10.890124 | tok/s 2626435.66 | norm 3.1054 | dt 199.62 ms | lr 4.0000e-04 | mlr 4.0000e-04 | mem 13.73/16.12 GiB step 200 | train 6.083564 | val 10.890124 | tok/s 2628556.45 | norm 3.3757 | dt 199.46 ms | lr 4.0200e-04 | mlr 4.0200e-04 | mem 13.73/16.12 GiB step 201 | train 5.933506 | val 10.890124 | tok/s 2629031.29 | norm 3.4152 | dt 199.42 ms | lr 4.0400e-04 | mlr 4.0400e-04 | mem 13.73/16.12 GiB step 202 | train 6.016769 | val 10.890124 | tok/s 2418420.77 | norm 4.6668 | dt 216.79 ms | lr 4.0600e-04 | mlr 4.0600e-04 | mem 13.73/16.12 GiB step 203 | train 5.914853 | val 10.890124 | tok/s 2626645.22 | norm 2.9346 | dt 199.60 ms | lr 4.0800e-04 | mlr 4.0800e-04 | mem 13.73/16.12 GiB step 204 | train 6.007089 | val 10.890124 | tok/s 2630325.44 | norm 4.1476 | dt 199.32 ms | lr 4.1000e-04 | mlr 4.1000e-04 | mem 13.73/16.12 GiB step 205 | train 5.883353 | val 10.890124 | tok/s 2631863.79 | norm 2.8221 | dt 199.21 ms | lr 4.1200e-04 | mlr 4.1200e-04 | mem 13.73/16.12 GiB step 206 | train 5.844732 | val 10.890124 | tok/s 2632165.35 | norm 3.5981 | dt 199.19 ms | lr 4.1400e-04 | mlr 4.1400e-04 | mem 13.73/16.12 GiB step 207 | train 5.885807 | val 10.890124 | tok/s 2633264.62 | norm 2.6837 | dt 199.10 ms | lr 4.1600e-04 | mlr 4.1600e-04 | mem 13.73/16.12 GiB step 208 | train 5.845690 | val 10.890124 | tok/s 2618579.36 | norm 3.7416 | dt 200.22 ms | lr 4.1800e-04 | mlr 4.1800e-04 | mem 13.73/16.12 GiB step 209 | train 5.867370 | val 10.890124 | tok/s 2628277.79 | norm 3.1563 | dt 199.48 ms | lr 4.2000e-04 | mlr 4.2000e-04 | mem 13.73/16.12 GiB step 210 | train 5.871372 | val 10.890124 | tok/s 2512760.57 | norm 3.2861 | dt 208.65 ms | lr 4.2200e-04 | mlr 4.2200e-04 | mem 13.73/16.12 GiB step 211 | train 5.773825 | val 10.890124 | tok/s 2631115.55 | norm 2.7563 | dt 199.26 ms | lr 4.2400e-04 | mlr 4.2400e-04 | mem 13.73/16.12 GiB step 212 | train 5.815445 | val 10.890124 | tok/s 2618868.07 | norm 3.0464 | dt 200.20 ms | lr 4.2600e-04 | mlr 4.2600e-04 | mem 13.73/16.12 GiB step 213 | train 5.844877 | val 10.890124 | tok/s 2623012.98 | norm 3.2866 | dt 199.88 ms | lr 4.2800e-04 | mlr 4.2800e-04 | mem 13.73/16.12 GiB step 214 | train 5.801294 | val 10.890124 | tok/s 2628120.46 | norm 3.2461 | dt 199.49 ms | lr 4.3000e-04 | mlr 4.3000e-04 | mem 13.73/16.12 GiB step 215 | train 5.787760 | val 10.890124 | tok/s 2629534.51 | norm 2.9296 | dt 199.38 ms | lr 4.3200e-04 | mlr 4.3200e-04 | mem 13.73/16.12 GiB step 216 | train 5.844946 | val 10.890124 | tok/s 2623146.82 | norm 3.3367 | dt 199.87 ms | lr 4.3400e-04 | mlr 4.3400e-04 | mem 13.73/16.12 GiB step 217 | train 5.825261 | val 10.890124 | tok/s 2620325.24 | norm 2.7478 | dt 200.09 ms | lr 4.3600e-04 | mlr 4.3600e-04 | mem 13.73/16.12 GiB step 218 | train 5.754633 | val 10.890124 | tok/s 2621769.12 | norm 3.2633 | dt 199.97 ms | lr 4.3800e-04 | mlr 4.3800e-04 | mem 13.73/16.12 GiB step 219 | train 5.775657 | val 10.890124 | tok/s 2632683.73 | norm 2.6769 | dt 199.15 ms | lr 4.4000e-04 | mlr 4.4000e-04 | mem 13.73/16.12 GiB step 220 | train 5.784617 | val 10.890124 | tok/s 2625407.24 | norm 3.0301 | dt 199.70 ms | lr 4.4200e-04 | mlr 4.4200e-04 | mem 13.73/16.12 GiB step 221 | train 5.718401 | val 10.890124 | tok/s 2629304.64 | norm 2.5292 | dt 199.40 ms | lr 4.4400e-04 | mlr 4.4400e-04 | mem 13.73/16.12 GiB step 222 | train 5.776261 | val 10.890124 | tok/s 2624178.41 | norm 3.8496 | dt 199.79 ms | lr 4.4600e-04 | mlr 4.4600e-04 | mem 13.73/16.12 GiB step 223 | train 5.784551 | val 10.890124 | tok/s 2626692.23 | norm 3.2915 | dt 199.60 ms | lr 4.4800e-04 | mlr 4.4800e-04 | mem 13.73/16.12 GiB step 224 | train 5.690553 | val 10.890124 | tok/s 2626422.23 | norm 4.0893 | dt 199.62 ms | lr 4.5000e-04 | mlr 4.5000e-04 | mem 13.73/16.12 GiB step 225 | train 5.718791 | val 10.890124 | tok/s 2621884.82 | norm 2.8543 | dt 199.97 ms | lr 4.5200e-04 | mlr 4.5200e-04 | mem 13.73/16.12 GiB step 226 | train 5.777402 | val 10.890124 | tok/s 2625926.88 | norm 3.7291 | dt 199.66 ms | lr 4.5400e-04 | mlr 4.5400e-04 | mem 13.73/16.12 GiB step 227 | train 5.725349 | val 10.890124 | tok/s 2622205.50 | norm 2.9209 | dt 199.94 ms | lr 4.5600e-04 | mlr 4.5600e-04 | mem 13.73/16.12 GiB step 228 | train 5.705347 | val 10.890124 | tok/s 2626567.61 | norm 3.9859 | dt 199.61 ms | lr 4.5800e-04 | mlr 4.5800e-04 | mem 13.73/16.12 GiB step 229 | train 5.739884 | val 10.890124 | tok/s 2630250.71 | norm 3.3577 | dt 199.33 ms | lr 4.6000e-04 | mlr 4.6000e-04 | mem 13.73/16.12 GiB step 230 | train 5.700396 | val 10.890124 | tok/s 2625279.68 | norm 4.3692 | dt 199.71 ms | lr 4.6200e-04 | mlr 4.6200e-04 | mem 13.73/16.12 GiB step 231 | train 5.729311 | val 10.890124 | tok/s 2626580.59 | norm 4.2910 | dt 199.61 ms | lr 4.6400e-04 | mlr 4.6400e-04 | mem 13.73/16.12 GiB step 232 | train 5.645062 | val 10.890124 | tok/s 2625327.01 | norm 3.8565 | dt 199.70 ms | lr 4.6600e-04 | mlr 4.6600e-04 | mem 13.73/16.12 GiB step 233 | train 5.678786 | val 10.890124 | tok/s 2617934.18 | norm 3.4501 | dt 200.27 ms | lr 4.6800e-04 | mlr 4.6800e-04 | mem 13.73/16.12 GiB step 234 | train 5.725953 | val 10.890124 | tok/s 2630157.95 | norm 3.5866 | dt 199.34 ms | lr 4.7000e-04 | mlr 4.7000e-04 | mem 13.73/16.12 GiB step 235 | train 5.698370 | val 10.890124 | tok/s 2631181.53 | norm 3.7486 | dt 199.26 ms | lr 4.7200e-04 | mlr 4.7200e-04 | mem 13.73/16.12 GiB step 236 | train 5.718711 | val 10.890124 | tok/s 2631946.37 | norm 3.7765 | dt 199.20 ms | lr 4.7400e-04 | mlr 4.7400e-04 | mem 13.73/16.12 GiB step 237 | train 5.624604 | val 10.890124 | tok/s 2624849.56 | norm 3.4018 | dt 199.74 ms | lr 4.7600e-04 | mlr 4.7600e-04 | mem 13.73/16.12 GiB step 238 | train 5.601611 | val 10.890124 | tok/s 2620841.05 | norm 3.3253 | dt 200.05 ms | lr 4.7800e-04 | mlr 4.7800e-04 | mem 13.73/16.12 GiB step 239 | train 5.607734 | val 10.890124 | tok/s 2626824.27 | norm 3.3476 | dt 199.59 ms | lr 4.8000e-04 | mlr 4.8000e-04 | mem 13.73/16.12 GiB step 240 | train 5.630656 | val 10.890124 | tok/s 2625017.72 | norm 3.8823 | dt 199.73 ms | lr 4.8200e-04 | mlr 4.8200e-04 | mem 13.73/16.12 GiB step 241 | train 5.611836 | val 10.890124 | tok/s 2609580.64 | norm 3.3998 | dt 200.91 ms | lr 4.8400e-04 | mlr 4.8400e-04 | mem 13.73/16.12 GiB step 242 | train 5.630474 | val 10.890124 | tok/s 2625309.05 | norm 4.2301 | dt 199.71 ms | lr 4.8600e-04 | mlr 4.8600e-04 | mem 13.73/16.12 GiB step 243 | train 5.656474 | val 10.890124 | tok/s 2625648.57 | norm 3.5930 | dt 199.68 ms | lr 4.8800e-04 | mlr 4.8800e-04 | mem 13.73/16.12 GiB step 244 | train 5.713233 | val 10.890124 | tok/s 2625739.49 | norm 3.8360 | dt 199.67 ms | lr 4.9000e-04 | mlr 4.9000e-04 | mem 13.73/16.12 GiB step 245 | train 5.631864 | val 10.890124 | tok/s 2627789.17 | norm 3.1903 | dt 199.52 ms | lr 4.9200e-04 | mlr 4.9200e-04 | mem 13.73/16.12 GiB step 246 | train 5.580963 | val 10.890124 | tok/s 2631595.10 | norm 4.3569 | dt 199.23 ms | lr 4.9400e-04 | mlr 4.9400e-04 | mem 13.73/16.12 GiB step 247 | train 5.500273 | val 10.890124 | tok/s 2618108.68 | norm 4.0516 | dt 200.25 ms | lr 4.9600e-04 | mlr 4.9600e-04 | mem 13.73/16.12 GiB step 248 | train 5.527167 | val 10.890124 | tok/s 2623023.34 | norm 3.8376 | dt 199.88 ms | lr 4.9800e-04 | mlr 4.9800e-04 | mem 13.73/16.12 GiB step 249 | train 5.545539 | val 10.890124 | tok/s 2626699.94 | norm 3.4564 | dt 199.60 ms | lr 5.0000e-04 | mlr 5.0000e-04 | mem 13.73/16.12 GiB [val] step 250 | val_loss 5.900655 step 250 | train 5.540193 | val 5.900655 | tok/s 2615083.44 | norm 3.8893 | dt 200.49 ms | lr 5.0200e-04 | mlr 5.0200e-04 | mem 13.73/16.12 GiB step 251 | train 5.585845 | val 5.900655 | tok/s 2626122.91 | norm 4.0401 | dt 199.64 ms | lr 5.0400e-04 | mlr 5.0400e-04 | mem 13.73/16.12 GiB step 252 | train 5.576172 | val 5.900655 | tok/s 2628006.94 | norm 3.8086 | dt 199.50 ms | lr 5.0600e-04 | mlr 5.0600e-04 | mem 13.73/16.12 GiB step 253 | train 5.556731 | val 5.900655 | tok/s 2617902.69 | norm 3.4559 | dt 200.27 ms | lr 5.0800e-04 | mlr 5.0800e-04 | mem 13.73/16.12 GiB step 254 | train 5.548102 | val 5.900655 | tok/s 2593253.09 | norm 3.2030 | dt 202.17 ms | lr 5.1000e-04 | mlr 5.1000e-04 | mem 13.73/16.12 GiB step 255 | train 5.507322 | val 5.900655 | tok/s 2631764.17 | norm 3.4391 | dt 199.22 ms | lr 5.1200e-04 | mlr 5.1200e-04 | mem 13.73/16.12 GiB step 256 | train 5.487433 | val 5.900655 | tok/s 2630431.83 | norm 3.3662 | dt 199.32 ms | lr 5.1400e-04 | mlr 5.1400e-04 | mem 13.73/16.12 GiB step 257 | train 5.574638 | val 5.900655 | tok/s 2629991.56 | norm 3.9695 | dt 199.35 ms | lr 5.1600e-04 | mlr 5.1600e-04 | mem 13.73/16.12 GiB step 258 | train 5.470649 | val 5.900655 | tok/s 2621981.44 | norm 3.8929 | dt 199.96 ms | lr 5.1800e-04 | mlr 5.1800e-04 | mem 13.73/16.12 GiB step 259 | train 5.536852 | val 5.900655 | tok/s 2625055.13 | norm 3.5880 | dt 199.72 ms | lr 5.2000e-04 | mlr 5.2000e-04 | mem 13.73/16.12 GiB step 260 | train 5.521070 | val 5.900655 | tok/s 2632103.93 | norm 3.1122 | dt 199.19 ms | lr 5.2200e-04 | mlr 5.2200e-04 | mem 13.73/16.12 GiB step 261 | train 5.456617 | val 5.900655 | tok/s 2623509.78 | norm 4.0929 | dt 199.84 ms | lr 5.2400e-04 | mlr 5.2400e-04 | mem 13.73/16.12 GiB step 262 | train 5.483427 | val 5.900655 | tok/s 2631762.10 | norm 3.3075 | dt 199.22 ms | lr 5.2600e-04 | mlr 5.2600e-04 | mem 13.73/16.12 GiB step 263 | train 5.479319 | val 5.900655 | tok/s 2621977.45 | norm 3.6869 | dt 199.96 ms | lr 5.2800e-04 | mlr 5.2800e-04 | mem 13.73/16.12 GiB step 264 | train 5.425973 | val 5.900655 | tok/s 2624537.68 | norm 3.0188 | dt 199.76 ms | lr 5.3000e-04 | mlr 5.3000e-04 | mem 13.73/16.12 GiB step 265 | train 5.522085 | val 5.900655 | tok/s 2628845.21 | norm 3.3616 | dt 199.44 ms | lr 5.3200e-04 | mlr 5.3200e-04 | mem 13.73/16.12 GiB step 266 | train 5.416916 | val 5.900655 | tok/s 2627137.09 | norm 3.4523 | dt 199.57 ms | lr 5.3400e-04 | mlr 5.3400e-04 | mem 13.73/16.12 GiB step 267 | train 5.386157 | val 5.900655 | tok/s 2627884.69 | norm 3.3229 | dt 199.51 ms | lr 5.3600e-04 | mlr 5.3600e-04 | mem 13.73/16.12 GiB step 268 | train 5.436084 | val 5.900655 | tok/s 2625740.08 | norm 3.2848 | dt 199.67 ms | lr 5.3800e-04 | mlr 5.3800e-04 | mem 13.73/16.12 GiB step 269 | train 5.358403 | val 5.900655 | tok/s 2628777.00 | norm 2.7211 | dt 199.44 ms | lr 5.4000e-04 | mlr 5.4000e-04 | mem 13.73/16.12 GiB step 270 | train 5.376176 | val 5.900655 | tok/s 2628173.63 | norm 3.0952 | dt 199.49 ms | lr 5.4200e-04 | mlr 5.4200e-04 | mem 13.73/16.12 GiB step 271 | train 5.408073 | val 5.900655 | tok/s 2629248.55 | norm 3.3415 | dt 199.41 ms | lr 5.4400e-04 | mlr 5.4400e-04 | mem 13.73/16.12 GiB step 272 | train 5.379280 | val 5.900655 | tok/s 2633819.13 | norm 3.6519 | dt 199.06 ms | lr 5.4600e-04 | mlr 5.4600e-04 | mem 13.73/16.12 GiB step 273 | train 5.408594 | val 5.900655 | tok/s 2625299.53 | norm 3.4653 | dt 199.71 ms | lr 5.4800e-04 | mlr 5.4800e-04 | mem 13.73/16.12 GiB step 274 | train 5.368109 | val 5.900655 | tok/s 2626055.62 | norm 3.6363 | dt 199.65 ms | lr 5.5000e-04 | mlr 5.5000e-04 | mem 13.73/16.12 GiB step 275 | train 5.342643 | val 5.900655 | tok/s 2627403.84 | norm 3.6823 | dt 199.55 ms | lr 5.5200e-04 | mlr 5.5200e-04 | mem 13.73/16.12 GiB step 276 | train 5.371880 | val 5.900655 | tok/s 2628692.62 | norm 3.5545 | dt 199.45 ms | lr 5.5400e-04 | mlr 5.5400e-04 | mem 13.73/16.12 GiB step 277 | train 5.397832 | val 5.900655 | tok/s 2631936.27 | norm 3.7456 | dt 199.20 ms | lr 5.5600e-04 | mlr 5.5600e-04 | mem 13.73/16.12 GiB step 278 | train 5.325366 | val 5.900655 | tok/s 2624508.89 | norm 3.3540 | dt 199.77 ms | lr 5.5800e-04 | mlr 5.5800e-04 | mem 13.73/16.12 GiB step 279 | train 5.592172 | val 5.900655 | tok/s 2629826.53 | norm 3.6817 | dt 199.36 ms | lr 5.6000e-04 | mlr 5.6000e-04 | mem 13.73/16.12 GiB step 280 | train 5.525098 | val 5.900655 | tok/s 2632255.65 | norm 3.0225 | dt 199.18 ms | lr 5.6200e-04 | mlr 5.6200e-04 | mem 13.73/16.12 GiB step 281 | train 5.330512 | val 5.900655 | tok/s 2626127.83 | norm 3.3410 | dt 199.64 ms | lr 5.6400e-04 | mlr 5.6400e-04 | mem 13.73/16.12 GiB step 282 | train 5.345813 | val 5.900655 | tok/s 2630154.60 | norm 3.6837 | dt 199.34 ms | lr 5.6600e-04 | mlr 5.6600e-04 | mem 13.73/16.12 GiB step 283 | train 5.273922 | val 5.900655 | tok/s 2616740.62 | norm 3.7303 | dt 200.36 ms | lr 5.6800e-04 | mlr 5.6800e-04 | mem 13.73/16.12 GiB step 284 | train 5.325618 | val 5.900655 | tok/s 2625896.41 | norm 4.0024 | dt 199.66 ms | lr 5.7000e-04 | mlr 5.7000e-04 | mem 13.73/16.12 GiB step 285 | train 5.411131 | val 5.900655 | tok/s 2631571.88 | norm 3.5273 | dt 199.23 ms | lr 5.7200e-04 | mlr 5.7200e-04 | mem 13.73/16.12 GiB step 286 | train 5.233435 | val 5.900655 | tok/s 2510470.60 | norm 3.3168 | dt 208.84 ms | lr 5.7400e-04 | mlr 5.7400e-04 | mem 13.73/16.12 GiB step 287 | train 5.243309 | val 5.900655 | tok/s 2630286.69 | norm 3.1555 | dt 199.33 ms | lr 5.7600e-04 | mlr 5.7600e-04 | mem 13.73/16.12 GiB step 288 | train 5.318348 | val 5.900655 | tok/s 2627784.54 | norm 3.4616 | dt 199.52 ms | lr 5.7800e-04 | mlr 5.7800e-04 | mem 13.73/16.12 GiB step 289 | train 5.281057 | val 5.900655 | tok/s 2627532.95 | norm 3.3918 | dt 199.54 ms | lr 5.8000e-04 | mlr 5.8000e-04 | mem 13.73/16.12 GiB step 290 | train 5.240214 | val 5.900655 | tok/s 2609525.55 | norm 3.0355 | dt 200.91 ms | lr 5.8200e-04 | mlr 5.8200e-04 | mem 13.73/16.12 GiB step 291 | train 5.287983 | val 5.900655 | tok/s 2626156.18 | norm 2.8270 | dt 199.64 ms | lr 5.8400e-04 | mlr 5.8400e-04 | mem 13.73/16.12 GiB step 292 | train 5.276385 | val 5.900655 | tok/s 2625578.34 | norm 2.5599 | dt 199.68 ms | lr 5.8600e-04 | mlr 5.8600e-04 | mem 13.73/16.12 GiB step 293 | train 5.168650 | val 5.900655 | tok/s 2626708.98 | norm 3.0127 | dt 199.60 ms | lr 5.8800e-04 | mlr 5.8800e-04 | mem 13.73/16.12 GiB step 294 | train 5.200440 | val 5.900655 | tok/s 2625941.34 | norm 4.5043 | dt 199.66 ms | lr 5.9000e-04 | mlr 5.9000e-04 | mem 13.73/16.12 GiB step 295 | train 5.233963 | val 5.900655 | tok/s 2622887.15 | norm 3.2321 | dt 199.89 ms | lr 5.9200e-04 | mlr 5.9200e-04 | mem 13.73/16.12 GiB step 296 | train 5.367909 | val 5.900655 | tok/s 2623845.10 | norm 2.9677 | dt 199.82 ms | lr 5.9400e-04 | mlr 5.9400e-04 | mem 13.73/16.12 GiB step 297 | train 5.091152 | val 5.900655 | tok/s 2629264.63 | norm 2.6048 | dt 199.40 ms | lr 5.9600e-04 | mlr 5.9600e-04 | mem 13.73/16.12 GiB step 298 | train 5.262006 | val 5.900655 | tok/s 2617022.53 | norm 3.1223 | dt 200.34 ms | lr 5.9800e-04 | mlr 5.9800e-04 | mem 13.73/16.12 GiB step 299 | train 5.258531 | val 5.900655 | tok/s 2619192.85 | norm 2.7532 | dt 200.17 ms | lr 6.0000e-04 | mlr 6.0000e-04 | mem 13.73/16.12 GiB step 300 | train 5.125353 | val 5.900655 | tok/s 2627632.17 | norm 2.8681 | dt 199.53 ms | lr 6.0200e-04 | mlr 6.0200e-04 | mem 13.73/16.12 GiB step 301 | train 5.181551 | val 5.900655 | tok/s 2626653.09 | norm 2.7377 | dt 199.60 ms | lr 6.0400e-04 | mlr 6.0400e-04 | mem 13.73/16.12 GiB step 302 | train 5.246376 | val 5.900655 | tok/s 2628560.80 | norm 2.8970 | dt 199.46 ms | lr 6.0600e-04 | mlr 6.0600e-04 | mem 13.73/16.12 GiB step 303 | train 5.091617 | val 5.900655 | tok/s 2623546.64 | norm 2.4734 | dt 199.84 ms | lr 6.0800e-04 | mlr 6.0800e-04 | mem 13.73/16.12 GiB step 304 | train 5.248963 | val 5.900655 | tok/s 2592607.02 | norm 3.1131 | dt 202.22 ms | lr 6.1000e-04 | mlr 6.1000e-04 | mem 13.73/16.12 GiB step 305 | train 5.117635 | val 5.900655 | tok/s 2629458.48 | norm 2.4184 | dt 199.39 ms | lr 6.1200e-04 | mlr 6.1200e-04 | mem 13.73/16.12 GiB step 306 | train 5.062715 | val 5.900655 | tok/s 2627261.71 | norm 2.9531 | dt 199.56 ms | lr 6.1400e-04 | mlr 6.1400e-04 | mem 13.73/16.12 GiB step 307 | train 5.262158 | val 5.900655 | tok/s 2627760.95 | norm 2.8057 | dt 199.52 ms | lr 6.1600e-04 | mlr 6.1600e-04 | mem 13.73/16.12 GiB step 308 | train 5.108143 | val 5.900655 | tok/s 2627538.74 | norm 2.8136 | dt 199.54 ms | lr 6.1800e-04 | mlr 6.1800e-04 | mem 13.73/16.12 GiB step 309 | train 5.168850 | val 5.900655 | tok/s 2628273.71 | norm 2.8718 | dt 199.48 ms | lr 6.2000e-04 | mlr 6.2000e-04 | mem 13.73/16.12 GiB step 310 | train 5.160002 | val 5.900655 | tok/s 2629330.29 | norm 3.0394 | dt 199.40 ms | lr 6.2200e-04 | mlr 6.2200e-04 | mem 13.73/16.12 GiB step 311 | train 5.235448 | val 5.900655 | tok/s 2628217.41 | norm 2.9367 | dt 199.48 ms | lr 6.2400e-04 | mlr 6.2400e-04 | mem 13.73/16.12 GiB step 312 | train 5.106264 | val 5.900655 | tok/s 2631898.73 | norm 2.6781 | dt 199.21 ms | lr 6.2600e-04 | mlr 6.2600e-04 | mem 13.73/16.12 GiB step 313 | train 5.078849 | val 5.900655 | tok/s 2625844.08 | norm 2.3941 | dt 199.66 ms | lr 6.2800e-04 | mlr 6.2800e-04 | mem 13.73/16.12 GiB step 314 | train 5.073913 | val 5.900655 | tok/s 2632296.58 | norm 2.3977 | dt 199.18 ms | lr 6.3000e-04 | mlr 6.3000e-04 | mem 13.73/16.12 GiB step 315 | train 5.142304 | val 5.900655 | tok/s 2492027.86 | norm 2.6790 | dt 210.39 ms | lr 6.3200e-04 | mlr 6.3200e-04 | mem 13.73/16.12 GiB step 316 | train 5.212749 | val 5.900655 | tok/s 2605413.15 | norm 2.6386 | dt 201.23 ms | lr 6.3400e-04 | mlr 6.3400e-04 | mem 13.73/16.12 GiB step 317 | train 5.169905 | val 5.900655 | tok/s 2628742.56 | norm 3.3870 | dt 199.44 ms | lr 6.3600e-04 | mlr 6.3600e-04 | mem 13.73/16.12 GiB step 318 | train 5.085989 | val 5.900655 | tok/s 2624651.17 | norm 3.1203 | dt 199.76 ms | lr 6.3800e-04 | mlr 6.3800e-04 | mem 13.73/16.12 GiB step 319 | train 5.091405 | val 5.900655 | tok/s 2629330.83 | norm 3.4046 | dt 199.40 ms | lr 6.4000e-04 | mlr 6.4000e-04 | mem 13.73/16.12 GiB step 320 | train 5.156659 | val 5.900655 | tok/s 2628370.63 | norm 3.0167 | dt 199.47 ms | lr 6.4200e-04 | mlr 6.4200e-04 | mem 13.73/16.12 GiB step 321 | train 5.086616 | val 5.900655 | tok/s 2550971.81 | norm 2.7746 | dt 205.52 ms | lr 6.4400e-04 | mlr 6.4400e-04 | mem 13.73/16.12 GiB step 322 | train 5.062883 | val 5.900655 | tok/s 2589409.38 | norm 2.8275 | dt 202.47 ms | lr 6.4600e-04 | mlr 6.4600e-04 | mem 13.73/16.12 GiB step 323 | train 5.127975 | val 5.900655 | tok/s 2563069.58 | norm 2.7282 | dt 204.55 ms | lr 6.4800e-04 | mlr 6.4800e-04 | mem 13.73/16.12 GiB step 324 | train 5.033067 | val 5.900655 | tok/s 2630154.33 | norm 2.4290 | dt 199.34 ms | lr 6.5000e-04 | mlr 6.5000e-04 | mem 13.73/16.12 GiB step 325 | train 5.068864 | val 5.900655 | tok/s 2623509.15 | norm 2.4906 | dt 199.84 ms | lr 6.5200e-04 | mlr 6.5200e-04 | mem 13.73/16.12 GiB step 326 | train 4.972438 | val 5.900655 | tok/s 2624884.90 | norm 2.3800 | dt 199.74 ms | lr 6.5400e-04 | mlr 6.5400e-04 | mem 13.73/16.12 GiB step 327 | train 5.012113 | val 5.900655 | tok/s 2626512.28 | norm 2.4899 | dt 199.61 ms | lr 6.5600e-04 | mlr 6.5600e-04 | mem 13.73/16.12 GiB step 328 | train 5.055187 | val 5.900655 | tok/s 2623318.96 | norm 2.3546 | dt 199.86 ms | lr 6.5800e-04 | mlr 6.5800e-04 | mem 13.73/16.12 GiB step 329 | train 5.035626 | val 5.900655 | tok/s 2471244.88 | norm 2.8072 | dt 212.16 ms | lr 6.6000e-04 | mlr 6.6000e-04 | mem 13.73/16.12 GiB step 330 | train 5.002388 | val 5.900655 | tok/s 2621949.27 | norm 2.9458 | dt 199.96 ms | lr 6.6200e-04 | mlr 6.6200e-04 | mem 13.73/16.12 GiB step 331 | train 4.961073 | val 5.900655 | tok/s 2623591.71 | norm 3.0254 | dt 199.84 ms | lr 6.6400e-04 | mlr 6.6400e-04 | mem 13.73/16.12 GiB step 332 | train 5.015147 | val 5.900655 | tok/s 2626952.98 | norm 2.5945 | dt 199.58 ms | lr 6.6600e-04 | mlr 6.6600e-04 | mem 13.73/16.12 GiB step 333 | train 4.989436 | val 5.900655 | tok/s 2621663.48 | norm 2.9187 | dt 199.98 ms | lr 6.6800e-04 | mlr 6.6800e-04 | mem 13.73/16.12 GiB step 334 | train 5.031321 | val 5.900655 | tok/s 2621584.75 | norm 2.4207 | dt 199.99 ms | lr 6.7000e-04 | mlr 6.7000e-04 | mem 13.73/16.12 GiB step 335 | train 5.088533 | val 5.900655 | tok/s 2623658.42 | norm 2.5814 | dt 199.83 ms | lr 6.7200e-04 | mlr 6.7200e-04 | mem 13.73/16.12 GiB step 336 | train 4.979982 | val 5.900655 | tok/s 2626550.23 | norm 2.1852 | dt 199.61 ms | lr 6.7400e-04 | mlr 6.7400e-04 | mem 13.73/16.12 GiB step 337 | train 5.055473 | val 5.900655 | tok/s 2627295.20 | norm 2.1662 | dt 199.55 ms | lr 6.7600e-04 | mlr 6.7600e-04 | mem 13.73/16.12 GiB step 338 | train 4.973167 | val 5.900655 | tok/s 2627279.63 | norm 2.2513 | dt 199.56 ms | lr 6.7800e-04 | mlr 6.7800e-04 | mem 13.73/16.12 GiB step 339 | train 5.057189 | val 5.900655 | tok/s 2628850.99 | norm 2.5777 | dt 199.44 ms | lr 6.8000e-04 | mlr 6.8000e-04 | mem 13.73/16.12 GiB step 340 | train 4.990711 | val 5.900655 | tok/s 2593450.78 | norm 2.2500 | dt 202.16 ms | lr 6.8200e-04 | mlr 6.8200e-04 | mem 13.73/16.12 GiB step 341 | train 5.177681 | val 5.900655 | tok/s 2604881.99 | norm 2.6584 | dt 201.27 ms | lr 6.8400e-04 | mlr 6.8400e-04 | mem 13.73/16.12 GiB step 342 | train 4.877884 | val 5.900655 | tok/s 2631558.10 | norm 1.9104 | dt 199.23 ms | lr 6.8600e-04 | mlr 6.8600e-04 | mem 13.73/16.12 GiB step 343 | train 4.966718 | val 5.900655 | tok/s 2625041.86 | norm 2.4993 | dt 199.73 ms | lr 6.8800e-04 | mlr 6.8800e-04 | mem 13.73/16.12 GiB step 344 | train 4.916270 | val 5.900655 | tok/s 2626543.43 | norm 2.0280 | dt 199.61 ms | lr 6.9000e-04 | mlr 6.9000e-04 | mem 13.73/16.12 GiB step 345 | train 4.917988 | val 5.900655 | tok/s 2621256.58 | norm 2.6362 | dt 200.01 ms | lr 6.9200e-04 | mlr 6.9200e-04 | mem 13.73/16.12 GiB step 346 | train 4.961685 | val 5.900655 | tok/s 2623514.69 | norm 2.5882 | dt 199.84 ms | lr 6.9400e-04 | mlr 6.9400e-04 | mem 13.73/16.12 GiB step 347 | train 5.064479 | val 5.900655 | tok/s 2620328.81 | norm 2.7467 | dt 200.08 ms | lr 6.9600e-04 | mlr 6.9600e-04 | mem 13.73/16.12 GiB step 348 | train 4.944697 | val 5.900655 | tok/s 2617578.54 | norm 2.8214 | dt 200.30 ms | lr 6.9800e-04 | mlr 6.9800e-04 | mem 13.73/16.12 GiB step 349 | train 4.981539 | val 5.900655 | tok/s 2629165.71 | norm 2.5083 | dt 199.41 ms | lr 7.0000e-04 | mlr 7.0000e-04 | mem 13.73/16.12 GiB step 350 | train 4.981514 | val 5.900655 | tok/s 2619711.16 | norm 2.5548 | dt 200.13 ms | lr 7.0200e-04 | mlr 7.0200e-04 | mem 13.73/16.12 GiB step 351 | train 4.904843 | val 5.900655 | tok/s 2624967.08 | norm 2.8100 | dt 199.73 ms | lr 7.0400e-04 | mlr 7.0400e-04 | mem 13.73/16.12 GiB step 352 | train 5.003916 | val 5.900655 | tok/s 2629391.07 | norm 2.7609 | dt 199.40 ms | lr 7.0600e-04 | mlr 7.0600e-04 | mem 13.73/16.12 GiB step 353 | train 4.947900 | val 5.900655 | tok/s 2628310.35 | norm 2.6178 | dt 199.48 ms | lr 7.0800e-04 | mlr 7.0800e-04 | mem 13.73/16.12 GiB step 354 | train 4.890911 | val 5.900655 | tok/s 2631680.24 | norm 2.9895 | dt 199.22 ms | lr 7.1000e-04 | mlr 7.1000e-04 | mem 13.73/16.12 GiB step 355 | train 4.910048 | val 5.900655 | tok/s 2622626.12 | norm 2.5772 | dt 199.91 ms | lr 7.1200e-04 | mlr 7.1200e-04 | mem 13.73/16.12 GiB step 356 | train 4.923357 | val 5.900655 | tok/s 2620953.26 | norm 2.4048 | dt 200.04 ms | lr 7.1400e-04 | mlr 7.1400e-04 | mem 13.73/16.12 GiB step 357 | train 4.867208 | val 5.900655 | tok/s 2611998.59 | norm 2.2639 | dt 200.72 ms | lr 7.1600e-04 | mlr 7.1600e-04 | mem 13.73/16.12 GiB step 358 | train 5.008941 | val 5.900655 | tok/s 2627086.87 | norm 2.5090 | dt 199.57 ms | lr 7.1800e-04 | mlr 7.1800e-04 | mem 13.73/16.12 GiB step 359 | train 4.899815 | val 5.900655 | tok/s 2627475.05 | norm 2.3160 | dt 199.54 ms | lr 7.2000e-04 | mlr 7.2000e-04 | mem 13.73/16.12 GiB step 360 | train 4.984812 | val 5.900655 | tok/s 2623694.84 | norm 2.2575 | dt 199.83 ms | lr 7.2200e-04 | mlr 7.2200e-04 | mem 13.73/16.12 GiB step 361 | train 4.808881 | val 5.900655 | tok/s 2626818.87 | norm 1.9990 | dt 199.59 ms | lr 7.2400e-04 | mlr 7.2400e-04 | mem 13.73/16.12 GiB step 362 | train 4.818131 | val 5.900655 | tok/s 2633110.94 | norm 2.4197 | dt 199.11 ms | lr 7.2600e-04 | mlr 7.2600e-04 | mem 13.73/16.12 GiB step 363 | train 4.928514 | val 5.900655 | tok/s 2629477.80 | norm 2.1169 | dt 199.39 ms | lr 7.2800e-04 | mlr 7.2800e-04 | mem 13.73/16.12 GiB step 364 | train 4.855689 | val 5.900655 | tok/s 2623693.53 | norm 2.0818 | dt 199.83 ms | lr 7.3000e-04 | mlr 7.3000e-04 | mem 13.73/16.12 GiB step 365 | train 4.862837 | val 5.900655 | tok/s 2622941.56 | norm 1.9063 | dt 199.89 ms | lr 7.3200e-04 | mlr 7.3200e-04 | mem 13.73/16.12 GiB step 366 | train 4.914121 | val 5.900655 | tok/s 2619149.44 | norm 1.8188 | dt 200.17 ms | lr 7.3400e-04 | mlr 7.3400e-04 | mem 13.73/16.12 GiB step 367 | train 4.895490 | val 5.900655 | tok/s 2629675.34 | norm 1.9710 | dt 199.37 ms | lr 7.3600e-04 | mlr 7.3600e-04 | mem 13.73/16.12 GiB step 368 | train 4.808915 | val 5.900655 | tok/s 2625161.07 | norm 2.2985 | dt 199.72 ms | lr 7.3800e-04 | mlr 7.3800e-04 | mem 13.73/16.12 GiB step 369 | train 4.857641 | val 5.900655 | tok/s 2626090.61 | norm 2.0079 | dt 199.65 ms | lr 7.4000e-04 | mlr 7.4000e-04 | mem 13.73/16.12 GiB step 370 | train 4.845081 | val 5.900655 | tok/s 2621666.30 | norm 2.0401 | dt 199.98 ms | lr 7.4200e-04 | mlr 7.4200e-04 | mem 13.73/16.12 GiB step 371 | train 4.795984 | val 5.900655 | tok/s 2619779.78 | norm 1.8383 | dt 200.13 ms | lr 7.4400e-04 | mlr 7.4400e-04 | mem 13.73/16.12 GiB step 372 | train 4.808123 | val 5.900655 | tok/s 2628606.37 | norm 2.0702 | dt 199.45 ms | lr 7.4600e-04 | mlr 7.4600e-04 | mem 13.73/16.12 GiB step 373 | train 4.735247 | val 5.900655 | tok/s 2626897.06 | norm 2.2821 | dt 199.58 ms | lr 7.4800e-04 | mlr 7.4800e-04 | mem 13.73/16.12 GiB step 374 | train 4.794974 | val 5.900655 | tok/s 2625718.91 | norm 2.1311 | dt 199.67 ms | lr 7.5000e-04 | mlr 7.5000e-04 | mem 13.73/16.12 GiB step 375 | train 4.747013 | val 5.900655 | tok/s 2622438.58 | norm 2.2380 | dt 199.92 ms | lr 7.5200e-04 | mlr 7.5200e-04 | mem 13.73/16.12 GiB step 376 | train 4.764320 | val 5.900655 | tok/s 2621766.87 | norm 2.0660 | dt 199.98 ms | lr 7.5400e-04 | mlr 7.5400e-04 | mem 13.73/16.12 GiB step 377 | train 4.623309 | val 5.900655 | tok/s 2629330.08 | norm 2.1707 | dt 199.40 ms | lr 7.5600e-04 | mlr 7.5600e-04 | mem 13.73/16.12 GiB step 378 | train 4.709623 | val 5.900655 | tok/s 2626510.36 | norm 1.8521 | dt 199.61 ms | lr 7.5800e-04 | mlr 7.5800e-04 | mem 13.73/16.12 GiB step 379 | train 4.824535 | val 5.900655 | tok/s 2630448.81 | norm 1.9554 | dt 199.32 ms | lr 7.6000e-04 | mlr 7.6000e-04 | mem 13.73/16.12 GiB step 380 | train 4.699613 | val 5.900655 | tok/s 2618067.62 | norm 1.8196 | dt 200.26 ms | lr 7.6200e-04 | mlr 7.6200e-04 | mem 13.73/16.12 GiB step 381 | train 4.766737 | val 5.900655 | tok/s 2622854.72 | norm 2.1111 | dt 199.89 ms | lr 7.6400e-04 | mlr 7.6400e-04 | mem 13.73/16.12 GiB step 382 | train 4.739814 | val 5.900655 | tok/s 2625774.52 | norm 2.1850 | dt 199.67 ms | lr 7.6600e-04 | mlr 7.6600e-04 | mem 13.73/16.12 GiB step 383 | train 4.727332 | val 5.900655 | tok/s 2624403.41 | norm 2.3764 | dt 199.77 ms | lr 7.6800e-04 | mlr 7.6800e-04 | mem 13.73/16.12 GiB step 384 | train 4.766382 | val 5.900655 | tok/s 2630432.26 | norm 1.8958 | dt 199.32 ms | lr 7.7000e-04 | mlr 7.7000e-04 | mem 13.73/16.12 GiB step 385 | train 4.731227 | val 5.900655 | tok/s 2621652.04 | norm 2.0556 | dt 199.98 ms | lr 7.7200e-04 | mlr 7.7200e-04 | mem 13.73/16.12 GiB step 386 | train 4.729277 | val 5.900655 | tok/s 2622331.08 | norm 1.8411 | dt 199.93 ms | lr 7.7400e-04 | mlr 7.7400e-04 | mem 13.73/16.12 GiB step 387 | train 4.625721 | val 5.900655 | tok/s 2631939.16 | norm 2.6563 | dt 199.20 ms | lr 7.7600e-04 | mlr 7.7600e-04 | mem 13.73/16.12 GiB step 388 | train 4.693658 | val 5.900655 | tok/s 2629875.95 | norm 1.8033 | dt 199.36 ms | lr 7.7800e-04 | mlr 7.7800e-04 | mem 13.73/16.12 GiB step 389 | train 4.810159 | val 5.900655 | tok/s 2629996.42 | norm 2.2266 | dt 199.35 ms | lr 7.8000e-04 | mlr 7.8000e-04 | mem 13.73/16.12 GiB step 390 | train 4.802468 | val 5.900655 | tok/s 2622546.92 | norm 1.9995 | dt 199.92 ms | lr 7.8200e-04 | mlr 7.8200e-04 | mem 13.73/16.12 GiB step 391 | train 4.820704 | val 5.900655 | tok/s 2627671.26 | norm 2.3094 | dt 199.53 ms | lr 7.8400e-04 | mlr 7.8400e-04 | mem 13.73/16.12 GiB step 392 | train 4.740274 | val 5.900655 | tok/s 2630215.50 | norm 1.6661 | dt 199.33 ms | lr 7.8600e-04 | mlr 7.8600e-04 | mem 13.73/16.12 GiB step 393 | train 4.674594 | val 5.900655 | tok/s 2624233.18 | norm 2.0143 | dt 199.79 ms | lr 7.8800e-04 | mlr 7.8800e-04 | mem 13.73/16.12 GiB step 394 | train 4.754473 | val 5.900655 | tok/s 2627518.59 | norm 1.6621 | dt 199.54 ms | lr 7.9000e-04 | mlr 7.9000e-04 | mem 13.73/16.12 GiB step 395 | train 4.573208 | val 5.900655 | tok/s 2620606.76 | norm 1.8730 | dt 200.06 ms | lr 7.9200e-04 | mlr 7.9200e-04 | mem 13.73/16.12 GiB step 396 | train 4.660645 | val 5.900655 | tok/s 2622923.92 | norm 1.4623 | dt 199.89 ms | lr 7.9400e-04 | mlr 7.9400e-04 | mem 13.73/16.12 GiB step 397 | train 4.681906 | val 5.900655 | tok/s 2629430.48 | norm 2.0869 | dt 199.39 ms | lr 7.9600e-04 | mlr 7.9600e-04 | mem 13.73/16.12 GiB step 398 | train 4.661083 | val 5.900655 | tok/s 2629325.56 | norm 1.8789 | dt 199.40 ms | lr 7.9800e-04 | mlr 7.9800e-04 | mem 13.73/16.12 GiB step 399 | train 4.665628 | val 5.900655 | tok/s 2628104.43 | norm 2.0578 | dt 199.49 ms | lr 8.0000e-04 | mlr 8.0000e-04 | mem 13.73/16.12 GiB step 400 | train 4.748775 | val 5.900655 | tok/s 2624181.40 | norm 1.8200 | dt 199.79 ms | lr 8.0200e-04 | mlr 8.0200e-04 | mem 13.73/16.12 GiB step 401 | train 4.635567 | val 5.900655 | tok/s 2626693.10 | norm 1.8502 | dt 199.60 ms | lr 8.0400e-04 | mlr 8.0400e-04 | mem 13.73/16.12 GiB step 402 | train 4.776889 | val 5.900655 | tok/s 2626791.37 | norm 1.9391 | dt 199.59 ms | lr 8.0600e-04 | mlr 8.0600e-04 | mem 13.73/16.12 GiB step 403 | train 4.649570 | val 5.900655 | tok/s 2622810.37 | norm 1.9875 | dt 199.90 ms | lr 8.0800e-04 | mlr 8.0800e-04 | mem 13.73/16.12 GiB step 404 | train 4.674883 | val 5.900655 | tok/s 2630112.27 | norm 1.9011 | dt 199.34 ms | lr 8.1000e-04 | mlr 8.1000e-04 | mem 13.73/16.12 GiB step 405 | train 4.631442 | val 5.900655 | tok/s 2627339.43 | norm 1.7210 | dt 199.55 ms | lr 8.1200e-04 | mlr 8.1200e-04 | mem 13.73/16.12 GiB step 406 | train 4.767846 | val 5.900655 | tok/s 2626283.34 | norm 1.8071 | dt 199.63 ms | lr 8.1400e-04 | mlr 8.1400e-04 | mem 13.73/16.12 GiB step 407 | train 4.562179 | val 5.900655 | tok/s 2625784.15 | norm 1.6908 | dt 199.67 ms | lr 8.1600e-04 | mlr 8.1600e-04 | mem 13.73/16.12 GiB step 408 | train 4.771635 | val 5.900655 | tok/s 2626774.23 | norm 2.2910 | dt 199.59 ms | lr 8.1800e-04 | mlr 8.1800e-04 | mem 13.73/16.12 GiB step 409 | train 4.630647 | val 5.900655 | tok/s 2628999.16 | norm 1.8920 | dt 199.42 ms | lr 8.2000e-04 | mlr 8.2000e-04 | mem 13.73/16.12 GiB step 410 | train 4.664135 | val 5.900655 | tok/s 2622841.56 | norm 2.1339 | dt 199.89 ms | lr 8.2200e-04 | mlr 8.2200e-04 | mem 13.73/16.12 GiB step 411 | train 4.668747 | val 5.900655 | tok/s 2625670.26 | norm 1.6240 | dt 199.68 ms | lr 8.2400e-04 | mlr 8.2400e-04 | mem 13.73/16.12 GiB step 412 | train 4.656586 | val 5.900655 | tok/s 2628939.64 | norm 1.8960 | dt 199.43 ms | lr 8.2600e-04 | mlr 8.2600e-04 | mem 13.73/16.12 GiB step 413 | train 4.539774 | val 5.900655 | tok/s 2624921.17 | norm 1.5717 | dt 199.73 ms | lr 8.2800e-04 | mlr 8.2800e-04 | mem 13.73/16.12 GiB step 414 | train 4.627558 | val 5.900655 | tok/s 2625115.81 | norm 1.7920 | dt 199.72 ms | lr 8.3000e-04 | mlr 8.3000e-04 | mem 13.73/16.12 GiB step 415 | train 4.686447 | val 5.900655 | tok/s 2626236.01 | norm 1.5180 | dt 199.63 ms | lr 8.3200e-04 | mlr 8.3200e-04 | mem 13.73/16.12 GiB step 416 | train 4.610098 | val 5.900655 | tok/s 2545889.20 | norm 1.9839 | dt 205.94 ms | lr 8.3400e-04 | mlr 8.3400e-04 | mem 13.73/16.12 GiB step 417 | train 4.625316 | val 5.900655 | tok/s 2626656.36 | norm 1.7342 | dt 199.60 ms | lr 8.3600e-04 | mlr 8.3600e-04 | mem 13.73/16.12 GiB step 418 | train 4.544060 | val 5.900655 | tok/s 2624197.23 | norm 1.7000 | dt 199.79 ms | lr 8.3800e-04 | mlr 8.3800e-04 | mem 13.73/16.12 GiB step 419 | train 4.545091 | val 5.900655 | tok/s 2627477.17 | norm 1.5376 | dt 199.54 ms | lr 8.4000e-04 | mlr 8.4000e-04 | mem 13.73/16.12 GiB step 420 | train 4.655261 | val 5.900655 | tok/s 2622809.85 | norm 1.4376 | dt 199.90 ms | lr 8.4200e-04 | mlr 8.4200e-04 | mem 13.73/16.12 GiB step 421 | train 4.594357 | val 5.900655 | tok/s 2625588.05 | norm 1.7746 | dt 199.68 ms | lr 8.4400e-04 | mlr 8.4400e-04 | mem 13.73/16.12 GiB step 422 | train 4.600797 | val 5.900655 | tok/s 2623688.28 | norm 1.6160 | dt 199.83 ms | lr 8.4600e-04 | mlr 8.4600e-04 | mem 13.73/16.12 GiB step 423 | train 4.614292 | val 5.900655 | tok/s 2621673.13 | norm 1.5845 | dt 199.98 ms | lr 8.4800e-04 | mlr 8.4800e-04 | mem 13.73/16.12 GiB step 424 | train 4.610989 | val 5.900655 | tok/s 2625342.37 | norm 1.4697 | dt 199.70 ms | lr 8.5000e-04 | mlr 8.5000e-04 | mem 13.73/16.12 GiB step 425 | train 4.567548 | val 5.900655 | tok/s 2623032.71 | norm 1.8933 | dt 199.88 ms | lr 8.5200e-04 | mlr 8.5200e-04 | mem 13.73/16.12 GiB step 426 | train 4.628592 | val 5.900655 | tok/s 2630963.42 | norm 1.5473 | dt 199.28 ms | lr 8.5400e-04 | mlr 8.5400e-04 | mem 13.73/16.12 GiB step 427 | train 4.539302 | val 5.900655 | tok/s 2631109.91 | norm 1.7381 | dt 199.26 ms | lr 8.5600e-04 | mlr 8.5600e-04 | mem 13.73/16.12 GiB step 428 | train 4.635339 | val 5.900655 | tok/s 2625305.67 | norm 1.4276 | dt 199.71 ms | lr 8.5800e-04 | mlr 8.5800e-04 | mem 13.73/16.12 GiB step 429 | train 4.505034 | val 5.900655 | tok/s 2630430.87 | norm 1.6459 | dt 199.32 ms | lr 8.6000e-04 | mlr 8.6000e-04 | mem 13.73/16.12 GiB step 430 | train 4.481323 | val 5.900655 | tok/s 2629609.24 | norm 1.6544 | dt 199.38 ms | lr 8.6200e-04 | mlr 8.6200e-04 | mem 13.73/16.12 GiB step 431 | train 4.632572 | val 5.900655 | tok/s 2629186.09 | norm 1.6035 | dt 199.41 ms | lr 8.6400e-04 | mlr 8.6400e-04 | mem 13.73/16.12 GiB step 432 | train 4.580534 | val 5.900655 | tok/s 2622330.85 | norm 1.6461 | dt 199.93 ms | lr 8.6600e-04 | mlr 8.6600e-04 | mem 13.73/16.12 GiB step 433 | train 4.565715 | val 5.900655 | tok/s 2616164.31 | norm 1.7625 | dt 200.40 ms | lr 8.6800e-04 | mlr 8.6800e-04 | mem 13.73/16.12 GiB step 434 | train 4.636227 | val 5.900655 | tok/s 2574982.97 | norm 1.6428 | dt 203.61 ms | lr 8.7000e-04 | mlr 8.7000e-04 | mem 13.73/16.12 GiB step 435 | train 4.645899 | val 5.900655 | tok/s 2523518.22 | norm 1.6448 | dt 207.76 ms | lr 8.7200e-04 | mlr 8.7200e-04 | mem 13.73/16.12 GiB step 436 | train 4.503733 | val 5.900655 | tok/s 2624496.99 | norm 1.5411 | dt 199.77 ms | lr 8.7400e-04 | mlr 8.7400e-04 | mem 13.73/16.12 GiB step 437 | train 4.486540 | val 5.900655 | tok/s 2623837.57 | norm 1.4733 | dt 199.82 ms | lr 8.7600e-04 | mlr 8.7600e-04 | mem 13.73/16.12 GiB step 438 | train 4.621307 | val 5.900655 | tok/s 2621951.33 | norm 1.7299 | dt 199.96 ms | lr 8.7800e-04 | mlr 8.7800e-04 | mem 13.73/16.12 GiB step 439 | train 4.539534 | val 5.900655 | tok/s 2624688.49 | norm 1.5281 | dt 199.75 ms | lr 8.8000e-04 | mlr 8.8000e-04 | mem 13.73/16.12 GiB step 440 | train 4.499877 | val 5.900655 | tok/s 2624928.33 | norm 1.5944 | dt 199.73 ms | lr 8.8200e-04 | mlr 8.8200e-04 | mem 13.73/16.12 GiB step 441 | train 4.535150 | val 5.900655 | tok/s 2486981.06 | norm 1.3774 | dt 210.81 ms | lr 8.8400e-04 | mlr 8.8400e-04 | mem 13.73/16.12 GiB step 442 | train 4.578764 | val 5.900655 | tok/s 2604265.68 | norm 1.4797 | dt 201.32 ms | lr 8.8600e-04 | mlr 8.8600e-04 | mem 13.73/16.12 GiB step 443 | train 4.528775 | val 5.900655 | tok/s 2617958.78 | norm 1.5682 | dt 200.27 ms | lr 8.8800e-04 | mlr 8.8800e-04 | mem 13.73/16.12 GiB step 444 | train 4.584009 | val 5.900655 | tok/s 2629419.24 | norm 1.8890 | dt 199.39 ms | lr 8.9000e-04 | mlr 8.9000e-04 | mem 13.73/16.12 GiB step 445 | train 4.515054 | val 5.900655 | tok/s 2625320.68 | norm 1.7085 | dt 199.70 ms | lr 8.9200e-04 | mlr 8.9200e-04 | mem 13.73/16.12 GiB step 446 | train 4.523479 | val 5.900655 | tok/s 2624865.86 | norm 1.5526 | dt 199.74 ms | lr 8.9400e-04 | mlr 8.9400e-04 | mem 13.73/16.12 GiB step 447 | train 4.423519 | val 5.900655 | tok/s 2625129.20 | norm 1.5070 | dt 199.72 ms | lr 8.9600e-04 | mlr 8.9600e-04 | mem 13.73/16.12 GiB step 448 | train 4.357097 | val 5.900655 | tok/s 2629744.30 | norm 1.5282 | dt 199.37 ms | lr 8.9800e-04 | mlr 8.9800e-04 | mem 13.73/16.12 GiB step 449 | train 4.575579 | val 5.900655 | tok/s 2627424.84 | norm 1.6200 | dt 199.54 ms | lr 9.0000e-04 | mlr 9.0000e-04 | mem 13.73/16.12 GiB step 450 | train 4.464789 | val 5.900655 | tok/s 2625414.55 | norm 1.9282 | dt 199.70 ms | lr 9.0200e-04 | mlr 9.0200e-04 | mem 13.73/16.12 GiB step 451 | train 4.436450 | val 5.900655 | tok/s 2623757.26 | norm 1.7662 | dt 199.82 ms | lr 9.0400e-04 | mlr 9.0400e-04 | mem 13.73/16.12 GiB step 452 | train 4.529160 | val 5.900655 | tok/s 2618825.98 | norm 1.9203 | dt 200.20 ms | lr 9.0600e-04 | mlr 9.0600e-04 | mem 13.73/16.12 GiB step 453 | train 4.487212 | val 5.900655 | tok/s 2623839.57 | norm 1.4802 | dt 199.82 ms | lr 9.0800e-04 | mlr 9.0800e-04 | mem 13.73/16.12 GiB step 454 | train 4.568995 | val 5.900655 | tok/s 2632143.86 | norm 1.5008 | dt 199.19 ms | lr 9.1000e-04 | mlr 9.1000e-04 | mem 13.73/16.12 GiB step 455 | train 4.517943 | val 5.900655 | tok/s 2625058.72 | norm 1.4289 | dt 199.72 ms | lr 9.1200e-04 | mlr 9.1200e-04 | mem 13.73/16.12 GiB step 456 | train 4.527650 | val 5.900655 | tok/s 2632352.13 | norm 1.5220 | dt 199.17 ms | lr 9.1400e-04 | mlr 9.1400e-04 | mem 13.73/16.12 GiB step 457 | train 4.498005 | val 5.900655 | tok/s 2620710.76 | norm 1.5926 | dt 200.06 ms | lr 9.1600e-04 | mlr 9.1600e-04 | mem 13.73/16.12 GiB step 458 | train 4.387330 | val 5.900655 | tok/s 2627243.15 | norm 1.4455 | dt 199.56 ms | lr 9.1800e-04 | mlr 9.1800e-04 | mem 13.73/16.12 GiB step 459 | train 4.376959 | val 5.900655 | tok/s 2627282.05 | norm 1.4187 | dt 199.56 ms | lr 9.2000e-04 | mlr 9.2000e-04 | mem 13.73/16.12 GiB step 460 | train 4.474253 | val 5.900655 | tok/s 2625589.84 | norm 1.3892 | dt 199.68 ms | lr 9.2200e-04 | mlr 9.2200e-04 | mem 13.73/16.12 GiB step 461 | train 4.414888 | val 5.900655 | tok/s 2628722.59 | norm 1.3681 | dt 199.45 ms | lr 9.2400e-04 | mlr 9.2400e-04 | mem 13.73/16.12 GiB step 462 | train 4.370182 | val 5.900655 | tok/s 2624527.42 | norm 1.6200 | dt 199.76 ms | lr 9.2600e-04 | mlr 9.2600e-04 | mem 13.73/16.12 GiB step 463 | train 4.512601 | val 5.900655 | tok/s 2624687.62 | norm 1.3635 | dt 199.75 ms | lr 9.2800e-04 | mlr 9.2800e-04 | mem 13.73/16.12 GiB step 464 | train 4.414896 | val 5.900655 | tok/s 2506787.72 | norm 1.4647 | dt 209.15 ms | lr 9.3000e-04 | mlr 9.3000e-04 | mem 13.73/16.12 GiB step 465 | train 4.480989 | val 5.900655 | tok/s 2624026.59 | norm 1.2725 | dt 199.80 ms | lr 9.3200e-04 | mlr 9.3200e-04 | mem 13.73/16.12 GiB step 466 | train 4.372841 | val 5.900655 | tok/s 2626223.09 | norm 1.4356 | dt 199.64 ms | lr 9.3400e-04 | mlr 9.3400e-04 | mem 13.73/16.12 GiB step 467 | train 4.472276 | val 5.900655 | tok/s 2621788.53 | norm 1.4143 | dt 199.97 ms | lr 9.3600e-04 | mlr 9.3600e-04 | mem 13.73/16.12 GiB step 468 | train 4.465769 | val 5.900655 | tok/s 2622261.16 | norm 1.4589 | dt 199.94 ms | lr 9.3800e-04 | mlr 9.3800e-04 | mem 13.73/16.12 GiB step 469 | train 4.502142 | val 5.900655 | tok/s 2630963.51 | norm 1.3847 | dt 199.28 ms | lr 9.4000e-04 | mlr 9.4000e-04 | mem 13.73/16.12 GiB step 470 | train 4.379955 | val 5.900655 | tok/s 2624629.77 | norm 1.4600 | dt 199.76 ms | lr 9.4200e-04 | mlr 9.4200e-04 | mem 13.73/16.12 GiB step 471 | train 4.453621 | val 5.900655 | tok/s 2630472.94 | norm 1.3547 | dt 199.31 ms | lr 9.4400e-04 | mlr 9.4400e-04 | mem 13.73/16.12 GiB step 472 | train 4.428102 | val 5.900655 | tok/s 2623333.26 | norm 1.3955 | dt 199.86 ms | lr 9.4600e-04 | mlr 9.4600e-04 | mem 13.73/16.12 GiB step 473 | train 4.291687 | val 5.900655 | tok/s 2624234.63 | norm 1.2466 | dt 199.79 ms | lr 9.4800e-04 | mlr 9.4800e-04 | mem 13.73/16.12 GiB step 474 | train 4.407279 | val 5.900655 | tok/s 2625513.06 | norm 1.2489 | dt 199.69 ms | lr 9.5000e-04 | mlr 9.5000e-04 | mem 13.73/16.12 GiB step 475 | train 4.447629 | val 5.900655 | tok/s 2602869.72 | norm 1.3493 | dt 201.43 ms | lr 9.5200e-04 | mlr 9.5200e-04 | mem 13.73/16.12 GiB step 476 | train 4.379354 | val 5.900655 | tok/s 2629127.33 | norm 1.4239 | dt 199.42 ms | lr 9.5400e-04 | mlr 9.5400e-04 | mem 13.73/16.12 GiB step 477 | train 4.445070 | val 5.900655 | tok/s 2626580.80 | norm 1.3389 | dt 199.61 ms | lr 9.5600e-04 | mlr 9.5600e-04 | mem 13.73/16.12 GiB step 478 | train 4.342992 | val 5.900655 | tok/s 2625658.47 | norm 1.4269 | dt 199.68 ms | lr 9.5800e-04 | mlr 9.5800e-04 | mem 13.73/16.12 GiB step 479 | train 4.357627 | val 5.900655 | tok/s 2626811.92 | norm 1.3604 | dt 199.59 ms | lr 9.6000e-04 | mlr 9.6000e-04 | mem 13.73/16.12 GiB step 480 | train 4.299726 | val 5.900655 | tok/s 2626352.59 | norm 1.4152 | dt 199.63 ms | lr 9.6200e-04 | mlr 9.6200e-04 | mem 13.73/16.12 GiB step 481 | train 4.378680 | val 5.900655 | tok/s 2624949.69 | norm 1.2191 | dt 199.73 ms | lr 9.6400e-04 | mlr 9.6400e-04 | mem 13.73/16.12 GiB step 482 | train 4.344148 | val 5.900655 | tok/s 2621789.38 | norm 1.4130 | dt 199.97 ms | lr 9.6600e-04 | mlr 9.6600e-04 | mem 13.73/16.12 GiB step 483 | train 4.430137 | val 5.900655 | tok/s 2627579.98 | norm 1.4284 | dt 199.53 ms | lr 9.6800e-04 | mlr 9.6800e-04 | mem 13.73/16.12 GiB step 484 | train 4.405962 | val 5.900655 | tok/s 2624146.01 | norm 1.2311 | dt 199.79 ms | lr 9.7000e-04 | mlr 9.7000e-04 | mem 13.73/16.12 GiB step 485 | train 4.191695 | val 5.900655 | tok/s 2620884.40 | norm 1.3308 | dt 200.04 ms | lr 9.7200e-04 | mlr 9.7200e-04 | mem 13.73/16.12 GiB step 486 | train 4.331272 | val 5.900655 | tok/s 2623561.88 | norm 1.2980 | dt 199.84 ms | lr 9.7400e-04 | mlr 9.7400e-04 | mem 13.73/16.12 GiB step 487 | train 4.287964 | val 5.900655 | tok/s 2618349.81 | norm 1.4202 | dt 200.24 ms | lr 9.7600e-04 | mlr 9.7600e-04 | mem 13.73/16.12 GiB step 488 | train 4.299291 | val 5.900655 | tok/s 2622478.46 | norm 1.2972 | dt 199.92 ms | lr 9.7800e-04 | mlr 9.7800e-04 | mem 13.73/16.12 GiB step 489 | train 4.315691 | val 5.900655 | tok/s 2622680.13 | norm 1.3900 | dt 199.91 ms | lr 9.8000e-04 | mlr 9.8000e-04 | mem 13.73/16.12 GiB step 490 | train 4.361405 | val 5.900655 | tok/s 2626027.84 | norm 1.1863 | dt 199.65 ms | lr 9.8200e-04 | mlr 9.8200e-04 | mem 13.73/16.12 GiB step 491 | train 4.291679 | val 5.900655 | tok/s 2627856.21 | norm 1.5314 | dt 199.51 ms | lr 9.8400e-04 | mlr 9.8400e-04 | mem 13.73/16.12 GiB step 492 | train 4.380963 | val 5.900655 | tok/s 2619606.97 | norm 1.1864 | dt 200.14 ms | lr 9.8600e-04 | mlr 9.8600e-04 | mem 13.73/16.12 GiB step 493 | train 4.330371 | val 5.900655 | tok/s 2627601.53 | norm 1.5782 | dt 199.53 ms | lr 9.8800e-04 | mlr 9.8800e-04 | mem 13.73/16.12 GiB step 494 | train 4.357635 | val 5.900655 | tok/s 2627275.93 | norm 1.2303 | dt 199.56 ms | lr 9.9000e-04 | mlr 9.9000e-04 | mem 13.73/16.12 GiB step 495 | train 4.350768 | val 5.900655 | tok/s 2623592.29 | norm 1.5091 | dt 199.84 ms | lr 9.9200e-04 | mlr 9.9200e-04 | mem 13.73/16.12 GiB step 496 | train 4.299978 | val 5.900655 | tok/s 2631834.91 | norm 1.2577 | dt 199.21 ms | lr 9.9400e-04 | mlr 9.9400e-04 | mem 13.73/16.12 GiB step 497 | train 4.285505 | val 5.900655 | tok/s 2619981.72 | norm 1.5614 | dt 200.11 ms | lr 9.9600e-04 | mlr 9.9600e-04 | mem 13.73/16.12 GiB step 498 | train 4.211067 | val 5.900655 | tok/s 2625383.10 | norm 1.2796 | dt 199.70 ms | lr 9.9800e-04 | mlr 9.9800e-04 | mem 13.73/16.12 GiB step 499 | train 4.335659 | val 5.900655 | tok/s 2626852.58 | norm 1.5958 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 500 | val_loss 4.668277 step 500 | train 4.316826 | val 4.668277 | tok/s 2544632.15 | norm 1.4678 | dt 206.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 501 | train 4.300688 | val 4.668277 | tok/s 2627135.25 | norm 1.6452 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 502 | train 4.261350 | val 4.668277 | tok/s 2624740.45 | norm 1.3364 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 503 | train 4.274572 | val 4.668277 | tok/s 2621252.40 | norm 1.4383 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 504 | train 4.266956 | val 4.668277 | tok/s 2622281.78 | norm 1.2417 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 505 | train 4.258319 | val 4.668277 | tok/s 2628780.19 | norm 1.4395 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 506 | train 4.448100 | val 4.668277 | tok/s 2624092.00 | norm 1.4031 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 507 | train 4.232776 | val 4.668277 | tok/s 2626702.16 | norm 1.2806 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 508 | train 4.333677 | val 4.668277 | tok/s 2621650.54 | norm 1.4428 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 509 | train 4.228677 | val 4.668277 | tok/s 2625364.18 | norm 1.4490 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 510 | train 4.256432 | val 4.668277 | tok/s 2631047.07 | norm 1.4643 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 511 | train 4.222443 | val 4.668277 | tok/s 2627746.00 | norm 1.3377 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 512 | train 4.170675 | val 4.668277 | tok/s 2630898.53 | norm 1.4728 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 513 | train 4.394073 | val 4.668277 | tok/s 2622260.22 | norm 1.3423 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 514 | train 4.189807 | val 4.668277 | tok/s 2626312.11 | norm 1.3253 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 515 | train 4.238643 | val 4.668277 | tok/s 2628391.39 | norm 1.1279 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 516 | train 4.224889 | val 4.668277 | tok/s 2621540.20 | norm 1.2181 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 517 | train 4.155628 | val 4.668277 | tok/s 2626619.55 | norm 1.1799 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 518 | train 4.306810 | val 4.668277 | tok/s 2625184.16 | norm 1.1359 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 519 | train 4.155418 | val 4.668277 | tok/s 2629250.00 | norm 1.1117 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 520 | train 4.160344 | val 4.668277 | tok/s 2634445.19 | norm 1.1997 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 521 | train 4.139393 | val 4.668277 | tok/s 2624978.47 | norm 1.0941 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 522 | train 4.185392 | val 4.668277 | tok/s 2632176.82 | norm 1.2576 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 523 | train 4.203897 | val 4.668277 | tok/s 2622827.36 | norm 1.2077 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 524 | train 4.158202 | val 4.668277 | tok/s 2624082.08 | norm 1.1110 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 525 | train 4.214368 | val 4.668277 | tok/s 2628401.56 | norm 1.2097 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 526 | train 4.187586 | val 4.668277 | tok/s 2624029.34 | norm 1.0956 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 527 | train 4.113723 | val 4.668277 | tok/s 2624392.69 | norm 1.3108 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 528 | train 4.099664 | val 4.668277 | tok/s 2620633.41 | norm 1.1619 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 529 | train 4.187577 | val 4.668277 | tok/s 2626547.45 | norm 1.1512 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 530 | train 4.094057 | val 4.668277 | tok/s 2628251.97 | norm 1.0103 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 531 | train 4.118253 | val 4.668277 | tok/s 2628373.14 | norm 1.0427 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 532 | train 4.040694 | val 4.668277 | tok/s 2630014.46 | norm 1.0584 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 533 | train 4.203362 | val 4.668277 | tok/s 2622404.68 | norm 0.9869 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 534 | train 4.102414 | val 4.668277 | tok/s 2621030.07 | norm 0.9370 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 535 | train 4.236063 | val 4.668277 | tok/s 2629755.72 | norm 1.0695 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 536 | train 4.017847 | val 4.668277 | tok/s 2621597.19 | norm 0.9470 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 537 | train 4.126219 | val 4.668277 | tok/s 2626834.45 | norm 1.0616 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 538 | train 4.084595 | val 4.668277 | tok/s 2623405.01 | norm 1.1282 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 539 | train 4.077645 | val 4.668277 | tok/s 2623303.67 | norm 1.0316 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 540 | train 4.088716 | val 4.668277 | tok/s 2622832.86 | norm 0.9701 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 541 | train 4.128730 | val 4.668277 | tok/s 2622639.29 | norm 1.0050 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 542 | train 4.209863 | val 4.668277 | tok/s 2626409.26 | norm 1.1712 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 543 | train 4.139517 | val 4.668277 | tok/s 2621817.38 | norm 0.9561 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 544 | train 4.119902 | val 4.668277 | tok/s 2623377.50 | norm 0.9939 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 545 | train 4.029914 | val 4.668277 | tok/s 2624830.45 | norm 1.0995 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 546 | train 3.998728 | val 4.668277 | tok/s 2628134.16 | norm 1.2984 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 547 | train 4.107809 | val 4.668277 | tok/s 2627730.29 | norm 1.1905 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 548 | train 4.126860 | val 4.668277 | tok/s 2626227.27 | norm 1.2108 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 549 | train 4.099660 | val 4.668277 | tok/s 2628653.16 | norm 1.1274 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 550 | train 4.025122 | val 4.668277 | tok/s 2631468.55 | norm 1.2118 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 551 | train 4.131793 | val 4.668277 | tok/s 2624901.87 | norm 1.1276 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 552 | train 4.121433 | val 4.668277 | tok/s 2629521.19 | norm 1.1542 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 553 | train 4.081304 | val 4.668277 | tok/s 2628996.97 | norm 1.0154 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 554 | train 4.059155 | val 4.668277 | tok/s 2625286.36 | norm 1.0428 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 555 | train 4.078403 | val 4.668277 | tok/s 2626204.97 | norm 0.9856 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 556 | train 4.058035 | val 4.668277 | tok/s 2625064.44 | norm 1.0942 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 557 | train 4.091643 | val 4.668277 | tok/s 2627045.35 | norm 0.9860 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 558 | train 4.020205 | val 4.668277 | tok/s 2623007.18 | norm 0.9968 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 559 | train 4.109781 | val 4.668277 | tok/s 2625878.01 | norm 1.0761 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 560 | train 4.129431 | val 4.668277 | tok/s 2625991.63 | norm 1.0405 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 561 | train 4.089879 | val 4.668277 | tok/s 2625354.71 | norm 1.1156 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 562 | train 4.076056 | val 4.668277 | tok/s 2630506.62 | norm 0.9905 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 563 | train 4.083436 | val 4.668277 | tok/s 2629444.79 | norm 0.9964 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 564 | train 4.003306 | val 4.668277 | tok/s 2627025.96 | norm 0.9650 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 565 | train 4.031120 | val 4.668277 | tok/s 2626738.59 | norm 1.0590 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 566 | train 3.987295 | val 4.668277 | tok/s 2570830.22 | norm 1.0276 | dt 203.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 567 | train 4.105005 | val 4.668277 | tok/s 2626840.73 | norm 1.0022 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 568 | train 3.985036 | val 4.668277 | tok/s 2628169.23 | norm 1.0987 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 569 | train 4.076259 | val 4.668277 | tok/s 2628714.99 | norm 1.0525 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 570 | train 4.024702 | val 4.668277 | tok/s 2630709.87 | norm 1.0099 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 571 | train 3.932115 | val 4.668277 | tok/s 2625931.95 | norm 1.0836 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 572 | train 3.966835 | val 4.668277 | tok/s 2628708.30 | norm 0.9283 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 573 | train 3.987101 | val 4.668277 | tok/s 2625863.71 | norm 0.8969 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 574 | train 3.976417 | val 4.668277 | tok/s 2626980.28 | norm 0.8792 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 575 | train 3.980408 | val 4.668277 | tok/s 2626143.03 | norm 0.8157 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 576 | train 4.039081 | val 4.668277 | tok/s 2627188.47 | norm 0.8272 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 577 | train 4.067899 | val 4.668277 | tok/s 2631026.92 | norm 0.9301 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 578 | train 3.949006 | val 4.668277 | tok/s 2628018.09 | norm 0.9716 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 579 | train 3.999165 | val 4.668277 | tok/s 2626216.67 | norm 1.0138 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 580 | train 4.014622 | val 4.668277 | tok/s 2625433.84 | norm 1.0054 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 581 | train 4.070064 | val 4.668277 | tok/s 2621777.69 | norm 0.9815 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 582 | train 4.059611 | val 4.668277 | tok/s 2628223.13 | norm 0.7965 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 583 | train 4.033815 | val 4.668277 | tok/s 2625374.36 | norm 0.7975 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 584 | train 3.903728 | val 4.668277 | tok/s 2625927.85 | norm 0.9109 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 585 | train 3.896596 | val 4.668277 | tok/s 2627641.44 | norm 1.0428 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 586 | train 4.071024 | val 4.668277 | tok/s 2624658.47 | norm 0.9438 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 587 | train 3.953628 | val 4.668277 | tok/s 2628687.82 | norm 0.7841 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 588 | train 3.918523 | val 4.668277 | tok/s 2627883.04 | norm 0.8747 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 589 | train 3.939748 | val 4.668277 | tok/s 2629073.75 | norm 1.0109 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 590 | train 4.031833 | val 4.668277 | tok/s 2620413.69 | norm 0.9895 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 591 | train 3.961544 | val 4.668277 | tok/s 2480438.36 | norm 0.9085 | dt 211.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 592 | train 3.880326 | val 4.668277 | tok/s 2615698.31 | norm 0.9008 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 593 | train 3.973832 | val 4.668277 | tok/s 2621122.65 | norm 1.0003 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 594 | train 4.003608 | val 4.668277 | tok/s 2629993.24 | norm 0.8790 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 595 | train 4.021238 | val 4.668277 | tok/s 2624775.54 | norm 0.8805 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 596 | train 3.874112 | val 4.668277 | tok/s 2625057.50 | norm 0.8498 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 597 | train 3.955700 | val 4.668277 | tok/s 2628285.87 | norm 0.8469 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 598 | train 4.033669 | val 4.668277 | tok/s 2623163.55 | norm 0.8988 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 599 | train 4.077478 | val 4.668277 | tok/s 2628949.88 | norm 0.8609 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 600 | train 3.990290 | val 4.668277 | tok/s 2623110.58 | norm 0.7939 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 601 | train 4.102081 | val 4.668277 | tok/s 2621171.13 | norm 0.8993 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 602 | train 4.014272 | val 4.668277 | tok/s 2628827.95 | norm 0.9792 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 603 | train 3.892226 | val 4.668277 | tok/s 2625398.23 | norm 0.9423 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 604 | train 3.922956 | val 4.668277 | tok/s 2607289.56 | norm 0.9271 | dt 201.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 605 | train 3.949077 | val 4.668277 | tok/s 2627868.36 | norm 0.9229 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 606 | train 3.884643 | val 4.668277 | tok/s 2626543.81 | norm 0.8353 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 607 | train 4.055140 | val 4.668277 | tok/s 2626655.80 | norm 0.7751 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 608 | train 3.950542 | val 4.668277 | tok/s 2627506.04 | norm 0.8297 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 609 | train 3.928501 | val 4.668277 | tok/s 2629342.61 | norm 0.8576 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 610 | train 3.888494 | val 4.668277 | tok/s 2625689.10 | norm 0.8354 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 611 | train 3.888389 | val 4.668277 | tok/s 2629978.20 | norm 0.7644 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 612 | train 3.881504 | val 4.668277 | tok/s 2632783.65 | norm 0.7953 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 613 | train 3.880549 | val 4.668277 | tok/s 2545517.09 | norm 0.8308 | dt 205.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 614 | train 3.864633 | val 4.668277 | tok/s 2622024.66 | norm 0.7883 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 615 | train 3.919463 | val 4.668277 | tok/s 2617428.97 | norm 0.7672 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 616 | train 3.853697 | val 4.668277 | tok/s 2617673.44 | norm 0.8519 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 617 | train 3.981305 | val 4.668277 | tok/s 2558696.58 | norm 1.0407 | dt 204.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 618 | train 3.977260 | val 4.668277 | tok/s 2559402.63 | norm 1.0311 | dt 204.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 619 | train 3.849630 | val 4.668277 | tok/s 2626671.28 | norm 0.9112 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 620 | train 3.782592 | val 4.668277 | tok/s 2622390.54 | norm 0.9562 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 621 | train 3.852819 | val 4.668277 | tok/s 2623812.21 | norm 0.8514 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 622 | train 3.867261 | val 4.668277 | tok/s 2622580.22 | norm 0.7215 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 623 | train 3.920959 | val 4.668277 | tok/s 2624218.29 | norm 0.7940 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 624 | train 3.762450 | val 4.668277 | tok/s 2626842.40 | norm 0.9557 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 625 | train 4.126504 | val 4.668277 | tok/s 2629035.92 | norm 0.9879 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 626 | train 3.902529 | val 4.668277 | tok/s 2622827.73 | norm 0.7142 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 627 | train 3.974154 | val 4.668277 | tok/s 2625461.93 | norm 0.7061 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 628 | train 4.287071 | val 4.668277 | tok/s 2620571.08 | norm 0.9957 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 629 | train 3.927895 | val 4.668277 | tok/s 2627296.90 | norm 0.6775 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 630 | train 3.805781 | val 4.668277 | tok/s 2624055.11 | norm 0.6930 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 631 | train 3.868862 | val 4.668277 | tok/s 2627215.87 | norm 0.7689 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 632 | train 3.769193 | val 4.668277 | tok/s 2628462.53 | norm 0.7414 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 633 | train 3.876476 | val 4.668277 | tok/s 2625765.61 | norm 0.7628 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 634 | train 3.905708 | val 4.668277 | tok/s 2632049.48 | norm 0.7868 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 635 | train 3.832778 | val 4.668277 | tok/s 2625717.53 | norm 0.8684 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 636 | train 3.812237 | val 4.668277 | tok/s 2623870.75 | norm 0.7942 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 637 | train 3.809842 | val 4.668277 | tok/s 2627448.48 | norm 0.6895 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 638 | train 3.809617 | val 4.668277 | tok/s 2624235.01 | norm 0.6866 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 639 | train 3.838969 | val 4.668277 | tok/s 2625176.24 | norm 0.7563 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 640 | train 3.866394 | val 4.668277 | tok/s 2621627.93 | norm 0.7394 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 641 | train 3.868803 | val 4.668277 | tok/s 2626916.61 | norm 0.8314 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 642 | train 3.854857 | val 4.668277 | tok/s 2625629.98 | norm 0.7800 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 643 | train 3.811792 | val 4.668277 | tok/s 2622355.46 | norm 0.7834 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 644 | train 3.846368 | val 4.668277 | tok/s 2618876.62 | norm 0.8660 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 645 | train 3.886734 | val 4.668277 | tok/s 2623829.84 | norm 0.8395 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 646 | train 3.857013 | val 4.668277 | tok/s 2622714.70 | norm 0.7958 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 647 | train 3.782981 | val 4.668277 | tok/s 2625840.49 | norm 0.7531 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 648 | train 3.922943 | val 4.668277 | tok/s 2620413.91 | norm 0.7223 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 649 | train 3.834312 | val 4.668277 | tok/s 2623049.28 | norm 0.7669 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 650 | train 3.814870 | val 4.668277 | tok/s 2619232.12 | norm 0.8860 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 651 | train 3.827457 | val 4.668277 | tok/s 2625497.13 | norm 0.8797 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 652 | train 3.924923 | val 4.668277 | tok/s 2626398.90 | norm 0.8094 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 653 | train 3.847766 | val 4.668277 | tok/s 2624036.16 | norm 0.8347 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 654 | train 3.921650 | val 4.668277 | tok/s 2626692.60 | norm 0.8192 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 655 | train 3.873645 | val 4.668277 | tok/s 2625259.11 | norm 0.8215 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 656 | train 3.899724 | val 4.668277 | tok/s 2625113.11 | norm 0.8635 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 657 | train 3.762188 | val 4.668277 | tok/s 2627628.76 | norm 0.8272 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 658 | train 3.838789 | val 4.668277 | tok/s 2620167.28 | norm 0.8121 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 659 | train 3.817315 | val 4.668277 | tok/s 2623891.80 | norm 0.7855 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 660 | train 3.824064 | val 4.668277 | tok/s 2625180.03 | norm 0.7595 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 661 | train 3.837954 | val 4.668277 | tok/s 2625504.94 | norm 0.7418 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 662 | train 3.857191 | val 4.668277 | tok/s 2619542.73 | norm 0.7800 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 663 | train 3.818662 | val 4.668277 | tok/s 2625988.28 | norm 0.7980 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 664 | train 3.880907 | val 4.668277 | tok/s 2626780.44 | norm 0.7357 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 665 | train 3.938243 | val 4.668277 | tok/s 2624843.10 | norm 0.7001 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 666 | train 3.899666 | val 4.668277 | tok/s 2624496.96 | norm 0.6677 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 667 | train 3.756188 | val 4.668277 | tok/s 2628399.66 | norm 0.7736 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 668 | train 3.802886 | val 4.668277 | tok/s 2629311.45 | norm 0.8500 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 669 | train 3.755080 | val 4.668277 | tok/s 2628674.19 | norm 0.8219 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 670 | train 3.817308 | val 4.668277 | tok/s 2621246.85 | norm 0.7336 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 671 | train 3.853007 | val 4.668277 | tok/s 2625968.29 | norm 0.7058 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 672 | train 3.757895 | val 4.668277 | tok/s 2625954.79 | norm 0.7355 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 673 | train 3.861626 | val 4.668277 | tok/s 2624463.66 | norm 0.7505 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 674 | train 3.746724 | val 4.668277 | tok/s 2625821.38 | norm 0.8714 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 675 | train 3.923374 | val 4.668277 | tok/s 2625441.84 | norm 0.7972 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 676 | train 3.867514 | val 4.668277 | tok/s 2629839.72 | norm 0.7019 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 677 | train 3.916810 | val 4.668277 | tok/s 2623758.25 | norm 0.6805 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 678 | train 3.730628 | val 4.668277 | tok/s 2620141.47 | norm 0.7073 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 679 | train 3.734154 | val 4.668277 | tok/s 2632076.63 | norm 0.7817 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 680 | train 3.788497 | val 4.668277 | tok/s 2629277.94 | norm 0.8043 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 681 | train 3.670501 | val 4.668277 | tok/s 2631666.18 | norm 0.7841 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 682 | train 3.802560 | val 4.668277 | tok/s 2626852.30 | norm 0.8132 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 683 | train 3.802541 | val 4.668277 | tok/s 2624635.93 | norm 0.8095 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 684 | train 3.782416 | val 4.668277 | tok/s 2630140.82 | norm 0.7622 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 685 | train 3.781924 | val 4.668277 | tok/s 2630417.14 | norm 0.7834 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 686 | train 3.782633 | val 4.668277 | tok/s 2632020.86 | norm 0.8161 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 687 | train 3.844516 | val 4.668277 | tok/s 2620699.46 | norm 0.7989 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 688 | train 3.751497 | val 4.668277 | tok/s 2619854.90 | norm 0.7208 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 689 | train 3.804320 | val 4.668277 | tok/s 2624175.90 | norm 0.6512 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 690 | train 3.784653 | val 4.668277 | tok/s 2624113.87 | norm 0.7319 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 691 | train 3.809075 | val 4.668277 | tok/s 2624059.80 | norm 0.7534 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 692 | train 3.757480 | val 4.668277 | tok/s 2619820.39 | norm 0.8013 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 693 | train 3.759750 | val 4.668277 | tok/s 2620413.58 | norm 0.7853 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 694 | train 3.867718 | val 4.668277 | tok/s 2620476.74 | norm 0.7231 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 695 | train 3.781687 | val 4.668277 | tok/s 2622926.31 | norm 0.7266 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 696 | train 3.787612 | val 4.668277 | tok/s 2626601.69 | norm 0.7642 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 697 | train 3.780165 | val 4.668277 | tok/s 2619494.36 | norm 0.6888 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 698 | train 3.665978 | val 4.668277 | tok/s 2622914.21 | norm 0.6790 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 699 | train 3.713553 | val 4.668277 | tok/s 2628108.52 | norm 0.7831 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 700 | train 3.772729 | val 4.668277 | tok/s 2627619.37 | norm 0.7757 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 701 | train 3.774741 | val 4.668277 | tok/s 2628124.19 | norm 0.7764 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 702 | train 3.768709 | val 4.668277 | tok/s 2618501.16 | norm 0.7885 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 703 | train 3.720154 | val 4.668277 | tok/s 2620306.48 | norm 0.8165 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 704 | train 3.830328 | val 4.668277 | tok/s 2626781.69 | norm 0.8418 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 705 | train 4.028046 | val 4.668277 | tok/s 2623750.24 | norm 1.9782 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 706 | train 3.723052 | val 4.668277 | tok/s 2626417.26 | norm 0.4614 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 707 | train 3.852570 | val 4.668277 | tok/s 2622761.86 | norm 0.5723 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 708 | train 3.675781 | val 4.668277 | tok/s 2627303.12 | norm 0.7330 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 709 | train 3.728961 | val 4.668277 | tok/s 2628819.46 | norm 0.7931 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 710 | train 3.744457 | val 4.668277 | tok/s 2625542.87 | norm 0.7573 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 711 | train 3.740811 | val 4.668277 | tok/s 2627796.43 | norm 0.7576 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 712 | train 3.745324 | val 4.668277 | tok/s 2620658.02 | norm 0.7408 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 713 | train 3.688409 | val 4.668277 | tok/s 2625137.82 | norm 0.7109 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 714 | train 3.753575 | val 4.668277 | tok/s 2632328.87 | norm 0.7108 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 715 | train 3.713949 | val 4.668277 | tok/s 2629258.46 | norm 0.6703 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 716 | train 3.759972 | val 4.668277 | tok/s 2626795.66 | norm 0.6616 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 717 | train 3.795742 | val 4.668277 | tok/s 2483959.41 | norm 0.7288 | dt 211.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 718 | train 3.692467 | val 4.668277 | tok/s 2608916.81 | norm 0.7562 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 719 | train 3.773548 | val 4.668277 | tok/s 2626888.21 | norm 0.7652 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 720 | train 3.696773 | val 4.668277 | tok/s 2626628.69 | norm 0.6991 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 721 | train 3.774031 | val 4.668277 | tok/s 2624901.22 | norm 0.7401 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 722 | train 3.793956 | val 4.668277 | tok/s 2624675.98 | norm 0.8202 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 723 | train 3.715510 | val 4.668277 | tok/s 2624198.77 | norm 0.7921 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 724 | train 3.682918 | val 4.668277 | tok/s 2625450.10 | norm 0.7535 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 725 | train 3.846183 | val 4.668277 | tok/s 2622919.50 | norm 0.7581 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 726 | train 3.704187 | val 4.668277 | tok/s 2627906.45 | norm 0.6814 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 727 | train 3.761331 | val 4.668277 | tok/s 2622220.41 | norm 0.6612 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 728 | train 3.799064 | val 4.668277 | tok/s 2625894.33 | norm 0.7732 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 729 | train 3.707645 | val 4.668277 | tok/s 2632124.94 | norm 0.7760 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 730 | train 3.691861 | val 4.668277 | tok/s 2630807.57 | norm 0.7208 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 731 | train 3.685873 | val 4.668277 | tok/s 2626898.41 | norm 0.7404 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 732 | train 3.741687 | val 4.668277 | tok/s 2623038.78 | norm 0.7548 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 733 | train 3.777710 | val 4.668277 | tok/s 2622345.08 | norm 0.7058 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 734 | train 3.707347 | val 4.668277 | tok/s 2629593.79 | norm 0.7906 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 735 | train 3.740838 | val 4.668277 | tok/s 2620882.92 | norm 0.8830 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 736 | train 3.695928 | val 4.668277 | tok/s 2625764.04 | norm 0.7938 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 737 | train 3.829300 | val 4.668277 | tok/s 2616053.79 | norm 0.7180 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 738 | train 3.918253 | val 4.668277 | tok/s 2627471.32 | norm 0.8269 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 739 | train 3.762169 | val 4.668277 | tok/s 2627178.24 | norm 0.6970 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 740 | train 3.726179 | val 4.668277 | tok/s 2627136.21 | norm 0.7147 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 741 | train 3.756289 | val 4.668277 | tok/s 2629604.50 | norm 0.8146 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 742 | train 3.669532 | val 4.668277 | tok/s 2625317.96 | norm 0.7410 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 743 | train 3.751471 | val 4.668277 | tok/s 2627059.53 | norm 0.6767 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 744 | train 3.746378 | val 4.668277 | tok/s 2632772.39 | norm 0.6550 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 745 | train 3.732464 | val 4.668277 | tok/s 2629909.86 | norm 0.6066 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 746 | train 3.759618 | val 4.668277 | tok/s 2626514.53 | norm 0.6763 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 747 | train 3.686565 | val 4.668277 | tok/s 2621190.12 | norm 0.6823 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 748 | train 3.808200 | val 4.668277 | tok/s 2621043.54 | norm 0.6805 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 749 | train 3.725515 | val 4.668277 | tok/s 2631044.48 | norm 0.7015 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 750 | val_loss 4.074482 step 750 | train 3.766136 | val 4.074482 | tok/s 2610904.73 | norm 0.7418 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 751 | train 3.703914 | val 4.074482 | tok/s 2626703.40 | norm 0.7201 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 752 | train 3.677206 | val 4.074482 | tok/s 2632789.85 | norm 0.6986 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 753 | train 3.643164 | val 4.074482 | tok/s 2629640.11 | norm 0.7042 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 754 | train 3.776541 | val 4.074482 | tok/s 2628094.56 | norm 0.6904 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 755 | train 3.749865 | val 4.074482 | tok/s 2627431.31 | norm 0.6336 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 756 | train 3.718285 | val 4.074482 | tok/s 2630386.21 | norm 0.6249 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 757 | train 3.619310 | val 4.074482 | tok/s 2631703.80 | norm 0.6787 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 758 | train 3.691758 | val 4.074482 | tok/s 2625318.08 | norm 0.8046 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 759 | train 3.902747 | val 4.074482 | tok/s 2624683.45 | norm 0.6435 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 760 | train 3.713020 | val 4.074482 | tok/s 2634247.31 | norm 0.6583 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 761 | train 3.750463 | val 4.074482 | tok/s 2623667.40 | norm 0.7420 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 762 | train 3.622888 | val 4.074482 | tok/s 2626321.98 | norm 0.7280 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 763 | train 3.724560 | val 4.074482 | tok/s 2622544.87 | norm 0.6656 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 764 | train 3.657043 | val 4.074482 | tok/s 2627436.78 | norm 0.6949 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 765 | train 3.781595 | val 4.074482 | tok/s 2631403.18 | norm 0.7349 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 766 | train 3.597582 | val 4.074482 | tok/s 2559067.54 | norm 0.6412 | dt 204.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 767 | train 3.647170 | val 4.074482 | tok/s 2616441.84 | norm 0.6826 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 768 | train 3.646318 | val 4.074482 | tok/s 2626092.81 | norm 0.7647 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 769 | train 3.835978 | val 4.074482 | tok/s 2623418.23 | norm 0.7723 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 770 | train 3.714850 | val 4.074482 | tok/s 2628892.94 | norm 0.6733 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 771 | train 3.636566 | val 4.074482 | tok/s 2628404.52 | norm 0.6132 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 772 | train 3.620339 | val 4.074482 | tok/s 2609674.21 | norm 0.6805 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 773 | train 3.712042 | val 4.074482 | tok/s 2626428.52 | norm 0.6824 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 774 | train 3.562425 | val 4.074482 | tok/s 2626739.37 | norm 0.6119 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 775 | train 3.701103 | val 4.074482 | tok/s 2625844.46 | norm 0.6147 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 776 | train 3.659492 | val 4.074482 | tok/s 2625819.67 | norm 0.5900 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 777 | train 3.554666 | val 4.074482 | tok/s 2604910.00 | norm 0.5834 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 778 | train 3.587970 | val 4.074482 | tok/s 2618366.25 | norm 0.6118 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 779 | train 3.685699 | val 4.074482 | tok/s 2606881.96 | norm 0.6337 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 780 | train 3.576395 | val 4.074482 | tok/s 2628164.94 | norm 0.7162 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 781 | train 3.774222 | val 4.074482 | tok/s 2625593.52 | norm 0.6842 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 782 | train 3.598204 | val 4.074482 | tok/s 2627391.55 | norm 0.7040 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 783 | train 3.654147 | val 4.074482 | tok/s 2619958.44 | norm 0.7294 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 784 | train 3.594877 | val 4.074482 | tok/s 2625850.96 | norm 0.5814 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 785 | train 3.754295 | val 4.074482 | tok/s 2625713.82 | norm 0.6400 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 786 | train 3.604938 | val 4.074482 | tok/s 2628451.86 | norm 0.6158 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 787 | train 3.640368 | val 4.074482 | tok/s 2622887.75 | norm 0.6261 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 788 | train 3.654783 | val 4.074482 | tok/s 2479334.07 | norm 0.6560 | dt 211.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 789 | train 3.662271 | val 4.074482 | tok/s 2628286.03 | norm 0.6509 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 790 | train 3.645328 | val 4.074482 | tok/s 2627161.76 | norm 0.6850 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 791 | train 3.682353 | val 4.074482 | tok/s 2621739.22 | norm 0.7028 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 792 | train 3.622504 | val 4.074482 | tok/s 2626474.57 | norm 0.7136 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 793 | train 3.641234 | val 4.074482 | tok/s 2628365.10 | norm 0.6693 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 794 | train 3.705154 | val 4.074482 | tok/s 2620140.99 | norm 0.6856 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 795 | train 3.611465 | val 4.074482 | tok/s 2624578.55 | norm 0.7419 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 796 | train 3.550003 | val 4.074482 | tok/s 2622016.66 | norm 0.7349 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 797 | train 3.629670 | val 4.074482 | tok/s 2576409.25 | norm 0.6633 | dt 203.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 798 | train 3.700650 | val 4.074482 | tok/s 2625437.78 | norm 0.5986 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 799 | train 3.626510 | val 4.074482 | tok/s 2625672.94 | norm 0.5888 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 800 | train 3.470042 | val 4.074482 | tok/s 2629640.43 | norm 0.6580 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 801 | train 3.676755 | val 4.074482 | tok/s 2627258.94 | norm 0.6860 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 802 | train 3.562716 | val 4.074482 | tok/s 2630874.08 | norm 0.6440 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 803 | train 3.643810 | val 4.074482 | tok/s 2625885.12 | norm 0.6241 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 804 | train 3.689595 | val 4.074482 | tok/s 2604146.81 | norm 0.6804 | dt 201.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 805 | train 3.649437 | val 4.074482 | tok/s 2617386.27 | norm 0.7247 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 806 | train 3.680723 | val 4.074482 | tok/s 2616874.86 | norm 0.6767 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 807 | train 3.608125 | val 4.074482 | tok/s 2630951.67 | norm 0.6447 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 808 | train 3.660034 | val 4.074482 | tok/s 2625861.40 | norm 0.6972 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 809 | train 3.886482 | val 4.074482 | tok/s 2624935.26 | norm 0.7392 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 810 | train 3.676351 | val 4.074482 | tok/s 2627039.55 | norm 0.6711 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 811 | train 3.716072 | val 4.074482 | tok/s 2623617.83 | norm 0.6701 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 812 | train 3.586182 | val 4.074482 | tok/s 2629928.03 | norm 0.6923 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 813 | train 3.802174 | val 4.074482 | tok/s 2623146.15 | norm 0.6787 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 814 | train 3.754052 | val 4.074482 | tok/s 2622222.01 | norm 0.6552 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 815 | train 3.643541 | val 4.074482 | tok/s 2619258.39 | norm 0.6930 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 816 | train 3.603579 | val 4.074482 | tok/s 2618656.02 | norm 0.6312 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 817 | train 3.665419 | val 4.074482 | tok/s 2623164.73 | norm 0.6022 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 818 | train 3.758702 | val 4.074482 | tok/s 2622119.97 | norm 0.6108 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 819 | train 3.693976 | val 4.074482 | tok/s 2623363.11 | norm 0.7254 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 820 | train 3.546338 | val 4.074482 | tok/s 2621649.96 | norm 0.7801 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 821 | train 3.679321 | val 4.074482 | tok/s 2594221.02 | norm 0.8060 | dt 202.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 822 | train 3.515792 | val 4.074482 | tok/s 2610167.61 | norm 0.7681 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 823 | train 3.645955 | val 4.074482 | tok/s 2625471.56 | norm 0.6914 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 824 | train 3.569147 | val 4.074482 | tok/s 2625704.68 | norm 0.6786 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 825 | train 3.546344 | val 4.074482 | tok/s 2619723.62 | norm 0.6711 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 826 | train 3.677567 | val 4.074482 | tok/s 2620954.26 | norm 0.6564 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 827 | train 3.614516 | val 4.074482 | tok/s 2624926.91 | norm 0.6426 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 828 | train 3.564019 | val 4.074482 | tok/s 2627976.91 | norm 0.7243 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 829 | train 3.688366 | val 4.074482 | tok/s 2630307.37 | norm 0.7489 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 830 | train 3.702591 | val 4.074482 | tok/s 2621191.05 | norm 0.6553 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 831 | train 3.660262 | val 4.074482 | tok/s 2622385.36 | norm 0.6048 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 832 | train 3.612564 | val 4.074482 | tok/s 2632674.44 | norm 0.6445 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 833 | train 3.575192 | val 4.074482 | tok/s 2624988.38 | norm 0.6993 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 834 | train 3.674481 | val 4.074482 | tok/s 2626029.80 | norm 0.6377 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 835 | train 3.609830 | val 4.074482 | tok/s 2620728.63 | norm 0.6619 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 836 | train 3.540087 | val 4.074482 | tok/s 2621750.22 | norm 0.7029 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 837 | train 3.549162 | val 4.074482 | tok/s 2630334.08 | norm 0.7133 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 838 | train 3.648551 | val 4.074482 | tok/s 2626459.65 | norm 0.6709 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 839 | train 3.701772 | val 4.074482 | tok/s 2628524.20 | norm 0.6937 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 840 | train 3.635744 | val 4.074482 | tok/s 2617733.94 | norm 0.6184 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 841 | train 3.615165 | val 4.074482 | tok/s 2626643.87 | norm 0.6165 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 842 | train 3.557515 | val 4.074482 | tok/s 2627916.25 | norm 0.6541 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 843 | train 3.588828 | val 4.074482 | tok/s 2627294.61 | norm 0.6269 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 844 | train 3.677911 | val 4.074482 | tok/s 2628884.92 | norm 0.6345 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 845 | train 3.601096 | val 4.074482 | tok/s 2623945.71 | norm 0.6815 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 846 | train 3.564557 | val 4.074482 | tok/s 2624704.53 | norm 0.6652 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 847 | train 3.484396 | val 4.074482 | tok/s 2627891.10 | norm 0.6358 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 848 | train 3.426534 | val 4.074482 | tok/s 2623764.87 | norm 0.6921 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 849 | train 3.597989 | val 4.074482 | tok/s 2629486.70 | norm 0.6817 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 850 | train 3.543520 | val 4.074482 | tok/s 2623565.80 | norm 0.6206 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 851 | train 3.626408 | val 4.074482 | tok/s 2622157.62 | norm 0.5129 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 852 | train 3.507702 | val 4.074482 | tok/s 2626604.67 | norm 0.5589 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 853 | train 3.704168 | val 4.074482 | tok/s 2627537.75 | norm 0.7532 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 854 | train 3.634641 | val 4.074482 | tok/s 2630894.96 | norm 0.5822 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 855 | train 3.653118 | val 4.074482 | tok/s 2614136.49 | norm 0.5866 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 856 | train 3.625977 | val 4.074482 | tok/s 2625727.22 | norm 0.6446 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 857 | train 3.643582 | val 4.074482 | tok/s 2630089.55 | norm 0.6731 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 858 | train 3.514526 | val 4.074482 | tok/s 2624859.10 | norm 0.6984 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 859 | train 3.571274 | val 4.074482 | tok/s 2627740.47 | norm 0.6719 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 860 | train 3.545528 | val 4.074482 | tok/s 2620199.99 | norm 0.6547 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 861 | train 3.635951 | val 4.074482 | tok/s 2623652.20 | norm 0.6701 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 862 | train 3.561945 | val 4.074482 | tok/s 2625727.76 | norm 0.6973 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 863 | train 3.566665 | val 4.074482 | tok/s 2622048.00 | norm 0.7024 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 864 | train 3.596319 | val 4.074482 | tok/s 2625129.57 | norm 0.6866 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 865 | train 3.750247 | val 4.074482 | tok/s 2626732.95 | norm 0.7730 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 866 | train 3.637876 | val 4.074482 | tok/s 2623894.14 | norm 0.7044 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 867 | train 3.592721 | val 4.074482 | tok/s 2490341.50 | norm 0.6513 | dt 210.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 868 | train 3.743566 | val 4.074482 | tok/s 2615788.08 | norm 0.7296 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 869 | train 3.590660 | val 4.074482 | tok/s 2627934.73 | norm 0.6779 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 870 | train 3.423596 | val 4.074482 | tok/s 2630184.32 | norm 0.6759 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 871 | train 3.554705 | val 4.074482 | tok/s 2628862.77 | norm 0.6547 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 872 | train 3.593631 | val 4.074482 | tok/s 2626214.67 | norm 0.6103 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 873 | train 3.539942 | val 4.074482 | tok/s 2613869.41 | norm 0.5808 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 874 | train 3.566913 | val 4.074482 | tok/s 2628112.30 | norm 0.5958 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 875 | train 3.576294 | val 4.074482 | tok/s 2621273.65 | norm 0.6138 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 876 | train 3.649354 | val 4.074482 | tok/s 2622854.13 | norm 0.5886 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 877 | train 3.565117 | val 4.074482 | tok/s 2629477.14 | norm 0.6054 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 878 | train 3.516394 | val 4.074482 | tok/s 2612930.80 | norm 0.7063 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 879 | train 3.646063 | val 4.074482 | tok/s 2628135.16 | norm 0.6485 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 880 | train 3.524774 | val 4.074482 | tok/s 2622112.91 | norm 0.5722 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 881 | train 3.606982 | val 4.074482 | tok/s 2628267.46 | norm 0.6806 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 882 | train 3.492498 | val 4.074482 | tok/s 2627312.58 | norm 0.6785 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 883 | train 3.507074 | val 4.074482 | tok/s 2625105.97 | norm 0.6725 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 884 | train 3.552071 | val 4.074482 | tok/s 2624208.68 | norm 0.6533 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 885 | train 3.634229 | val 4.074482 | tok/s 2619946.53 | norm 0.6252 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 886 | train 3.588961 | val 4.074482 | tok/s 2623621.49 | norm 0.6242 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 887 | train 3.670666 | val 4.074482 | tok/s 2625901.34 | norm 0.6254 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 888 | train 3.635458 | val 4.074482 | tok/s 2626477.56 | norm 0.6441 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 889 | train 3.531858 | val 4.074482 | tok/s 2631704.00 | norm 0.6730 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 890 | train 3.628102 | val 4.074482 | tok/s 2621067.40 | norm 0.7108 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 891 | train 3.518112 | val 4.074482 | tok/s 2626631.48 | norm 0.7007 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 892 | train 3.395990 | val 4.074482 | tok/s 2626278.87 | norm 0.7026 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 893 | train 3.490472 | val 4.074482 | tok/s 2627107.09 | norm 0.6622 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 894 | train 3.639368 | val 4.074482 | tok/s 2620276.14 | norm 0.6317 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 895 | train 3.463339 | val 4.074482 | tok/s 2417633.09 | norm 0.6819 | dt 216.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 896 | train 3.559119 | val 4.074482 | tok/s 2619292.59 | norm 0.6339 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 897 | train 3.706877 | val 4.074482 | tok/s 2608501.32 | norm 0.5645 | dt 200.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 898 | train 3.516055 | val 4.074482 | tok/s 2625004.28 | norm 0.5815 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 899 | train 3.607533 | val 4.074482 | tok/s 2627692.13 | norm 0.6332 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 900 | train 3.513888 | val 4.074482 | tok/s 2624140.91 | norm 0.6433 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 901 | train 3.639174 | val 4.074482 | tok/s 2627505.87 | norm 0.5777 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 902 | train 3.547716 | val 4.074482 | tok/s 2507692.38 | norm 0.6139 | dt 209.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 903 | train 3.492152 | val 4.074482 | tok/s 2582072.51 | norm 0.6198 | dt 203.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 904 | train 3.577939 | val 4.074482 | tok/s 2628023.05 | norm 0.6058 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 905 | train 3.573887 | val 4.074482 | tok/s 2626386.35 | norm 0.5542 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 906 | train 3.532206 | val 4.074482 | tok/s 2626380.35 | norm 0.6107 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 907 | train 3.510843 | val 4.074482 | tok/s 2623368.82 | norm 0.7139 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 908 | train 3.688719 | val 4.074482 | tok/s 2619102.45 | norm 0.7597 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 909 | train 3.627825 | val 4.074482 | tok/s 2625041.23 | norm 0.6491 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 910 | train 3.553327 | val 4.074482 | tok/s 2622089.18 | norm 0.6640 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 911 | train 3.689809 | val 4.074482 | tok/s 2624033.24 | norm 0.7174 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 912 | train 3.479917 | val 4.074482 | tok/s 2627594.64 | norm 0.6238 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 913 | train 3.656314 | val 4.074482 | tok/s 2628389.14 | norm 0.6312 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 914 | train 3.506105 | val 4.074482 | tok/s 2629876.43 | norm 0.6727 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 915 | train 3.681330 | val 4.074482 | tok/s 2625665.23 | norm 0.6456 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 916 | train 3.562542 | val 4.074482 | tok/s 2629648.51 | norm 0.6208 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 917 | train 3.497435 | val 4.074482 | tok/s 2625037.30 | norm 0.6579 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 918 | train 3.619285 | val 4.074482 | tok/s 2626865.36 | norm 0.6636 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 919 | train 3.589073 | val 4.074482 | tok/s 2627789.88 | norm 0.6802 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 920 | train 3.496777 | val 4.074482 | tok/s 2627306.53 | norm 0.6528 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 921 | train 3.562116 | val 4.074482 | tok/s 2626818.65 | norm 0.6942 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 922 | train 3.504057 | val 4.074482 | tok/s 2613391.37 | norm 0.6613 | dt 200.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 923 | train 3.544642 | val 4.074482 | tok/s 2631445.54 | norm 0.6133 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 924 | train 3.546339 | val 4.074482 | tok/s 2632355.09 | norm 0.6406 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 925 | train 3.574512 | val 4.074482 | tok/s 2624636.01 | norm 0.6032 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 926 | train 3.603702 | val 4.074482 | tok/s 2630336.52 | norm 0.6340 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 927 | train 3.527861 | val 4.074482 | tok/s 2620852.80 | norm 0.6292 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 928 | train 3.538318 | val 4.074482 | tok/s 2624518.32 | norm 0.6538 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 929 | train 3.553675 | val 4.074482 | tok/s 2630525.11 | norm 0.6781 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 930 | train 3.467846 | val 4.074482 | tok/s 2624825.20 | norm 0.6231 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 931 | train 3.502321 | val 4.074482 | tok/s 2631564.70 | norm 0.6740 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 932 | train 3.454759 | val 4.074482 | tok/s 2615904.64 | norm 0.5910 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 933 | train 3.433489 | val 4.074482 | tok/s 2626564.31 | norm 0.5423 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 934 | train 3.520880 | val 4.074482 | tok/s 2628956.32 | norm 0.6048 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 935 | train 3.586425 | val 4.074482 | tok/s 2627663.04 | norm 0.6694 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 936 | train 3.558432 | val 4.074482 | tok/s 2627005.47 | norm 0.6422 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 937 | train 3.488485 | val 4.074482 | tok/s 2615144.01 | norm 0.6536 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 938 | train 3.538702 | val 4.074482 | tok/s 2622486.46 | norm 0.6820 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 939 | train 3.460770 | val 4.074482 | tok/s 2628768.98 | norm 0.6570 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 940 | train 3.524443 | val 4.074482 | tok/s 2625378.00 | norm 0.6550 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 941 | train 3.446010 | val 4.074482 | tok/s 2624429.74 | norm 0.6047 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 942 | train 3.529171 | val 4.074482 | tok/s 2625217.98 | norm 0.6313 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 943 | train 3.545877 | val 4.074482 | tok/s 2627209.68 | norm 0.7405 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 944 | train 3.580227 | val 4.074482 | tok/s 2629603.13 | norm 0.6412 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 945 | train 3.559858 | val 4.074482 | tok/s 2627961.47 | norm 0.7455 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 946 | train 3.575393 | val 4.074482 | tok/s 2624360.17 | norm 0.7481 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 947 | train 3.583807 | val 4.074482 | tok/s 2620346.65 | norm 0.6200 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 948 | train 3.505953 | val 4.074482 | tok/s 2611555.43 | norm 0.6121 | dt 200.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 949 | train 3.564546 | val 4.074482 | tok/s 2630396.34 | norm 0.6150 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 950 | train 3.574524 | val 4.074482 | tok/s 2624846.42 | norm 0.5792 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 951 | train 3.563112 | val 4.074482 | tok/s 2628887.93 | norm 0.5648 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 952 | train 3.534284 | val 4.074482 | tok/s 2601207.24 | norm 0.6156 | dt 201.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 953 | train 3.455349 | val 4.074482 | tok/s 2624029.13 | norm 0.6863 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 954 | train 3.622349 | val 4.074482 | tok/s 2627335.36 | norm 0.6833 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 955 | train 3.548556 | val 4.074482 | tok/s 2627684.28 | norm 0.6211 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 956 | train 3.440020 | val 4.074482 | tok/s 2624714.32 | norm 0.6075 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 957 | train 3.606035 | val 4.074482 | tok/s 2611275.06 | norm 0.5317 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 958 | train 3.492346 | val 4.074482 | tok/s 2621331.56 | norm 0.5609 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 959 | train 3.548168 | val 4.074482 | tok/s 2628621.01 | norm 0.5963 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 960 | train 3.505885 | val 4.074482 | tok/s 2627514.72 | norm 0.6146 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 961 | train 3.592726 | val 4.074482 | tok/s 2626785.22 | norm 0.5678 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 962 | train 3.544016 | val 4.074482 | tok/s 2622568.82 | norm 0.5572 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 963 | train 3.403620 | val 4.074482 | tok/s 2623931.05 | norm 0.6626 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 964 | train 3.408619 | val 4.074482 | tok/s 2630661.35 | norm 0.6857 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 965 | train 3.500378 | val 4.074482 | tok/s 2623659.08 | norm 0.6759 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 966 | train 3.539987 | val 4.074482 | tok/s 2627143.17 | norm 0.5468 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 967 | train 3.509277 | val 4.074482 | tok/s 2620820.96 | norm 0.5836 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 968 | train 3.537544 | val 4.074482 | tok/s 2606244.33 | norm 0.6133 | dt 201.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 969 | train 3.518055 | val 4.074482 | tok/s 2631317.35 | norm 0.6021 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 970 | train 3.480733 | val 4.074482 | tok/s 2628083.41 | norm 0.5445 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 971 | train 3.444403 | val 4.074482 | tok/s 2626205.05 | norm 0.5454 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 972 | train 3.476296 | val 4.074482 | tok/s 2628816.50 | norm 0.6010 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 973 | train 3.471402 | val 4.074482 | tok/s 2626840.94 | norm 0.5729 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 974 | train 3.491663 | val 4.074482 | tok/s 2626805.91 | norm 0.6299 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 975 | train 3.404969 | val 4.074482 | tok/s 2628402.28 | norm 0.5960 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 976 | train 3.563771 | val 4.074482 | tok/s 2627369.92 | norm 0.5622 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 977 | train 3.541383 | val 4.074482 | tok/s 2623068.76 | norm 0.6163 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 978 | train 3.479420 | val 4.074482 | tok/s 2626291.26 | norm 0.6109 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 979 | train 3.427200 | val 4.074482 | tok/s 2629823.98 | norm 0.6264 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 980 | train 3.495589 | val 4.074482 | tok/s 2628128.35 | norm 0.6716 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 981 | train 3.501885 | val 4.074482 | tok/s 2626228.40 | norm 0.6585 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 982 | train 3.425220 | val 4.074482 | tok/s 2626512.28 | norm 0.6170 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 983 | train 3.367781 | val 4.074482 | tok/s 2629213.41 | norm 0.5773 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 984 | train 3.524565 | val 4.074482 | tok/s 2633196.78 | norm 0.6234 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 985 | train 3.606841 | val 4.074482 | tok/s 2625157.47 | norm 0.6222 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 986 | train 3.481511 | val 4.074482 | tok/s 2625509.15 | norm 0.6114 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 987 | train 3.515833 | val 4.074482 | tok/s 2623771.42 | norm 0.5779 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 988 | train 3.509953 | val 4.074482 | tok/s 2623602.54 | norm 0.5900 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 989 | train 3.487846 | val 4.074482 | tok/s 2627532.01 | norm 0.5756 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 990 | train 3.496485 | val 4.074482 | tok/s 2624430.84 | norm 0.6785 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 991 | train 3.570571 | val 4.074482 | tok/s 2627112.32 | norm 0.6662 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 992 | train 3.506269 | val 4.074482 | tok/s 2624952.92 | norm 0.6802 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 993 | train 3.441044 | val 4.074482 | tok/s 2449831.35 | norm 0.6996 | dt 214.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 994 | train 3.510621 | val 4.074482 | tok/s 2403195.93 | norm 0.6059 | dt 218.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 995 | train 3.548063 | val 4.074482 | tok/s 2620870.96 | norm 0.6173 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 996 | train 3.483313 | val 4.074482 | tok/s 2625718.04 | norm 0.6768 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 997 | train 3.720533 | val 4.074482 | tok/s 2614133.38 | norm 0.6768 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 998 | train 3.460853 | val 4.074482 | tok/s 2627933.81 | norm 0.6150 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 999 | train 3.439345 | val 4.074482 | tok/s 2622752.29 | norm 0.5937 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 1000 | val_loss 3.870685 step 1000 | train 3.389673 | val 3.870685 | tok/s 2610466.68 | norm 0.5915 | dt 200.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1001 | train 3.609277 | val 3.870685 | tok/s 2625482.97 | norm 0.6473 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1002 | train 3.648104 | val 3.870685 | tok/s 2626261.48 | norm 0.6467 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1003 | train 3.412315 | val 3.870685 | tok/s 2625135.09 | norm 0.7143 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1004 | train 3.537769 | val 3.870685 | tok/s 2490523.29 | norm 0.6890 | dt 210.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1005 | train 3.461332 | val 3.870685 | tok/s 2624226.75 | norm 0.6353 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1006 | train 3.504386 | val 3.870685 | tok/s 2624955.93 | norm 0.6624 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1007 | train 3.617195 | val 3.870685 | tok/s 2628004.59 | norm 0.5833 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1008 | train 3.561660 | val 3.870685 | tok/s 2624434.72 | norm 0.5367 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1009 | train 3.340429 | val 3.870685 | tok/s 2626500.93 | norm 0.6568 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1010 | train 3.460144 | val 3.870685 | tok/s 2622782.76 | norm 0.6440 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1011 | train 3.554680 | val 3.870685 | tok/s 2621689.69 | norm 0.5827 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1012 | train 3.499402 | val 3.870685 | tok/s 2629028.53 | norm 0.5414 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1013 | train 3.674577 | val 3.870685 | tok/s 2621740.27 | norm 0.5253 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1014 | train 3.437227 | val 3.870685 | tok/s 2624577.20 | norm 0.5016 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1015 | train 3.564874 | val 3.870685 | tok/s 2626877.22 | norm 0.6828 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1016 | train 3.542128 | val 3.870685 | tok/s 2619253.83 | norm 0.5825 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1017 | train 3.454875 | val 3.870685 | tok/s 2632038.77 | norm 0.5089 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1018 | train 3.416531 | val 3.870685 | tok/s 2630164.03 | norm 0.5740 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1019 | train 3.418116 | val 3.870685 | tok/s 2633094.74 | norm 0.5714 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1020 | train 3.446357 | val 3.870685 | tok/s 2597233.01 | norm 0.6222 | dt 201.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1021 | train 3.593620 | val 3.870685 | tok/s 2601196.81 | norm 0.7248 | dt 201.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1022 | train 3.540863 | val 3.870685 | tok/s 2628407.05 | norm 0.6236 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1023 | train 3.489328 | val 3.870685 | tok/s 2626132.97 | norm 0.5729 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1024 | train 3.498195 | val 3.870685 | tok/s 2630028.60 | norm 0.6343 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1025 | train 3.552389 | val 3.870685 | tok/s 2618333.37 | norm 0.6219 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1026 | train 3.480545 | val 3.870685 | tok/s 2618469.58 | norm 0.6021 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1027 | train 3.562796 | val 3.870685 | tok/s 2625754.77 | norm 0.6078 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1028 | train 3.814649 | val 3.870685 | tok/s 2622653.84 | norm 0.6841 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1029 | train 3.549479 | val 3.870685 | tok/s 2629793.93 | norm 0.6232 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1030 | train 3.478454 | val 3.870685 | tok/s 2619849.96 | norm 0.5351 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1031 | train 3.476972 | val 3.870685 | tok/s 2622739.15 | norm 0.5370 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1032 | train 3.479324 | val 3.870685 | tok/s 2626652.42 | norm 0.5889 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1033 | train 3.463008 | val 3.870685 | tok/s 2626843.02 | norm 0.6822 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1034 | train 3.446759 | val 3.870685 | tok/s 2628986.70 | norm 0.6184 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1035 | train 3.508769 | val 3.870685 | tok/s 2612622.15 | norm 0.6415 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1036 | train 3.508299 | val 3.870685 | tok/s 2616625.16 | norm 0.6967 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1037 | train 3.549200 | val 3.870685 | tok/s 2623436.11 | norm 0.5814 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1038 | train 3.574319 | val 3.870685 | tok/s 2625992.93 | norm 0.5676 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1039 | train 3.388131 | val 3.870685 | tok/s 2629960.39 | norm 0.6234 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1040 | train 3.556899 | val 3.870685 | tok/s 2623736.74 | norm 0.6629 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1041 | train 3.372978 | val 3.870685 | tok/s 2626970.52 | norm 0.5825 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1042 | train 3.440350 | val 3.870685 | tok/s 2482978.43 | norm 0.5647 | dt 211.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1043 | train 3.551368 | val 3.870685 | tok/s 2614184.45 | norm 0.6222 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1044 | train 3.546173 | val 3.870685 | tok/s 2624936.15 | norm 0.6120 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1045 | train 3.518224 | val 3.870685 | tok/s 2621626.20 | norm 0.6211 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1046 | train 3.401336 | val 3.870685 | tok/s 2623829.38 | norm 0.6064 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1047 | train 3.490934 | val 3.870685 | tok/s 2629224.37 | norm 0.6250 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1048 | train 3.387688 | val 3.870685 | tok/s 2627411.98 | norm 0.6422 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1049 | train 3.449147 | val 3.870685 | tok/s 2628288.93 | norm 0.6961 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1050 | train 3.492275 | val 3.870685 | tok/s 2621134.90 | norm 0.6607 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1051 | train 3.438686 | val 3.870685 | tok/s 2622819.29 | norm 0.6003 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1052 | train 3.533046 | val 3.870685 | tok/s 2630991.51 | norm 0.6283 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1053 | train 3.596446 | val 3.870685 | tok/s 2628068.12 | norm 0.6127 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1054 | train 3.492380 | val 3.870685 | tok/s 2626883.53 | norm 0.6269 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1055 | train 3.485270 | val 3.870685 | tok/s 2623622.68 | norm 0.6066 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1056 | train 3.521378 | val 3.870685 | tok/s 2624135.05 | norm 0.5957 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1057 | train 3.496435 | val 3.870685 | tok/s 2624928.25 | norm 0.6567 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1058 | train 3.491122 | val 3.870685 | tok/s 2624560.73 | norm 0.6274 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1059 | train 3.544453 | val 3.870685 | tok/s 2629178.01 | norm 0.6642 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1060 | train 3.460207 | val 3.870685 | tok/s 2625120.96 | norm 0.6919 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1061 | train 3.400415 | val 3.870685 | tok/s 2625689.24 | norm 0.6630 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1062 | train 3.654870 | val 3.870685 | tok/s 2630660.17 | norm 0.7400 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1063 | train 3.543853 | val 3.870685 | tok/s 2623421.89 | norm 0.7065 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1064 | train 3.529198 | val 3.870685 | tok/s 2627613.58 | norm 0.6054 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1065 | train 3.488980 | val 3.870685 | tok/s 2626178.58 | norm 0.6223 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1066 | train 3.402706 | val 3.870685 | tok/s 2627259.70 | norm 0.6235 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1067 | train 3.499988 | val 3.870685 | tok/s 2631286.31 | norm 0.6205 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1068 | train 3.460642 | val 3.870685 | tok/s 2623967.27 | norm 0.6354 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1069 | train 3.458523 | val 3.870685 | tok/s 2630530.40 | norm 0.6200 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1070 | train 3.497108 | val 3.870685 | tok/s 2620552.71 | norm 0.5717 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1071 | train 3.598595 | val 3.870685 | tok/s 2622502.90 | norm 0.5959 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1072 | train 3.495518 | val 3.870685 | tok/s 2632916.92 | norm 0.6590 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1073 | train 3.433498 | val 3.870685 | tok/s 2629253.23 | norm 0.6414 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1074 | train 3.454595 | val 3.870685 | tok/s 2627824.58 | norm 0.6519 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1075 | train 3.446945 | val 3.870685 | tok/s 2623625.18 | norm 0.6449 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1076 | train 3.448822 | val 3.870685 | tok/s 2626138.15 | norm 0.6028 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1077 | train 3.430471 | val 3.870685 | tok/s 2626320.15 | norm 0.6703 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1078 | train 3.479289 | val 3.870685 | tok/s 2628883.93 | norm 0.6020 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1079 | train 3.539729 | val 3.870685 | tok/s 2629857.77 | norm 0.5667 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1080 | train 3.444886 | val 3.870685 | tok/s 2626242.20 | norm 0.6373 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1081 | train 3.610861 | val 3.870685 | tok/s 2626412.48 | norm 0.6428 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1082 | train 3.379628 | val 3.870685 | tok/s 2628487.58 | norm 0.6098 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1083 | train 3.446274 | val 3.870685 | tok/s 2624283.04 | norm 0.5777 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1084 | train 3.400024 | val 3.870685 | tok/s 2628082.15 | norm 0.6174 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1085 | train 3.554640 | val 3.870685 | tok/s 2613459.21 | norm 0.6300 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1086 | train 3.323396 | val 3.870685 | tok/s 2622819.28 | norm 0.6781 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1087 | train 3.566473 | val 3.870685 | tok/s 2630481.54 | norm 0.6827 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1088 | train 3.586385 | val 3.870685 | tok/s 2627733.71 | norm 0.6417 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1089 | train 3.296965 | val 3.870685 | tok/s 2630653.03 | norm 0.5868 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1090 | train 3.590943 | val 3.870685 | tok/s 2626453.95 | norm 0.5321 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1091 | train 3.545790 | val 3.870685 | tok/s 2626422.34 | norm 0.6135 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1092 | train 3.438735 | val 3.870685 | tok/s 2631542.94 | norm 0.6587 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1093 | train 3.479119 | val 3.870685 | tok/s 2623033.83 | norm 0.7310 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1094 | train 3.327485 | val 3.870685 | tok/s 2625695.75 | norm 0.6915 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1095 | train 3.411783 | val 3.870685 | tok/s 2625993.33 | norm 0.6900 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1096 | train 3.508961 | val 3.870685 | tok/s 2627667.27 | norm 0.6250 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1097 | train 3.443126 | val 3.870685 | tok/s 2629769.08 | norm 0.5523 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1098 | train 3.409515 | val 3.870685 | tok/s 2626038.26 | norm 0.5496 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1099 | train 3.441302 | val 3.870685 | tok/s 2621328.69 | norm 0.5401 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1100 | train 3.480762 | val 3.870685 | tok/s 2618325.14 | norm 0.5547 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1101 | train 3.439858 | val 3.870685 | tok/s 2626272.97 | norm 0.5673 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1102 | train 3.431652 | val 3.870685 | tok/s 2628068.91 | norm 0.5569 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1103 | train 3.467036 | val 3.870685 | tok/s 2628367.68 | norm 0.4952 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1104 | train 3.401400 | val 3.870685 | tok/s 2605712.53 | norm 0.6167 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1105 | train 3.475143 | val 3.870685 | tok/s 2622881.85 | norm 0.6643 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1106 | train 3.352602 | val 3.870685 | tok/s 2623173.87 | norm 0.6957 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1107 | train 3.356756 | val 3.870685 | tok/s 2632624.69 | norm 0.5788 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1108 | train 3.473350 | val 3.870685 | tok/s 2629552.58 | norm 0.5654 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1109 | train 3.417134 | val 3.870685 | tok/s 2631866.02 | norm 0.5587 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1110 | train 3.361214 | val 3.870685 | tok/s 2625807.77 | norm 0.5581 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1111 | train 3.362256 | val 3.870685 | tok/s 2622651.19 | norm 0.5393 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1112 | train 3.409544 | val 3.870685 | tok/s 2629485.04 | norm 0.5678 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1113 | train 3.454697 | val 3.870685 | tok/s 2628075.58 | norm 0.5816 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1114 | train 3.513757 | val 3.870685 | tok/s 2623917.10 | norm 1.1256 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1115 | train 3.525466 | val 3.870685 | tok/s 2624240.00 | norm 0.6135 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1116 | train 3.528140 | val 3.870685 | tok/s 2625545.34 | norm 0.7898 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1117 | train 3.455424 | val 3.870685 | tok/s 2630462.61 | norm 0.6919 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1118 | train 3.490755 | val 3.870685 | tok/s 2622659.09 | norm 0.6423 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1119 | train 3.524646 | val 3.870685 | tok/s 2628935.85 | norm 0.6733 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1120 | train 3.410101 | val 3.870685 | tok/s 2623635.13 | norm 0.5892 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1121 | train 3.402771 | val 3.870685 | tok/s 2608615.59 | norm 0.6233 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1122 | train 3.747418 | val 3.870685 | tok/s 2631327.96 | norm 0.7118 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1123 | train 3.443274 | val 3.870685 | tok/s 2623570.55 | norm 0.5500 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1124 | train 3.359668 | val 3.870685 | tok/s 2629105.49 | norm 0.5311 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1125 | train 3.386295 | val 3.870685 | tok/s 2624005.03 | norm 0.5613 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1126 | train 3.414097 | val 3.870685 | tok/s 2627071.26 | norm 0.5594 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1127 | train 3.477014 | val 3.870685 | tok/s 2627331.17 | norm 0.5752 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1128 | train 3.444040 | val 3.870685 | tok/s 2628203.33 | norm 0.5913 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1129 | train 3.541844 | val 3.870685 | tok/s 2632576.03 | norm 0.6362 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1130 | train 3.412940 | val 3.870685 | tok/s 2627158.13 | norm 0.6191 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1131 | train 3.403104 | val 3.870685 | tok/s 2626820.89 | norm 0.5773 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1132 | train 3.428858 | val 3.870685 | tok/s 2627524.45 | norm 0.5837 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1133 | train 3.358716 | val 3.870685 | tok/s 2626188.02 | norm 0.5953 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1134 | train 3.376993 | val 3.870685 | tok/s 2628287.75 | norm 0.6413 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1135 | train 3.654180 | val 3.870685 | tok/s 2624226.99 | norm 0.6228 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1136 | train 3.439405 | val 3.870685 | tok/s 2593966.00 | norm 0.5083 | dt 202.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1137 | train 3.418950 | val 3.870685 | tok/s 2626083.41 | norm 0.6004 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1138 | train 3.421454 | val 3.870685 | tok/s 2513970.90 | norm 0.6471 | dt 208.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1139 | train 3.380275 | val 3.870685 | tok/s 2482057.03 | norm 0.6357 | dt 211.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1140 | train 3.436966 | val 3.870685 | tok/s 2626947.65 | norm 0.6867 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1141 | train 3.366491 | val 3.870685 | tok/s 2625702.39 | norm 0.6152 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1142 | train 3.352711 | val 3.870685 | tok/s 2619263.21 | norm 0.6445 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1143 | train 3.441885 | val 3.870685 | tok/s 2490819.02 | norm 0.6755 | dt 210.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1144 | train 3.441188 | val 3.870685 | tok/s 2628590.19 | norm 0.7083 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1145 | train 3.467328 | val 3.870685 | tok/s 2628704.58 | norm 0.6664 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1146 | train 3.463429 | val 3.870685 | tok/s 2627398.27 | norm 0.6724 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1147 | train 3.488237 | val 3.870685 | tok/s 2627037.02 | norm 0.6120 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1148 | train 3.418971 | val 3.870685 | tok/s 2626817.04 | norm 0.5537 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1149 | train 3.393727 | val 3.870685 | tok/s 2631068.65 | norm 0.5921 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1150 | train 3.382384 | val 3.870685 | tok/s 2629410.37 | norm 0.5900 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1151 | train 3.370505 | val 3.870685 | tok/s 2631167.88 | norm 0.6079 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1152 | train 3.488939 | val 3.870685 | tok/s 2626303.99 | norm 0.5842 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1153 | train 3.484698 | val 3.870685 | tok/s 2625245.04 | norm 0.5424 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1154 | train 3.380459 | val 3.870685 | tok/s 2631293.13 | norm 0.6459 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1155 | train 3.392387 | val 3.870685 | tok/s 2628387.84 | norm 0.6921 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1156 | train 3.497232 | val 3.870685 | tok/s 2631224.50 | norm 0.6453 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1157 | train 3.434686 | val 3.870685 | tok/s 2622111.37 | norm 0.6534 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1158 | train 3.413221 | val 3.870685 | tok/s 2624024.53 | norm 0.8077 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1159 | train 3.408965 | val 3.870685 | tok/s 2627188.43 | norm 0.5984 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1160 | train 3.419522 | val 3.870685 | tok/s 2619562.44 | norm 0.6167 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1161 | train 3.532456 | val 3.870685 | tok/s 2633050.89 | norm 0.6639 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1162 | train 3.488533 | val 3.870685 | tok/s 2624452.70 | norm 0.6162 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1163 | train 3.662909 | val 3.870685 | tok/s 2623695.56 | norm 0.6384 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1164 | train 3.460418 | val 3.870685 | tok/s 2630918.91 | norm 0.5961 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1165 | train 3.428856 | val 3.870685 | tok/s 2626699.55 | norm 0.5638 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1166 | train 3.303771 | val 3.870685 | tok/s 2622375.87 | norm 0.5570 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1167 | train 3.506882 | val 3.870685 | tok/s 2618307.04 | norm 0.6498 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1168 | train 3.485091 | val 3.870685 | tok/s 2547957.96 | norm 0.6543 | dt 205.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1169 | train 3.311281 | val 3.870685 | tok/s 2615482.04 | norm 0.6788 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1170 | train 3.466444 | val 3.870685 | tok/s 2629917.82 | norm 0.6309 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1171 | train 3.484782 | val 3.870685 | tok/s 2629015.85 | norm 0.5720 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1172 | train 3.367883 | val 3.870685 | tok/s 2605422.95 | norm 0.5956 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1173 | train 3.409466 | val 3.870685 | tok/s 2608804.46 | norm 0.6151 | dt 200.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1174 | train 3.485294 | val 3.870685 | tok/s 2630136.93 | norm 0.6043 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1175 | train 3.339512 | val 3.870685 | tok/s 2628988.75 | norm 0.5823 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1176 | train 3.433883 | val 3.870685 | tok/s 2626141.08 | norm 0.5854 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1177 | train 3.368370 | val 3.870685 | tok/s 2623356.74 | norm 0.5917 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1178 | train 3.465603 | val 3.870685 | tok/s 2628425.71 | norm 0.6143 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1179 | train 3.309094 | val 3.870685 | tok/s 2632784.22 | norm 0.6168 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1180 | train 3.449255 | val 3.870685 | tok/s 2626973.30 | norm 0.6342 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1181 | train 3.403376 | val 3.870685 | tok/s 2630653.43 | norm 0.6515 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1182 | train 3.477128 | val 3.870685 | tok/s 2620049.85 | norm 0.7133 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1183 | train 3.392256 | val 3.870685 | tok/s 2624259.22 | norm 0.6607 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1184 | train 3.501768 | val 3.870685 | tok/s 2627612.60 | norm 0.6202 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1185 | train 3.565171 | val 3.870685 | tok/s 2622673.65 | norm 0.5724 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1186 | train 3.405022 | val 3.870685 | tok/s 2630072.03 | norm 0.5527 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1187 | train 3.413854 | val 3.870685 | tok/s 2622972.24 | norm 0.5221 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1188 | train 3.383511 | val 3.870685 | tok/s 2625004.48 | norm 0.5349 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1189 | train 3.469924 | val 3.870685 | tok/s 2626053.75 | norm 0.6662 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1190 | train 3.327689 | val 3.870685 | tok/s 2624548.63 | norm 0.6722 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1191 | train 3.492104 | val 3.870685 | tok/s 2627206.96 | norm 0.7399 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1192 | train 3.472705 | val 3.870685 | tok/s 2626098.69 | norm 0.6336 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1193 | train 3.504941 | val 3.870685 | tok/s 2627363.56 | norm 0.5213 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1194 | train 3.271201 | val 3.870685 | tok/s 2630454.53 | norm 0.6255 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1195 | train 3.274288 | val 3.870685 | tok/s 2624848.12 | norm 0.6802 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1196 | train 3.298740 | val 3.870685 | tok/s 2623818.92 | norm 0.6255 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1197 | train 3.388129 | val 3.870685 | tok/s 2622746.64 | norm 0.6090 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1198 | train 3.435655 | val 3.870685 | tok/s 2622518.50 | norm 0.6177 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1199 | train 3.498928 | val 3.870685 | tok/s 2628045.30 | norm 0.5790 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1200 | train 3.374859 | val 3.870685 | tok/s 2626853.27 | norm 0.6545 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1201 | train 3.371513 | val 3.870685 | tok/s 2625830.01 | norm 0.5972 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1202 | train 3.478488 | val 3.870685 | tok/s 2622972.29 | norm 0.5381 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1203 | train 3.418434 | val 3.870685 | tok/s 2627005.15 | norm 0.5681 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1204 | train 3.554848 | val 3.870685 | tok/s 2626273.39 | norm 0.6204 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1205 | train 3.329326 | val 3.870685 | tok/s 2623337.41 | norm 0.6511 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1206 | train 3.390281 | val 3.870685 | tok/s 2627340.34 | norm 0.6399 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1207 | train 3.395631 | val 3.870685 | tok/s 2624085.97 | norm 0.5786 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1208 | train 3.379253 | val 3.870685 | tok/s 2622722.71 | norm 0.5910 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1209 | train 3.408833 | val 3.870685 | tok/s 2629486.72 | norm 0.5953 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1210 | train 3.499866 | val 3.870685 | tok/s 2624764.50 | norm 0.5082 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1211 | train 3.321486 | val 3.870685 | tok/s 2629967.01 | norm 0.5222 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1212 | train 3.437219 | val 3.870685 | tok/s 2625753.08 | norm 0.6123 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1213 | train 3.463917 | val 3.870685 | tok/s 2620059.66 | norm 0.6410 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1214 | train 3.425872 | val 3.870685 | tok/s 2625708.68 | norm 0.6356 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1215 | train 3.337157 | val 3.870685 | tok/s 2625516.93 | norm 0.6599 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1216 | train 3.311141 | val 3.870685 | tok/s 2627533.01 | norm 0.5883 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1217 | train 3.472373 | val 3.870685 | tok/s 2623270.49 | norm 0.5324 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1218 | train 3.430558 | val 3.870685 | tok/s 2627467.92 | norm 0.5922 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1219 | train 3.444755 | val 3.870685 | tok/s 2625808.81 | norm 0.5996 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1220 | train 3.448884 | val 3.870685 | tok/s 2623347.91 | norm 0.6088 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1221 | train 3.442683 | val 3.870685 | tok/s 2626040.48 | norm 0.6250 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1222 | train 3.419314 | val 3.870685 | tok/s 2624161.61 | norm 0.6277 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1223 | train 3.392621 | val 3.870685 | tok/s 2628690.32 | norm 0.5709 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1224 | train 3.332681 | val 3.870685 | tok/s 2627922.44 | norm 0.5938 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1225 | train 3.478416 | val 3.870685 | tok/s 2628238.52 | norm 0.5977 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1226 | train 3.321202 | val 3.870685 | tok/s 2624638.80 | norm 0.5916 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1227 | train 3.451236 | val 3.870685 | tok/s 2622196.57 | norm 0.5978 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1228 | train 3.471695 | val 3.870685 | tok/s 2623058.54 | norm 0.5408 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1229 | train 3.429816 | val 3.870685 | tok/s 2625124.84 | norm 0.5628 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1230 | train 3.378404 | val 3.870685 | tok/s 2625291.09 | norm 0.5845 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1231 | train 3.403901 | val 3.870685 | tok/s 2627958.24 | norm 0.5544 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1232 | train 3.388306 | val 3.870685 | tok/s 2620393.90 | norm 0.5626 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1233 | train 3.475174 | val 3.870685 | tok/s 2624095.55 | norm 0.5708 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1234 | train 3.499947 | val 3.870685 | tok/s 2630399.77 | norm 0.5190 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1235 | train 3.458884 | val 3.870685 | tok/s 2626975.26 | norm 0.5762 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1236 | train 3.499746 | val 3.870685 | tok/s 2627587.66 | norm 0.6331 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1237 | train 3.451859 | val 3.870685 | tok/s 2623016.88 | norm 0.6198 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1238 | train 3.430968 | val 3.870685 | tok/s 2622163.28 | norm 0.5777 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1239 | train 3.360754 | val 3.870685 | tok/s 2629363.52 | norm 0.6095 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1240 | train 3.486353 | val 3.870685 | tok/s 2625266.98 | norm 0.7201 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1241 | train 3.417418 | val 3.870685 | tok/s 2630386.38 | norm 0.7489 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1242 | train 3.546230 | val 3.870685 | tok/s 2629349.10 | norm 0.8509 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1243 | train 3.409833 | val 3.870685 | tok/s 2405267.60 | norm 0.6738 | dt 217.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1244 | train 3.524415 | val 3.870685 | tok/s 2615460.64 | norm 0.5466 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1245 | train 3.334909 | val 3.870685 | tok/s 2619419.21 | norm 0.6364 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1246 | train 3.323983 | val 3.870685 | tok/s 2630207.45 | norm 0.6375 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1247 | train 3.330649 | val 3.870685 | tok/s 2624636.15 | norm 0.6276 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1248 | train 3.319884 | val 3.870685 | tok/s 2633226.40 | norm 0.5657 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1249 | train 3.474391 | val 3.870685 | tok/s 2630428.21 | norm 0.5356 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 1250 | val_loss 3.769978 step 1250 | train 3.529299 | val 3.769978 | tok/s 2562097.52 | norm 0.5597 | dt 204.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1251 | train 3.434431 | val 3.769978 | tok/s 2623605.22 | norm 0.5512 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1252 | train 3.475252 | val 3.769978 | tok/s 2627725.98 | norm 0.6761 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1253 | train 3.357172 | val 3.769978 | tok/s 2624129.21 | norm 0.5320 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1254 | train 3.391140 | val 3.769978 | tok/s 2627405.38 | norm 0.5972 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1255 | train 3.380583 | val 3.769978 | tok/s 2633028.96 | norm 0.6371 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1256 | train 3.337764 | val 3.769978 | tok/s 2629111.22 | norm 0.6419 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1257 | train 3.209898 | val 3.769978 | tok/s 2631030.62 | norm 0.7075 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1258 | train 3.488981 | val 3.769978 | tok/s 2627877.51 | norm 0.6048 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1259 | train 3.450264 | val 3.769978 | tok/s 2486794.38 | norm 0.6106 | dt 210.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1260 | train 3.529947 | val 3.769978 | tok/s 2509547.26 | norm 0.6908 | dt 208.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1261 | train 3.351157 | val 3.769978 | tok/s 2620496.11 | norm 0.5991 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1262 | train 3.410028 | val 3.769978 | tok/s 2624436.08 | norm 0.5520 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1263 | train 3.466485 | val 3.769978 | tok/s 2625874.60 | norm 0.5793 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1264 | train 3.443648 | val 3.769978 | tok/s 2627052.76 | norm 0.6203 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1265 | train 3.349795 | val 3.769978 | tok/s 2617543.08 | norm 0.6383 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1266 | train 3.374279 | val 3.769978 | tok/s 2622588.88 | norm 0.5754 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1267 | train 3.424521 | val 3.769978 | tok/s 2626395.54 | norm 0.5484 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1268 | train 3.308572 | val 3.769978 | tok/s 2624794.63 | norm 0.6040 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1269 | train 3.476630 | val 3.769978 | tok/s 2627186.15 | norm 0.6369 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1270 | train 3.281389 | val 3.769978 | tok/s 2619745.52 | norm 0.6056 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1271 | train 3.383750 | val 3.769978 | tok/s 2621998.39 | norm 0.6152 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1272 | train 3.476763 | val 3.769978 | tok/s 2627758.95 | norm 0.6060 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1273 | train 3.365907 | val 3.769978 | tok/s 2624939.06 | norm 0.5612 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1274 | train 3.409656 | val 3.769978 | tok/s 2594871.05 | norm 0.5782 | dt 202.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1275 | train 3.476615 | val 3.769978 | tok/s 2599906.70 | norm 0.6361 | dt 201.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1276 | train 3.539194 | val 3.769978 | tok/s 2629461.42 | norm 0.6238 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1277 | train 3.487165 | val 3.769978 | tok/s 2626551.69 | norm 0.6315 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1278 | train 3.431888 | val 3.769978 | tok/s 2627616.63 | norm 0.6337 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1279 | train 3.321153 | val 3.769978 | tok/s 2630396.00 | norm 0.6149 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1280 | train 3.479332 | val 3.769978 | tok/s 2621822.88 | norm 0.6170 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1281 | train 3.384511 | val 3.769978 | tok/s 2624332.71 | norm 0.5919 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1282 | train 3.330480 | val 3.769978 | tok/s 2631674.54 | norm 0.6621 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1283 | train 3.317028 | val 3.769978 | tok/s 2618090.74 | norm 0.7273 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1284 | train 3.395584 | val 3.769978 | tok/s 2630189.36 | norm 0.6080 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1285 | train 3.345239 | val 3.769978 | tok/s 2614774.81 | norm 0.5992 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1286 | train 3.443703 | val 3.769978 | tok/s 2624301.62 | norm 0.5550 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1287 | train 3.413791 | val 3.769978 | tok/s 2627159.89 | norm 0.5738 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1288 | train 3.456911 | val 3.769978 | tok/s 2628098.85 | norm 0.5988 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1289 | train 3.345895 | val 3.769978 | tok/s 2628258.82 | norm 0.5811 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1290 | train 3.385254 | val 3.769978 | tok/s 2624118.15 | norm 0.5518 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1291 | train 3.447100 | val 3.769978 | tok/s 2620669.12 | norm 0.5592 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1292 | train 3.428581 | val 3.769978 | tok/s 2629528.20 | norm 0.7355 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1293 | train 3.380057 | val 3.769978 | tok/s 2627827.26 | norm 0.6025 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1294 | train 3.394220 | val 3.769978 | tok/s 2629045.94 | norm 0.5220 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1295 | train 3.459254 | val 3.769978 | tok/s 2617446.25 | norm 0.5972 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1296 | train 3.264148 | val 3.769978 | tok/s 2623410.69 | norm 0.6175 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1297 | train 3.519107 | val 3.769978 | tok/s 2627850.05 | norm 0.6056 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1298 | train 3.430078 | val 3.769978 | tok/s 2623631.11 | norm 0.6207 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1299 | train 3.370327 | val 3.769978 | tok/s 2619292.53 | norm 0.6810 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1300 | train 3.524644 | val 3.769978 | tok/s 2618378.47 | norm 0.6120 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1301 | train 3.402357 | val 3.769978 | tok/s 2624237.80 | norm 0.5209 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1302 | train 3.544794 | val 3.769978 | tok/s 2627461.23 | norm 0.6877 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1303 | train 3.656939 | val 3.769978 | tok/s 2626335.95 | norm 0.6490 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1304 | train 3.297119 | val 3.769978 | tok/s 2625856.43 | norm 0.5247 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1305 | train 3.309986 | val 3.769978 | tok/s 2615478.20 | norm 0.6135 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1306 | train 3.305932 | val 3.769978 | tok/s 2626320.93 | norm 0.6731 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1307 | train 3.262618 | val 3.769978 | tok/s 2626539.99 | norm 0.6710 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1308 | train 3.315893 | val 3.769978 | tok/s 2629352.15 | norm 0.6458 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1309 | train 3.375593 | val 3.769978 | tok/s 2626657.74 | norm 0.5788 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1310 | train 3.420256 | val 3.769978 | tok/s 2619148.77 | norm 0.5754 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1311 | train 3.371065 | val 3.769978 | tok/s 2623075.57 | norm 0.6147 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1312 | train 3.385745 | val 3.769978 | tok/s 2627720.17 | norm 0.5742 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1313 | train 3.384381 | val 3.769978 | tok/s 2626719.65 | norm 0.6119 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1314 | train 3.430751 | val 3.769978 | tok/s 2629683.74 | norm 0.5026 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1315 | train 3.430319 | val 3.769978 | tok/s 2622748.67 | norm 0.4960 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1316 | train 3.444310 | val 3.769978 | tok/s 2624811.83 | norm 0.5857 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1317 | train 3.397654 | val 3.769978 | tok/s 2628760.51 | norm 0.5956 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1318 | train 3.440176 | val 3.769978 | tok/s 2486984.60 | norm 0.5813 | dt 210.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1319 | train 3.412734 | val 3.769978 | tok/s 2613959.41 | norm 0.5707 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1320 | train 3.321890 | val 3.769978 | tok/s 2623543.74 | norm 0.5626 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1321 | train 3.359397 | val 3.769978 | tok/s 2625164.61 | norm 0.5726 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1322 | train 3.344713 | val 3.769978 | tok/s 2623425.79 | norm 0.5498 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1323 | train 3.337824 | val 3.769978 | tok/s 2628765.49 | norm 0.6228 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1324 | train 3.462063 | val 3.769978 | tok/s 2630794.29 | norm 0.6370 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1325 | train 3.364587 | val 3.769978 | tok/s 2630503.39 | norm 0.6291 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1326 | train 3.368057 | val 3.769978 | tok/s 2628950.90 | norm 0.5878 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1327 | train 3.384193 | val 3.769978 | tok/s 2633204.80 | norm 0.5968 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1328 | train 3.443131 | val 3.769978 | tok/s 2627511.82 | norm 0.6602 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1329 | train 3.430245 | val 3.769978 | tok/s 2621895.52 | norm 0.6601 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1330 | train 3.336584 | val 3.769978 | tok/s 2624018.30 | norm 0.6487 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1331 | train 3.356226 | val 3.769978 | tok/s 2626647.00 | norm 0.5703 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1332 | train 3.380425 | val 3.769978 | tok/s 2628072.48 | norm 0.6293 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1333 | train 3.297719 | val 3.769978 | tok/s 2624748.71 | norm 0.6443 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1334 | train 3.440533 | val 3.769978 | tok/s 2626369.42 | norm 0.5853 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1335 | train 3.294075 | val 3.769978 | tok/s 2627420.88 | norm 0.5812 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB /home/operator/envv/lib/python3.12/site-packages/torch/distributed/c10d_logger.py:83: UserWarning: barrier(): using the device under current context. You can specify `device_id` in `init_process_group` to mute this warning. return func(*args, **kwargs) step 1336 | train 3.391844 | val 3.769978 | tok/s 2608250.25 | norm 0.5551 | dt 201.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1337 | train 3.339034 | val 3.769978 | tok/s 2628060.83 | norm 0.5784 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1338 | train 3.452218 | val 3.769978 | tok/s 2481053.30 | norm 0.5817 | dt 211.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1339 | train 3.448184 | val 3.769978 | tok/s 2521094.22 | norm 0.5426 | dt 207.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1340 | train 3.302336 | val 3.769978 | tok/s 2486740.94 | norm 0.6534 | dt 210.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1341 | train 3.364227 | val 3.769978 | tok/s 2605080.17 | norm 0.6609 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1342 | train 3.327419 | val 3.769978 | tok/s 2627559.90 | norm 0.6105 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1343 | train 3.393849 | val 3.769978 | tok/s 2625620.50 | norm 0.5859 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1344 | train 3.301219 | val 3.769978 | tok/s 2621061.89 | norm 0.6984 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1345 | train 3.315124 | val 3.769978 | tok/s 2623100.88 | norm 0.7192 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1346 | train 3.375969 | val 3.769978 | tok/s 2627625.28 | norm 0.7103 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1347 | train 3.335752 | val 3.769978 | tok/s 2628798.45 | norm 0.6425 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1348 | train 3.287660 | val 3.769978 | tok/s 2628276.92 | norm 0.6106 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1349 | train 3.368250 | val 3.769978 | tok/s 2630932.85 | norm 0.5888 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1350 | train 3.452970 | val 3.769978 | tok/s 2625179.45 | norm 0.5861 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1351 | train 3.479050 | val 3.769978 | tok/s 2627529.97 | norm 0.5239 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1352 | train 3.413699 | val 3.769978 | tok/s 2626232.39 | norm 0.5523 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1353 | train 3.362742 | val 3.769978 | tok/s 2628391.45 | norm 0.6048 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1354 | train 3.402488 | val 3.769978 | tok/s 2626274.21 | norm 0.5411 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1355 | train 3.382812 | val 3.769978 | tok/s 2624081.30 | norm 0.5276 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1356 | train 3.392373 | val 3.769978 | tok/s 2627303.85 | norm 0.5057 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1357 | train 3.195592 | val 3.769978 | tok/s 2626895.25 | norm 0.5124 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1358 | train 3.368666 | val 3.769978 | tok/s 2631161.38 | norm 0.5301 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1359 | train 3.373928 | val 3.769978 | tok/s 2624670.80 | norm 0.5277 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1360 | train 3.486189 | val 3.769978 | tok/s 2623661.14 | norm 0.5626 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1361 | train 3.436513 | val 3.769978 | tok/s 2627040.76 | norm 0.5368 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1362 | train 3.311543 | val 3.769978 | tok/s 2624120.37 | norm 0.5789 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1363 | train 3.335380 | val 3.769978 | tok/s 2625881.47 | norm 0.5624 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1364 | train 3.282250 | val 3.769978 | tok/s 2616734.26 | norm 0.5457 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1365 | train 3.523833 | val 3.769978 | tok/s 2622277.45 | norm 0.5691 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1366 | train 3.377005 | val 3.769978 | tok/s 2628515.40 | norm 0.5656 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1367 | train 3.401951 | val 3.769978 | tok/s 2625191.41 | norm 0.5727 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1368 | train 3.362667 | val 3.769978 | tok/s 2624620.73 | norm 0.5892 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1369 | train 3.532859 | val 3.769978 | tok/s 2621354.78 | norm 0.6785 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1370 | train 3.334823 | val 3.769978 | tok/s 2624693.65 | norm 0.6077 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1371 | train 3.364518 | val 3.769978 | tok/s 2626132.89 | norm 0.5253 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1372 | train 3.437950 | val 3.769978 | tok/s 2620309.98 | norm 0.5844 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1373 | train 3.408905 | val 3.769978 | tok/s 2630566.01 | norm 0.5938 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1374 | train 3.360898 | val 3.769978 | tok/s 2622304.68 | norm 0.5732 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1375 | train 3.326766 | val 3.769978 | tok/s 2622309.19 | norm 0.6146 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1376 | train 3.325206 | val 3.769978 | tok/s 2625822.43 | norm 0.5936 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1377 | train 3.276022 | val 3.769978 | tok/s 2627337.32 | norm 0.5934 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1378 | train 3.364105 | val 3.769978 | tok/s 2630439.19 | norm 0.5912 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1379 | train 3.363771 | val 3.769978 | tok/s 2619106.51 | norm 0.6184 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1380 | train 3.342265 | val 3.769978 | tok/s 2622424.30 | norm 0.5421 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1381 | train 3.363729 | val 3.769978 | tok/s 2629181.18 | norm 0.5368 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1382 | train 3.343671 | val 3.769978 | tok/s 2622936.35 | norm 0.5904 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1383 | train 3.412188 | val 3.769978 | tok/s 2623269.73 | norm 0.5999 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1384 | train 3.365966 | val 3.769978 | tok/s 2624649.12 | norm 0.5894 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1385 | train 3.466952 | val 3.769978 | tok/s 2591737.03 | norm 0.5731 | dt 202.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1386 | train 3.311512 | val 3.769978 | tok/s 2630534.86 | norm 0.5775 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1387 | train 3.395812 | val 3.769978 | tok/s 2626106.29 | norm 0.6047 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1388 | train 3.353508 | val 3.769978 | tok/s 2625298.59 | norm 0.6325 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1389 | train 3.308119 | val 3.769978 | tok/s 2626126.38 | norm 0.6172 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1390 | train 3.524157 | val 3.769978 | tok/s 2625845.76 | norm 0.6759 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1391 | train 3.420991 | val 3.769978 | tok/s 2629393.32 | norm 0.6223 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1392 | train 3.482131 | val 3.769978 | tok/s 2627536.16 | norm 0.5333 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1393 | train 3.259448 | val 3.769978 | tok/s 2628149.53 | norm 0.5050 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1394 | train 3.409018 | val 3.769978 | tok/s 2624375.02 | norm 0.6181 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1395 | train 3.311108 | val 3.769978 | tok/s 2622764.38 | norm 0.5971 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1396 | train 3.394856 | val 3.769978 | tok/s 2631298.67 | norm 0.5593 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1397 | train 3.392547 | val 3.769978 | tok/s 2629788.07 | norm 0.5569 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1398 | train 3.337701 | val 3.769978 | tok/s 2628809.66 | norm 0.5964 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1399 | train 3.397215 | val 3.769978 | tok/s 2619132.37 | norm 0.6650 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1400 | train 3.343069 | val 3.769978 | tok/s 2607931.02 | norm 0.5983 | dt 201.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1401 | train 3.316703 | val 3.769978 | tok/s 2628841.50 | norm 0.5763 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1402 | train 3.307509 | val 3.769978 | tok/s 2622678.53 | norm 0.5802 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1403 | train 3.424161 | val 3.769978 | tok/s 2623639.57 | norm 0.5388 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1404 | train 3.350426 | val 3.769978 | tok/s 2595298.04 | norm 0.5859 | dt 202.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1405 | train 3.385812 | val 3.769978 | tok/s 2623339.81 | norm 0.6343 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1406 | train 3.349831 | val 3.769978 | tok/s 2601848.99 | norm 0.6680 | dt 201.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1407 | train 3.307482 | val 3.769978 | tok/s 2619878.54 | norm 0.7735 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1408 | train 3.211350 | val 3.769978 | tok/s 2628138.05 | norm 0.6512 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1409 | train 3.520590 | val 3.769978 | tok/s 2624989.21 | norm 0.5529 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1410 | train 3.388992 | val 3.769978 | tok/s 2627329.03 | norm 0.5473 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1411 | train 3.288833 | val 3.769978 | tok/s 2622163.60 | norm 0.6300 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1412 | train 3.323785 | val 3.769978 | tok/s 2484038.55 | norm 0.6742 | dt 211.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1413 | train 3.272497 | val 3.769978 | tok/s 2627448.57 | norm 0.5999 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1414 | train 3.427448 | val 3.769978 | tok/s 2627426.29 | norm 0.5699 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1415 | train 3.430038 | val 3.769978 | tok/s 2624284.86 | norm 0.5231 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1416 | train 3.426623 | val 3.769978 | tok/s 2609113.95 | norm 0.5986 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1417 | train 3.266694 | val 3.769978 | tok/s 2625520.66 | norm 0.5873 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1418 | train 3.308791 | val 3.769978 | tok/s 2627966.50 | norm 0.5274 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1419 | train 3.321525 | val 3.769978 | tok/s 2624968.25 | norm 0.5568 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1420 | train 3.210672 | val 3.769978 | tok/s 2629584.81 | norm 0.6138 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1421 | train 3.348801 | val 3.769978 | tok/s 2578281.88 | norm 0.6992 | dt 203.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1422 | train 3.464446 | val 3.769978 | tok/s 2624877.97 | norm 0.6233 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1423 | train 3.280013 | val 3.769978 | tok/s 2628106.69 | norm 0.5918 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1424 | train 3.330755 | val 3.769978 | tok/s 2628063.84 | norm 0.6528 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1425 | train 3.269643 | val 3.769978 | tok/s 2631230.15 | norm 0.5905 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1426 | train 3.412563 | val 3.769978 | tok/s 2626477.99 | norm 0.5590 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1427 | train 3.326725 | val 3.769978 | tok/s 2624075.39 | norm 0.5902 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1428 | train 3.341604 | val 3.769978 | tok/s 2629710.81 | norm 0.5860 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1429 | train 3.332392 | val 3.769978 | tok/s 2586011.91 | norm 0.6056 | dt 202.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1430 | train 3.376164 | val 3.769978 | tok/s 2518255.45 | norm 0.5820 | dt 208.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1431 | train 3.430482 | val 3.769978 | tok/s 2623750.45 | norm 0.6011 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1432 | train 3.302656 | val 3.769978 | tok/s 2624442.61 | norm 0.5940 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1433 | train 3.451067 | val 3.769978 | tok/s 2620983.53 | norm 0.5265 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1434 | train 3.325893 | val 3.769978 | tok/s 2627506.36 | norm 0.5307 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1435 | train 3.344898 | val 3.769978 | tok/s 2625848.42 | norm 0.5552 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1436 | train 3.366061 | val 3.769978 | tok/s 2620739.95 | norm 0.5336 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1437 | train 3.343950 | val 3.769978 | tok/s 2621111.84 | norm 0.5497 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1438 | train 3.332172 | val 3.769978 | tok/s 2628938.17 | norm 0.5668 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1439 | train 3.269129 | val 3.769978 | tok/s 2628689.83 | norm 0.5467 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1440 | train 3.373998 | val 3.769978 | tok/s 2624451.95 | norm 0.6070 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1441 | train 3.254347 | val 3.769978 | tok/s 2618193.24 | norm 0.5509 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1442 | train 3.370569 | val 3.769978 | tok/s 2611015.46 | norm 0.5258 | dt 200.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1443 | train 3.393715 | val 3.769978 | tok/s 2625797.43 | norm 0.5813 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1444 | train 3.386547 | val 3.769978 | tok/s 2626647.66 | norm 0.5791 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1445 | train 3.478620 | val 3.769978 | tok/s 2624830.59 | norm 0.5787 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1446 | train 3.300581 | val 3.769978 | tok/s 2621041.84 | norm 0.5033 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1447 | train 3.364462 | val 3.769978 | tok/s 2624513.94 | norm 0.4951 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1448 | train 3.373235 | val 3.769978 | tok/s 2627368.75 | norm 0.5568 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1449 | train 3.346331 | val 3.769978 | tok/s 2624438.92 | norm 0.5827 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1450 | train 3.320409 | val 3.769978 | tok/s 2622987.89 | norm 0.5453 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1451 | train 3.406705 | val 3.769978 | tok/s 2622757.17 | norm 0.5603 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1452 | train 3.348113 | val 3.769978 | tok/s 2623021.42 | norm 0.6124 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1453 | train 3.355012 | val 3.769978 | tok/s 2632167.73 | norm 0.5799 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1454 | train 3.335813 | val 3.769978 | tok/s 2628276.29 | norm 0.5848 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1455 | train 3.335024 | val 3.769978 | tok/s 2630112.82 | norm 0.6270 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1456 | train 3.355636 | val 3.769978 | tok/s 2628788.08 | norm 0.6029 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1457 | train 3.408762 | val 3.769978 | tok/s 2626759.03 | norm 0.6177 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1458 | train 3.286146 | val 3.769978 | tok/s 2628266.17 | norm 0.6568 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1459 | train 3.285548 | val 3.769978 | tok/s 2627289.36 | norm 0.6203 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1460 | train 3.290171 | val 3.769978 | tok/s 2628242.04 | norm 0.5844 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1461 | train 3.326959 | val 3.769978 | tok/s 2628829.63 | norm 0.5748 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1462 | train 3.334278 | val 3.769978 | tok/s 2627480.18 | norm 0.5699 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1463 | train 3.397139 | val 3.769978 | tok/s 2630442.49 | norm 0.5936 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1464 | train 3.324487 | val 3.769978 | tok/s 2627189.71 | norm 0.5272 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1465 | train 3.359197 | val 3.769978 | tok/s 2623218.21 | norm 0.4945 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1466 | train 3.373964 | val 3.769978 | tok/s 2623691.93 | norm 0.5316 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1467 | train 3.420140 | val 3.769978 | tok/s 2625529.54 | norm 0.5556 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1468 | train 3.273913 | val 3.769978 | tok/s 2629961.25 | norm 0.5674 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1469 | train 3.382990 | val 3.769978 | tok/s 2629935.02 | norm 0.5390 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1470 | train 3.301420 | val 3.769978 | tok/s 2630860.67 | norm 0.5070 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1471 | train 3.300112 | val 3.769978 | tok/s 2624834.90 | norm 0.5340 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1472 | train 3.400715 | val 3.769978 | tok/s 2627878.59 | norm 0.5759 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1473 | train 3.365757 | val 3.769978 | tok/s 2628879.80 | norm 0.6638 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1474 | train 3.331656 | val 3.769978 | tok/s 2628488.90 | norm 0.6037 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1475 | train 3.311262 | val 3.769978 | tok/s 2627465.98 | norm 0.5846 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1476 | train 3.222646 | val 3.769978 | tok/s 2620549.42 | norm 0.5578 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1477 | train 3.310718 | val 3.769978 | tok/s 2620901.95 | norm 0.5728 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1478 | train 3.334438 | val 3.769978 | tok/s 2626925.99 | norm 0.5242 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1479 | train 3.521915 | val 3.769978 | tok/s 2623957.84 | norm 0.4852 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1480 | train 3.481335 | val 3.769978 | tok/s 2627468.96 | norm 0.5556 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1481 | train 3.324287 | val 3.769978 | tok/s 2619084.76 | norm 0.5939 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1482 | train 3.370891 | val 3.769978 | tok/s 2622378.61 | norm 0.5782 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1483 | train 3.412215 | val 3.769978 | tok/s 2622689.93 | norm 0.6314 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1484 | train 3.438323 | val 3.769978 | tok/s 2627101.76 | norm 0.6266 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1485 | train 3.414174 | val 3.769978 | tok/s 2628267.52 | norm 0.5471 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1486 | train 3.328181 | val 3.769978 | tok/s 2620111.83 | norm 0.5604 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1487 | train 3.308166 | val 3.769978 | tok/s 2626565.82 | norm 0.5473 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1488 | train 3.309757 | val 3.769978 | tok/s 2624790.74 | norm 0.5650 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1489 | train 3.334567 | val 3.769978 | tok/s 2627241.72 | norm 0.5895 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1490 | train 3.344042 | val 3.769978 | tok/s 2629303.83 | norm 0.5861 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1491 | train 3.371655 | val 3.769978 | tok/s 2622714.60 | norm 0.6378 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1492 | train 3.259025 | val 3.769978 | tok/s 2622124.68 | norm 0.6858 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1493 | train 3.317502 | val 3.769978 | tok/s 2628154.28 | norm 0.6049 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1494 | train 3.306720 | val 3.769978 | tok/s 2623118.99 | norm 0.5450 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1495 | train 3.320020 | val 3.769978 | tok/s 2627587.78 | norm 0.6088 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1496 | train 3.311014 | val 3.769978 | tok/s 2621243.70 | norm 0.6214 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1497 | train 3.329439 | val 3.769978 | tok/s 2625377.67 | norm 0.6371 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1498 | train 3.154581 | val 3.769978 | tok/s 2629573.20 | norm 0.6519 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1499 | train 3.384635 | val 3.769978 | tok/s 2626809.94 | norm 0.6241 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 1500 | val_loss 3.706957 step 1500 | train 3.413533 | val 3.706957 | tok/s 2612087.13 | norm 0.5850 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1501 | train 3.247970 | val 3.706957 | tok/s 2627518.25 | norm 0.6081 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1502 | train 3.353180 | val 3.706957 | tok/s 2622415.81 | norm 0.5658 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1503 | train 3.357442 | val 3.706957 | tok/s 2625953.70 | norm 0.5579 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1504 | train 3.344928 | val 3.706957 | tok/s 2626009.65 | norm 0.6034 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1505 | train 3.335760 | val 3.706957 | tok/s 2624323.87 | norm 0.5603 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1506 | train 3.466209 | val 3.706957 | tok/s 2627915.58 | norm 0.5535 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1507 | train 3.404523 | val 3.706957 | tok/s 2622188.40 | norm 0.6055 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1508 | train 3.348160 | val 3.706957 | tok/s 2626461.11 | norm 0.6002 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1509 | train 3.211601 | val 3.706957 | tok/s 2628645.18 | norm 0.6027 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1510 | train 3.296829 | val 3.706957 | tok/s 2628487.41 | norm 0.5962 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1511 | train 3.293043 | val 3.706957 | tok/s 2623837.18 | norm 0.5096 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1512 | train 3.333329 | val 3.706957 | tok/s 2621653.40 | norm 0.5533 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1513 | train 3.307860 | val 3.706957 | tok/s 2622937.08 | norm 0.6156 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1514 | train 3.312256 | val 3.706957 | tok/s 2632194.65 | norm 0.5638 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1515 | train 3.380832 | val 3.706957 | tok/s 2623581.46 | norm 0.5712 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1516 | train 3.246342 | val 3.706957 | tok/s 2629318.32 | norm 0.5793 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1517 | train 3.524168 | val 3.706957 | tok/s 2618839.39 | norm 0.6004 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1518 | train 3.290164 | val 3.706957 | tok/s 2606527.37 | norm 0.5248 | dt 201.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1519 | train 3.369443 | val 3.706957 | tok/s 2628349.46 | norm 0.4824 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1520 | train 3.295330 | val 3.706957 | tok/s 2629420.94 | norm 0.5298 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1521 | train 3.414200 | val 3.706957 | tok/s 2624999.14 | norm 0.5680 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1522 | train 3.451725 | val 3.706957 | tok/s 2625431.00 | norm 0.5802 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1523 | train 3.346748 | val 3.706957 | tok/s 2624863.25 | norm 0.5948 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1524 | train 3.359877 | val 3.706957 | tok/s 2630059.63 | norm 0.5790 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1525 | train 3.385085 | val 3.706957 | tok/s 2625878.35 | norm 0.5055 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1526 | train 3.404245 | val 3.706957 | tok/s 2627928.09 | norm 0.5041 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1527 | train 3.292934 | val 3.706957 | tok/s 2621903.34 | norm 0.5545 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1528 | train 3.438358 | val 3.706957 | tok/s 2622542.49 | norm 0.5544 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1529 | train 3.292866 | val 3.706957 | tok/s 2627635.17 | norm 0.6203 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1530 | train 3.337361 | val 3.706957 | tok/s 2626191.95 | norm 0.6526 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1531 | train 3.389017 | val 3.706957 | tok/s 2620997.64 | norm 0.6539 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1532 | train 3.361905 | val 3.706957 | tok/s 2621466.08 | norm 0.6250 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1533 | train 3.327634 | val 3.706957 | tok/s 2628190.77 | norm 0.6551 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1534 | train 3.217712 | val 3.706957 | tok/s 2626647.80 | norm 0.5606 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1535 | train 3.310372 | val 3.706957 | tok/s 2622845.16 | norm 0.5988 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1536 | train 3.316282 | val 3.706957 | tok/s 2629466.80 | norm 0.6124 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1537 | train 3.340097 | val 3.706957 | tok/s 2621389.42 | norm 0.5878 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1538 | train 3.286820 | val 3.706957 | tok/s 2625107.28 | norm 0.6721 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1539 | train 3.299098 | val 3.706957 | tok/s 2631132.88 | norm 0.5970 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1540 | train 3.331320 | val 3.706957 | tok/s 2614227.82 | norm 0.5311 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1541 | train 3.212027 | val 3.706957 | tok/s 2629761.97 | norm 0.6024 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1542 | train 3.357456 | val 3.706957 | tok/s 2624818.93 | norm 0.5947 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1543 | train 3.390352 | val 3.706957 | tok/s 2628273.89 | norm 0.5498 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1544 | train 3.338011 | val 3.706957 | tok/s 2630186.70 | norm 0.6554 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1545 | train 3.164950 | val 3.706957 | tok/s 2629235.16 | norm 0.6872 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1546 | train 3.612734 | val 3.706957 | tok/s 2628330.75 | norm 0.6951 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1547 | train 3.280909 | val 3.706957 | tok/s 2615911.15 | norm 0.6125 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1548 | train 3.340056 | val 3.706957 | tok/s 2623890.44 | norm 0.5845 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1549 | train 3.290190 | val 3.706957 | tok/s 2627843.40 | norm 0.5997 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1550 | train 3.228041 | val 3.706957 | tok/s 2623827.84 | norm 0.5946 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1551 | train 3.260012 | val 3.706957 | tok/s 2619604.43 | norm 0.5364 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1552 | train 3.249400 | val 3.706957 | tok/s 2621015.00 | norm 0.5123 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1553 | train 3.319405 | val 3.706957 | tok/s 2608913.06 | norm 0.5449 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1554 | train 3.346539 | val 3.706957 | tok/s 2626399.84 | norm 0.6085 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1555 | train 3.284810 | val 3.706957 | tok/s 2621283.94 | norm 0.6037 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1556 | train 3.414270 | val 3.706957 | tok/s 2628353.40 | norm 0.5447 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1557 | train 3.172961 | val 3.706957 | tok/s 2624263.96 | norm 0.5455 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1558 | train 3.342870 | val 3.706957 | tok/s 2626476.66 | norm 0.5560 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1559 | train 3.354578 | val 3.706957 | tok/s 2618195.44 | norm 0.5728 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1560 | train 3.261646 | val 3.706957 | tok/s 2620352.51 | norm 0.5395 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1561 | train 3.275184 | val 3.706957 | tok/s 2625866.88 | norm 0.5591 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1562 | train 3.385671 | val 3.706957 | tok/s 2624607.69 | norm 0.5918 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1563 | train 3.314353 | val 3.706957 | tok/s 2630137.10 | norm 0.5544 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1564 | train 3.367030 | val 3.706957 | tok/s 2630049.94 | norm 0.5544 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1565 | train 3.373555 | val 3.706957 | tok/s 2602416.71 | norm 0.6478 | dt 201.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1566 | train 3.477612 | val 3.706957 | tok/s 2613993.63 | norm 0.5795 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1567 | train 3.532982 | val 3.706957 | tok/s 2478339.69 | norm 0.5980 | dt 211.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1568 | train 3.398910 | val 3.706957 | tok/s 2627205.15 | norm 0.5792 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1569 | train 3.374225 | val 3.706957 | tok/s 2626218.02 | norm 0.5609 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1570 | train 3.301277 | val 3.706957 | tok/s 2626321.24 | norm 0.5360 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1571 | train 3.469878 | val 3.706957 | tok/s 2629473.24 | norm 0.4980 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1572 | train 3.362785 | val 3.706957 | tok/s 2626885.97 | norm 0.5074 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1573 | train 3.331395 | val 3.706957 | tok/s 2625360.57 | norm 0.5243 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1574 | train 3.269092 | val 3.706957 | tok/s 2620006.32 | norm 0.5449 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1575 | train 3.328581 | val 3.706957 | tok/s 2623867.92 | norm 0.5801 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1576 | train 3.430805 | val 3.706957 | tok/s 2627487.10 | norm 0.5624 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1577 | train 3.396976 | val 3.706957 | tok/s 2625231.28 | norm 0.5834 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1578 | train 3.300834 | val 3.706957 | tok/s 2628299.06 | norm 0.5861 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1579 | train 3.407205 | val 3.706957 | tok/s 2619348.62 | norm 0.5421 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1580 | train 3.241089 | val 3.706957 | tok/s 2620886.70 | norm 0.5289 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1581 | train 3.295948 | val 3.706957 | tok/s 2627554.00 | norm 0.5004 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1582 | train 3.326613 | val 3.706957 | tok/s 2622260.37 | norm 0.5295 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1583 | train 3.402293 | val 3.706957 | tok/s 2627221.13 | norm 0.5681 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1584 | train 3.505701 | val 3.706957 | tok/s 2618672.72 | norm 0.6107 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1585 | train 3.278146 | val 3.706957 | tok/s 2626005.94 | norm 0.4851 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1586 | train 3.458746 | val 3.706957 | tok/s 2630412.23 | norm 0.5196 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1587 | train 3.455929 | val 3.706957 | tok/s 2629585.15 | norm 0.5812 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1588 | train 3.354233 | val 3.706957 | tok/s 2625085.71 | norm 0.5452 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1589 | train 3.422368 | val 3.706957 | tok/s 2596188.43 | norm 0.6078 | dt 201.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1590 | train 3.329205 | val 3.706957 | tok/s 2624007.57 | norm 0.5666 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1591 | train 3.297126 | val 3.706957 | tok/s 2629383.05 | norm 0.5818 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1592 | train 3.426830 | val 3.706957 | tok/s 2624205.19 | norm 0.5972 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1593 | train 3.273391 | val 3.706957 | tok/s 2628783.62 | norm 0.5225 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1594 | train 3.271018 | val 3.706957 | tok/s 2629167.43 | norm 0.6513 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1595 | train 3.337097 | val 3.706957 | tok/s 2625784.36 | norm 0.6477 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1596 | train 3.262687 | val 3.706957 | tok/s 2627738.53 | norm 0.6098 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1597 | train 3.335865 | val 3.706957 | tok/s 2624382.58 | norm 0.6626 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1598 | train 3.248476 | val 3.706957 | tok/s 2628899.17 | norm 0.6571 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1599 | train 3.359269 | val 3.706957 | tok/s 2613751.78 | norm 0.5676 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1600 | train 3.339366 | val 3.706957 | tok/s 2627820.26 | norm 0.5519 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1601 | train 3.338325 | val 3.706957 | tok/s 2627301.29 | norm 0.5395 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1602 | train 3.333721 | val 3.706957 | tok/s 2622439.41 | norm 0.5253 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1603 | train 3.221170 | val 3.706957 | tok/s 2628853.38 | norm 0.5900 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1604 | train 3.380642 | val 3.706957 | tok/s 2617519.31 | norm 0.6356 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1605 | train 3.348062 | val 3.706957 | tok/s 2598816.56 | norm 0.6434 | dt 201.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1606 | train 3.364679 | val 3.706957 | tok/s 2631015.56 | norm 0.6042 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1607 | train 3.330073 | val 3.706957 | tok/s 2626598.80 | norm 0.5424 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1608 | train 3.309587 | val 3.706957 | tok/s 2629190.14 | norm 0.5576 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1609 | train 3.294467 | val 3.706957 | tok/s 2623190.03 | norm 0.5556 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1610 | train 3.310291 | val 3.706957 | tok/s 2626258.76 | norm 0.5625 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1611 | train 3.329554 | val 3.706957 | tok/s 2627902.75 | norm 0.5783 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1612 | train 3.242226 | val 3.706957 | tok/s 2627386.58 | norm 0.5605 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1613 | train 3.229500 | val 3.706957 | tok/s 2628641.23 | norm 0.6018 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1614 | train 3.315341 | val 3.706957 | tok/s 2628432.83 | norm 0.5502 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1615 | train 3.257379 | val 3.706957 | tok/s 2621981.87 | norm 0.5387 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1616 | train 3.287853 | val 3.706957 | tok/s 2628295.57 | norm 0.5387 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1617 | train 3.261869 | val 3.706957 | tok/s 2630246.47 | norm 0.5534 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1618 | train 3.252558 | val 3.706957 | tok/s 2628383.33 | norm 0.5874 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1619 | train 3.234795 | val 3.706957 | tok/s 2627088.73 | norm 0.6157 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1620 | train 3.318273 | val 3.706957 | tok/s 2626773.93 | norm 0.5718 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1621 | train 3.395857 | val 3.706957 | tok/s 2631109.67 | norm 0.6728 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1622 | train 3.209364 | val 3.706957 | tok/s 2624069.72 | norm 0.6361 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1623 | train 3.252431 | val 3.706957 | tok/s 2626320.81 | norm 0.6553 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1624 | train 3.244887 | val 3.706957 | tok/s 2623736.35 | norm 0.6252 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1625 | train 3.411565 | val 3.706957 | tok/s 2621091.07 | norm 0.6227 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1626 | train 3.392301 | val 3.706957 | tok/s 2628190.53 | norm 0.5745 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1627 | train 3.382596 | val 3.706957 | tok/s 2628800.79 | norm 0.5207 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1628 | train 3.397344 | val 3.706957 | tok/s 2626500.35 | norm 0.5190 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1629 | train 3.175608 | val 3.706957 | tok/s 2620945.15 | norm 0.5587 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1630 | train 3.368533 | val 3.706957 | tok/s 2626616.46 | norm 0.5170 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1631 | train 3.276584 | val 3.706957 | tok/s 2623643.23 | norm 0.5393 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1632 | train 3.135668 | val 3.706957 | tok/s 2626015.41 | norm 0.5690 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1633 | train 3.261595 | val 3.706957 | tok/s 2627331.83 | norm 0.5068 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1634 | train 3.337341 | val 3.706957 | tok/s 2625898.16 | norm 0.5408 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1635 | train 3.404458 | val 3.706957 | tok/s 2628058.36 | norm 0.5700 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1636 | train 3.466503 | val 3.706957 | tok/s 2631427.86 | norm 0.5561 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1637 | train 3.278266 | val 3.706957 | tok/s 2625081.53 | norm 0.5384 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1638 | train 3.274303 | val 3.706957 | tok/s 2627028.29 | norm 0.5703 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1639 | train 3.209727 | val 3.706957 | tok/s 2624182.18 | norm 0.5708 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1640 | train 3.345907 | val 3.706957 | tok/s 2613007.91 | norm 0.5305 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1641 | train 3.309513 | val 3.706957 | tok/s 2565305.47 | norm 0.5522 | dt 204.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1642 | train 3.280504 | val 3.706957 | tok/s 2623125.24 | norm 0.6123 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1643 | train 3.320982 | val 3.706957 | tok/s 2634407.04 | norm 0.6288 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1644 | train 3.284450 | val 3.706957 | tok/s 2632378.40 | norm 0.6036 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1645 | train 3.326488 | val 3.706957 | tok/s 2633693.85 | norm 0.6209 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1646 | train 3.301198 | val 3.706957 | tok/s 2632073.80 | norm 0.5785 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1647 | train 3.368814 | val 3.706957 | tok/s 2624448.31 | norm 0.5992 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1648 | train 3.230479 | val 3.706957 | tok/s 2627662.32 | norm 0.6062 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1649 | train 3.355461 | val 3.706957 | tok/s 2625488.69 | norm 0.6035 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1650 | train 3.224339 | val 3.706957 | tok/s 2627166.18 | norm 0.5810 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1651 | train 3.274875 | val 3.706957 | tok/s 2627944.44 | norm 0.5913 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1652 | train 3.279874 | val 3.706957 | tok/s 2625314.11 | norm 0.5690 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1653 | train 3.245044 | val 3.706957 | tok/s 2632735.89 | norm 0.5759 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1654 | train 3.326290 | val 3.706957 | tok/s 2626777.63 | norm 0.5635 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1655 | train 3.201783 | val 3.706957 | tok/s 2625436.86 | norm 0.5354 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1656 | train 3.296250 | val 3.706957 | tok/s 2630709.75 | norm 0.5339 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1657 | train 3.265336 | val 3.706957 | tok/s 2613260.86 | norm 0.5484 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1658 | train 3.346572 | val 3.706957 | tok/s 2627846.93 | norm 0.5146 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1659 | train 3.328740 | val 3.706957 | tok/s 2624665.28 | norm 0.5553 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1660 | train 3.350898 | val 3.706957 | tok/s 2625893.20 | norm 0.5946 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1661 | train 3.326691 | val 3.706957 | tok/s 2603598.41 | norm 0.5600 | dt 201.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1662 | train 3.245902 | val 3.706957 | tok/s 2625339.13 | norm 0.5209 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1663 | train 3.251842 | val 3.706957 | tok/s 2629186.15 | norm 0.5223 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1664 | train 3.237655 | val 3.706957 | tok/s 2626232.60 | norm 0.5381 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1665 | train 3.283841 | val 3.706957 | tok/s 2609467.41 | norm 0.5638 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1666 | train 3.305604 | val 3.706957 | tok/s 2411956.33 | norm 0.5704 | dt 217.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1667 | train 3.290545 | val 3.706957 | tok/s 2611728.56 | norm 0.6312 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1668 | train 3.320396 | val 3.706957 | tok/s 2631637.49 | norm 0.5620 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1669 | train 3.392073 | val 3.706957 | tok/s 2630147.12 | norm 0.5805 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1670 | train 3.239769 | val 3.706957 | tok/s 2629470.72 | norm 0.6249 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1671 | train 3.287856 | val 3.706957 | tok/s 2623193.07 | norm 0.6352 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1672 | train 3.297346 | val 3.706957 | tok/s 2494439.35 | norm 0.5811 | dt 210.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1673 | train 3.337463 | val 3.706957 | tok/s 2628151.79 | norm 0.5907 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1674 | train 3.467018 | val 3.706957 | tok/s 2627094.80 | norm 0.6153 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1675 | train 3.304200 | val 3.706957 | tok/s 2629883.03 | norm 0.5545 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1676 | train 3.307281 | val 3.706957 | tok/s 2623424.31 | norm 0.5639 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1677 | train 3.238327 | val 3.706957 | tok/s 2625751.42 | norm 0.5897 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1678 | train 3.263513 | val 3.706957 | tok/s 2629060.38 | norm 0.5550 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1679 | train 3.290689 | val 3.706957 | tok/s 2628119.64 | norm 0.6279 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1680 | train 3.412910 | val 3.706957 | tok/s 2591970.51 | norm 0.6349 | dt 202.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1681 | train 3.431152 | val 3.706957 | tok/s 2622711.28 | norm 0.5629 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1682 | train 3.345296 | val 3.706957 | tok/s 2619152.38 | norm 0.5957 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1683 | train 3.235668 | val 3.706957 | tok/s 2623015.03 | norm 0.5924 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1684 | train 3.211256 | val 3.706957 | tok/s 2630041.26 | norm 0.5453 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1685 | train 3.355500 | val 3.706957 | tok/s 2628903.52 | norm 0.5648 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1686 | train 3.284102 | val 3.706957 | tok/s 2623012.61 | norm 0.5271 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1687 | train 3.336087 | val 3.706957 | tok/s 2626970.68 | norm 0.5380 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1688 | train 3.323610 | val 3.706957 | tok/s 2630517.67 | norm 0.6086 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1689 | train 3.256067 | val 3.706957 | tok/s 2624157.66 | norm 0.5864 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1690 | train 3.254990 | val 3.706957 | tok/s 2619126.43 | norm 0.5222 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1691 | train 3.305974 | val 3.706957 | tok/s 2608641.44 | norm 0.5281 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1692 | train 3.323467 | val 3.706957 | tok/s 2605615.46 | norm 0.6020 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1693 | train 3.257022 | val 3.706957 | tok/s 2626313.83 | norm 0.5819 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1694 | train 3.206339 | val 3.706957 | tok/s 2625312.65 | norm 0.5203 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1695 | train 3.327655 | val 3.706957 | tok/s 2625749.16 | norm 0.5028 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1696 | train 3.234641 | val 3.706957 | tok/s 2625225.31 | norm 0.5442 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1697 | train 3.213796 | val 3.706957 | tok/s 2619793.17 | norm 0.5511 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1698 | train 3.230750 | val 3.706957 | tok/s 2629505.79 | norm 0.5946 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1699 | train 3.443699 | val 3.706957 | tok/s 2624044.84 | norm 0.5581 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1700 | train 3.254558 | val 3.706957 | tok/s 2626609.64 | norm 0.5391 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1701 | train 3.403961 | val 3.706957 | tok/s 2622950.74 | norm 0.5608 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1702 | train 3.307786 | val 3.706957 | tok/s 2622837.68 | norm 0.6129 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1703 | train 3.450749 | val 3.706957 | tok/s 2625710.95 | norm 0.6500 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1704 | train 3.302409 | val 3.706957 | tok/s 2626568.76 | norm 0.6211 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1705 | train 3.327158 | val 3.706957 | tok/s 2622390.35 | norm 0.5744 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1706 | train 3.465897 | val 3.706957 | tok/s 2624261.00 | norm 0.5717 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1707 | train 3.181834 | val 3.706957 | tok/s 2629557.94 | norm 0.4954 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1708 | train 3.381294 | val 3.706957 | tok/s 2626483.65 | norm 0.5318 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1709 | train 3.207835 | val 3.706957 | tok/s 2627102.15 | norm 0.6185 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1710 | train 3.269584 | val 3.706957 | tok/s 2622950.62 | norm 0.6296 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1711 | train 3.261223 | val 3.706957 | tok/s 2623369.76 | norm 0.6041 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1712 | train 3.275032 | val 3.706957 | tok/s 2599449.12 | norm 0.5754 | dt 201.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1713 | train 3.319016 | val 3.706957 | tok/s 2628217.11 | norm 0.5838 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1714 | train 3.276791 | val 3.706957 | tok/s 2624883.91 | norm 0.5560 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1715 | train 3.304650 | val 3.706957 | tok/s 2616948.62 | norm 0.5444 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1716 | train 3.359328 | val 3.706957 | tok/s 2627838.16 | norm 0.5529 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1717 | train 3.364459 | val 3.706957 | tok/s 2629005.17 | norm 0.6014 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1718 | train 3.194199 | val 3.706957 | tok/s 2627170.28 | norm 0.5868 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1719 | train 3.291366 | val 3.706957 | tok/s 2625332.17 | norm 0.5579 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1720 | train 3.397671 | val 3.706957 | tok/s 2632823.29 | norm 0.5817 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1721 | train 3.385037 | val 3.706957 | tok/s 2627200.33 | norm 0.6456 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1722 | train 3.321589 | val 3.706957 | tok/s 2629294.15 | norm 0.5807 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1723 | train 3.374287 | val 3.706957 | tok/s 2628649.37 | norm 0.5689 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1724 | train 3.253617 | val 3.706957 | tok/s 2622860.79 | norm 0.4887 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1725 | train 3.247783 | val 3.706957 | tok/s 2625147.75 | norm 0.4963 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1726 | train 3.264086 | val 3.706957 | tok/s 2623891.74 | norm 0.5282 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1727 | train 3.334012 | val 3.706957 | tok/s 2622526.20 | norm 0.5830 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1728 | train 3.319426 | val 3.706957 | tok/s 2622584.48 | norm 0.6089 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1729 | train 3.313072 | val 3.706957 | tok/s 2623825.81 | norm 0.5907 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1730 | train 3.368315 | val 3.706957 | tok/s 2629173.57 | norm 0.6305 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1731 | train 3.106957 | val 3.706957 | tok/s 2620786.57 | norm 0.5331 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1732 | train 3.268814 | val 3.706957 | tok/s 2627913.22 | norm 0.5248 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1733 | train 3.320192 | val 3.706957 | tok/s 2624008.62 | norm 0.5691 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1734 | train 3.339844 | val 3.706957 | tok/s 2621075.17 | norm 0.5686 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1735 | train 3.300886 | val 3.706957 | tok/s 2551441.27 | norm 0.5423 | dt 205.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1736 | train 3.348020 | val 3.706957 | tok/s 2627872.07 | norm 0.5497 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1737 | train 3.334203 | val 3.706957 | tok/s 2623558.59 | norm 0.5312 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1738 | train 3.376482 | val 3.706957 | tok/s 2620176.18 | norm 0.5420 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1739 | train 3.227139 | val 3.706957 | tok/s 2622608.96 | norm 0.5706 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1740 | train 3.290849 | val 3.706957 | tok/s 2626729.58 | norm 0.5833 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1741 | train 3.333088 | val 3.706957 | tok/s 2466208.27 | norm 0.5993 | dt 212.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1742 | train 3.364773 | val 3.706957 | tok/s 2547752.25 | norm 0.5841 | dt 205.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1743 | train 3.211503 | val 3.706957 | tok/s 2630615.56 | norm 0.6126 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1744 | train 3.312348 | val 3.706957 | tok/s 2627788.37 | norm 0.6120 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1745 | train 3.318096 | val 3.706957 | tok/s 2629795.50 | norm 0.5541 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1746 | train 3.326241 | val 3.706957 | tok/s 2626922.19 | norm 0.5463 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1747 | train 3.212059 | val 3.706957 | tok/s 2630027.43 | norm 0.5288 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1748 | train 3.452163 | val 3.706957 | tok/s 2627565.14 | norm 0.5773 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1749 | train 3.324066 | val 3.706957 | tok/s 2627271.32 | norm 0.6037 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 1750 | val_loss 3.654962 step 1750 | train 3.309192 | val 3.654962 | tok/s 2615434.74 | norm 0.7456 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1751 | train 3.271366 | val 3.654962 | tok/s 2625264.88 | norm 0.5934 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1752 | train 3.282329 | val 3.654962 | tok/s 2626734.76 | norm 0.5222 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1753 | train 3.263268 | val 3.654962 | tok/s 2628286.75 | norm 0.5367 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1754 | train 3.215505 | val 3.654962 | tok/s 2625732.09 | norm 0.5429 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1755 | train 3.158096 | val 3.654962 | tok/s 2513336.60 | norm 0.4949 | dt 208.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1756 | train 3.282689 | val 3.654962 | tok/s 2586985.29 | norm 0.5009 | dt 202.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1757 | train 3.451252 | val 3.654962 | tok/s 2626388.42 | norm 0.5354 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1758 | train 3.410537 | val 3.654962 | tok/s 2627936.81 | norm 0.5731 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1759 | train 3.123994 | val 3.654962 | tok/s 2618740.29 | norm 0.5369 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1760 | train 3.300517 | val 3.654962 | tok/s 2624493.23 | norm 0.5454 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1761 | train 3.636325 | val 3.654962 | tok/s 2623909.02 | norm 0.6910 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1762 | train 3.322333 | val 3.654962 | tok/s 2627003.13 | norm 0.5248 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1763 | train 3.283262 | val 3.654962 | tok/s 2614478.56 | norm 0.5097 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1764 | train 3.292135 | val 3.654962 | tok/s 2630030.80 | norm 0.5515 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1765 | train 3.382766 | val 3.654962 | tok/s 2616026.94 | norm 0.5671 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1766 | train 3.215454 | val 3.654962 | tok/s 2626590.61 | norm 0.5813 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1767 | train 3.442961 | val 3.654962 | tok/s 2624940.35 | norm 0.7571 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1768 | train 3.230663 | val 3.654962 | tok/s 2628973.44 | norm 0.6181 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1769 | train 3.259977 | val 3.654962 | tok/s 2623398.63 | norm 0.5625 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1770 | train 3.198632 | val 3.654962 | tok/s 2607137.62 | norm 0.6644 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1771 | train 3.315935 | val 3.654962 | tok/s 2623079.81 | norm 0.6206 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1772 | train 3.513497 | val 3.654962 | tok/s 2625385.06 | norm 0.5361 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1773 | train 3.270582 | val 3.654962 | tok/s 2623624.42 | norm 0.5221 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1774 | train 3.389852 | val 3.654962 | tok/s 2618174.23 | norm 0.5009 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1775 | train 3.264145 | val 3.654962 | tok/s 2626012.73 | norm 0.5392 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1776 | train 3.403466 | val 3.654962 | tok/s 2628154.40 | norm 0.6089 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1777 | train 3.297624 | val 3.654962 | tok/s 2627020.49 | norm 0.6001 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1778 | train 3.292504 | val 3.654962 | tok/s 2625347.24 | norm 0.5261 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1779 | train 3.342236 | val 3.654962 | tok/s 2621443.81 | norm 0.4878 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1780 | train 3.391271 | val 3.654962 | tok/s 2626379.14 | norm 0.5953 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1781 | train 3.249122 | val 3.654962 | tok/s 2620641.36 | norm 0.5944 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1782 | train 3.266791 | val 3.654962 | tok/s 2622208.60 | norm 0.5799 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1783 | train 3.250468 | val 3.654962 | tok/s 2625437.48 | norm 0.5648 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1784 | train 3.298150 | val 3.654962 | tok/s 2629845.62 | norm 0.5247 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1785 | train 3.298893 | val 3.654962 | tok/s 2627160.54 | norm 0.4807 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1786 | train 3.346138 | val 3.654962 | tok/s 2631968.57 | norm 0.4670 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1787 | train 3.319700 | val 3.654962 | tok/s 2624196.02 | norm 0.5311 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1788 | train 3.308261 | val 3.654962 | tok/s 2628573.78 | norm 0.4849 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1789 | train 3.260578 | val 3.654962 | tok/s 2633518.94 | norm 0.5336 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1790 | train 3.414736 | val 3.654962 | tok/s 2552416.47 | norm 0.5570 | dt 205.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1791 | train 3.361079 | val 3.654962 | tok/s 2626956.28 | norm 0.5502 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1792 | train 3.202260 | val 3.654962 | tok/s 2628717.20 | norm 0.5605 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1793 | train 3.282817 | val 3.654962 | tok/s 2630882.26 | norm 0.5184 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1794 | train 3.237272 | val 3.654962 | tok/s 2625740.96 | norm 0.5671 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1795 | train 3.374764 | val 3.654962 | tok/s 2627797.58 | norm 0.5951 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1796 | train 3.359521 | val 3.654962 | tok/s 2630269.00 | norm 0.6342 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1797 | train 3.244250 | val 3.654962 | tok/s 2628074.50 | norm 0.5985 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1798 | train 3.112842 | val 3.654962 | tok/s 2631908.43 | norm 0.5363 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1799 | train 3.394446 | val 3.654962 | tok/s 2619905.53 | norm 0.6376 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1800 | train 3.201563 | val 3.654962 | tok/s 2604598.76 | norm 0.6163 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1801 | train 3.376327 | val 3.654962 | tok/s 2627666.80 | norm 0.5735 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1802 | train 3.362523 | val 3.654962 | tok/s 2626805.05 | norm 0.6013 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1803 | train 3.165904 | val 3.654962 | tok/s 2628694.58 | norm 0.5749 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1804 | train 3.271758 | val 3.654962 | tok/s 2627177.27 | norm 0.4740 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1805 | train 3.277307 | val 3.654962 | tok/s 2631087.14 | norm 0.4922 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1806 | train 3.255941 | val 3.654962 | tok/s 2629602.42 | norm 0.5361 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1807 | train 3.199339 | val 3.654962 | tok/s 2627855.24 | norm 0.5199 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1808 | train 3.223369 | val 3.654962 | tok/s 2628558.58 | norm 0.5365 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1809 | train 3.256004 | val 3.654962 | tok/s 2624907.48 | norm 0.6123 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1810 | train 3.292636 | val 3.654962 | tok/s 2626637.64 | norm 0.6426 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1811 | train 3.242272 | val 3.654962 | tok/s 2625694.89 | norm 0.6450 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1812 | train 3.302968 | val 3.654962 | tok/s 2627358.53 | norm 0.6673 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1813 | train 3.294612 | val 3.654962 | tok/s 2617524.08 | norm 0.5994 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1814 | train 3.233073 | val 3.654962 | tok/s 2625841.63 | norm 0.5622 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1815 | train 3.468102 | val 3.654962 | tok/s 2628633.53 | norm 0.6595 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1816 | train 3.249386 | val 3.654962 | tok/s 2629010.60 | norm 0.5510 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1817 | train 3.268769 | val 3.654962 | tok/s 2628799.01 | norm 0.5010 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1818 | train 3.225191 | val 3.654962 | tok/s 2628298.99 | norm 0.5243 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1819 | train 3.346389 | val 3.654962 | tok/s 2626583.72 | norm 0.4926 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1820 | train 3.268997 | val 3.654962 | tok/s 2625639.70 | norm 0.4919 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1821 | train 3.220451 | val 3.654962 | tok/s 2628379.21 | norm 0.5460 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1822 | train 3.245982 | val 3.654962 | tok/s 2627606.69 | norm 0.5491 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1823 | train 3.330033 | val 3.654962 | tok/s 2628849.07 | norm 0.5077 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1824 | train 3.328417 | val 3.654962 | tok/s 2622078.63 | norm 0.5218 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1825 | train 3.269313 | val 3.654962 | tok/s 2630563.94 | norm 0.5592 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1826 | train 3.283639 | val 3.654962 | tok/s 2631172.68 | norm 0.4829 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1827 | train 3.160972 | val 3.654962 | tok/s 2629102.53 | norm 0.5231 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1828 | train 3.365443 | val 3.654962 | tok/s 2631813.13 | norm 0.5296 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1829 | train 3.338644 | val 3.654962 | tok/s 2626185.61 | norm 0.4978 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1830 | train 3.219672 | val 3.654962 | tok/s 2626643.53 | norm 0.5755 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1831 | train 3.347743 | val 3.654962 | tok/s 2631383.22 | norm 0.5860 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1832 | train 3.258915 | val 3.654962 | tok/s 2624978.35 | norm 0.5738 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1833 | train 3.221633 | val 3.654962 | tok/s 2629707.58 | norm 0.5407 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1834 | train 3.341566 | val 3.654962 | tok/s 2625336.67 | norm 0.5184 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1835 | train 3.298315 | val 3.654962 | tok/s 2627826.12 | norm 0.5305 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1836 | train 3.283936 | val 3.654962 | tok/s 2631405.84 | norm 0.5911 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1837 | train 3.321958 | val 3.654962 | tok/s 2598537.26 | norm 0.5774 | dt 201.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1838 | train 3.314397 | val 3.654962 | tok/s 2630754.84 | norm 0.5795 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1839 | train 3.302859 | val 3.654962 | tok/s 2627990.34 | norm 0.5786 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1840 | train 3.367068 | val 3.654962 | tok/s 2625365.66 | norm 0.4862 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1841 | train 3.246991 | val 3.654962 | tok/s 2485102.27 | norm 0.5015 | dt 210.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1842 | train 3.330955 | val 3.654962 | tok/s 2602814.64 | norm 0.5353 | dt 201.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1843 | train 3.362274 | val 3.654962 | tok/s 2626738.30 | norm 0.5977 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1844 | train 3.222030 | val 3.654962 | tok/s 2626883.41 | norm 0.6028 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1845 | train 3.252418 | val 3.654962 | tok/s 2623579.21 | norm 0.5933 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1846 | train 3.333106 | val 3.654962 | tok/s 2549388.10 | norm 0.6097 | dt 205.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1847 | train 3.327051 | val 3.654962 | tok/s 2614899.94 | norm 0.5796 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1848 | train 3.208014 | val 3.654962 | tok/s 2628086.01 | norm 0.6350 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1849 | train 3.332729 | val 3.654962 | tok/s 2619721.34 | norm 0.6401 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1850 | train 3.377634 | val 3.654962 | tok/s 2628682.36 | norm 0.7014 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1851 | train 3.375467 | val 3.654962 | tok/s 2623019.03 | norm 0.5632 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1852 | train 3.186883 | val 3.654962 | tok/s 2622861.81 | norm 0.5521 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1853 | train 3.337413 | val 3.654962 | tok/s 2623630.88 | norm 0.5498 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1854 | train 3.325196 | val 3.654962 | tok/s 2624189.00 | norm 0.6067 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1855 | train 3.351019 | val 3.654962 | tok/s 2626171.89 | norm 0.5624 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1856 | train 3.238655 | val 3.654962 | tok/s 2627133.72 | norm 0.5536 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1857 | train 3.195476 | val 3.654962 | tok/s 2626652.59 | norm 0.6101 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1858 | train 3.315076 | val 3.654962 | tok/s 2628475.09 | norm 0.6267 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1859 | train 3.234021 | val 3.654962 | tok/s 2625975.07 | norm 0.6053 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1860 | train 3.301753 | val 3.654962 | tok/s 2630565.67 | norm 0.5179 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1861 | train 3.374020 | val 3.654962 | tok/s 2630288.18 | norm 0.5722 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1862 | train 3.336307 | val 3.654962 | tok/s 2446308.52 | norm 0.6107 | dt 214.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1863 | train 3.318312 | val 3.654962 | tok/s 2627837.19 | norm 0.5460 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1864 | train 3.255174 | val 3.654962 | tok/s 2624797.43 | norm 0.5276 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1865 | train 3.417754 | val 3.654962 | tok/s 2627720.92 | norm 0.4931 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1866 | train 3.308258 | val 3.654962 | tok/s 2477356.84 | norm 0.5123 | dt 211.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1867 | train 3.244952 | val 3.654962 | tok/s 2608416.07 | norm 0.4774 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1868 | train 3.492191 | val 3.654962 | tok/s 2625811.15 | norm 0.5601 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1869 | train 3.274379 | val 3.654962 | tok/s 2605293.43 | norm 0.5842 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1870 | train 3.225126 | val 3.654962 | tok/s 2626757.52 | norm 0.4864 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1871 | train 3.381772 | val 3.654962 | tok/s 2624221.45 | norm 0.5031 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1872 | train 3.330374 | val 3.654962 | tok/s 2624154.37 | norm 0.5456 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1873 | train 3.193003 | val 3.654962 | tok/s 2621390.18 | norm 0.5994 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1874 | train 3.219044 | val 3.654962 | tok/s 2625600.04 | norm 0.6041 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1875 | train 3.271055 | val 3.654962 | tok/s 2628219.95 | norm 0.5466 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1876 | train 3.347881 | val 3.654962 | tok/s 2624340.26 | norm 0.5166 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1877 | train 3.445379 | val 3.654962 | tok/s 2627033.64 | norm 0.5112 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1878 | train 3.447146 | val 3.654962 | tok/s 2631728.74 | norm 0.5257 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1879 | train 3.239705 | val 3.654962 | tok/s 2630810.59 | norm 0.5226 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1880 | train 3.301630 | val 3.654962 | tok/s 2631668.19 | norm 0.5458 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1881 | train 3.269819 | val 3.654962 | tok/s 2626611.54 | norm 0.5662 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1882 | train 3.165233 | val 3.654962 | tok/s 2627514.47 | norm 0.5764 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1883 | train 3.295373 | val 3.654962 | tok/s 2631400.77 | norm 0.5591 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1884 | train 3.213148 | val 3.654962 | tok/s 2626393.50 | norm 0.5330 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1885 | train 3.303582 | val 3.654962 | tok/s 2626064.11 | norm 0.5146 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1886 | train 3.223207 | val 3.654962 | tok/s 2626240.31 | norm 0.5338 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1887 | train 3.398300 | val 3.654962 | tok/s 2628414.20 | norm 0.5473 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1888 | train 3.144114 | val 3.654962 | tok/s 2629291.01 | norm 0.6045 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1889 | train 3.261333 | val 3.654962 | tok/s 2628380.17 | norm 0.5920 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1890 | train 3.239243 | val 3.654962 | tok/s 2624581.53 | norm 0.6126 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1891 | train 3.322121 | val 3.654962 | tok/s 2623003.69 | norm 0.5886 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1892 | train 3.247911 | val 3.654962 | tok/s 2627141.54 | norm 0.6153 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1893 | train 3.378056 | val 3.654962 | tok/s 2633433.08 | norm 0.6084 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1894 | train 3.211511 | val 3.654962 | tok/s 2625459.12 | norm 0.5630 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1895 | train 3.387166 | val 3.654962 | tok/s 2630550.53 | norm 0.5825 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1896 | train 3.297137 | val 3.654962 | tok/s 2629579.07 | norm 0.6049 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1897 | train 3.289710 | val 3.654962 | tok/s 2626659.76 | norm 0.5701 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1898 | train 3.288247 | val 3.654962 | tok/s 2628376.09 | norm 0.5592 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1899 | train 3.270048 | val 3.654962 | tok/s 2624087.59 | norm 0.5216 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1900 | train 3.212817 | val 3.654962 | tok/s 2625526.86 | norm 0.4985 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1901 | train 3.271648 | val 3.654962 | tok/s 2626978.33 | norm 0.5718 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1902 | train 3.525166 | val 3.654962 | tok/s 2626062.16 | norm 0.6116 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1903 | train 3.348289 | val 3.654962 | tok/s 2632084.68 | norm 0.6026 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1904 | train 3.265121 | val 3.654962 | tok/s 2629787.13 | norm 0.5784 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1905 | train 3.292575 | val 3.654962 | tok/s 2632924.87 | norm 0.5656 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1906 | train 3.256449 | val 3.654962 | tok/s 2626116.80 | norm 0.5233 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1907 | train 3.156877 | val 3.654962 | tok/s 2629676.78 | norm 0.5002 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1908 | train 3.269507 | val 3.654962 | tok/s 2628057.03 | norm 0.5206 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1909 | train 3.274234 | val 3.654962 | tok/s 2624913.10 | norm 0.5869 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1910 | train 3.290385 | val 3.654962 | tok/s 2629426.98 | norm 0.5683 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1911 | train 3.366461 | val 3.654962 | tok/s 2624373.31 | norm 0.5149 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1912 | train 3.246645 | val 3.654962 | tok/s 2626370.51 | norm 0.5075 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1913 | train 3.268464 | val 3.654962 | tok/s 2626129.57 | norm 0.5476 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1914 | train 3.247942 | val 3.654962 | tok/s 2625967.39 | norm 0.5606 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1915 | train 3.313413 | val 3.654962 | tok/s 2624139.52 | norm 0.5716 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1916 | train 3.157885 | val 3.654962 | tok/s 2624859.82 | norm 0.5792 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1917 | train 3.278382 | val 3.654962 | tok/s 2624823.20 | norm 0.5955 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1918 | train 3.218269 | val 3.654962 | tok/s 2630663.09 | norm 0.5964 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1919 | train 3.256487 | val 3.654962 | tok/s 2628773.12 | norm 0.5200 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1920 | train 3.325045 | val 3.654962 | tok/s 2629287.74 | norm 0.5031 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1921 | train 3.516371 | val 3.654962 | tok/s 2620729.51 | norm 0.6762 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1922 | train 3.134867 | val 3.654962 | tok/s 2625180.76 | norm 0.5671 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1923 | train 3.373647 | val 3.654962 | tok/s 2632717.84 | norm 0.5533 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1924 | train 3.281417 | val 3.654962 | tok/s 2626561.39 | norm 0.6213 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1925 | train 3.294174 | val 3.654962 | tok/s 2628889.27 | norm 0.5442 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1926 | train 3.176300 | val 3.654962 | tok/s 2627417.88 | norm 0.6018 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1927 | train 3.112108 | val 3.654962 | tok/s 2625645.09 | norm 0.6092 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1928 | train 3.127458 | val 3.654962 | tok/s 2631100.39 | norm 0.5591 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1929 | train 3.269734 | val 3.654962 | tok/s 2627418.11 | norm 0.5748 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1930 | train 3.163176 | val 3.654962 | tok/s 2628333.61 | norm 0.5820 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1931 | train 3.392087 | val 3.654962 | tok/s 2626793.51 | norm 0.5154 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1932 | train 3.340224 | val 3.654962 | tok/s 2627086.91 | norm 0.5048 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1933 | train 3.293380 | val 3.654962 | tok/s 2628525.16 | norm 0.5890 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1934 | train 3.275941 | val 3.654962 | tok/s 2626519.97 | norm 0.5649 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1935 | train 3.311420 | val 3.654962 | tok/s 2626510.20 | norm 0.5375 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1936 | train 3.255919 | val 3.654962 | tok/s 2621997.66 | norm 0.5682 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1937 | train 3.346402 | val 3.654962 | tok/s 2622966.75 | norm 0.5474 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1938 | train 3.220146 | val 3.654962 | tok/s 2627322.79 | norm 0.5150 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1939 | train 3.215311 | val 3.654962 | tok/s 2627099.27 | norm 0.5584 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1940 | train 3.352792 | val 3.654962 | tok/s 2623707.32 | norm 0.5645 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1941 | train 3.280632 | val 3.654962 | tok/s 2626285.13 | norm 0.5630 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1942 | train 3.199691 | val 3.654962 | tok/s 2630528.98 | norm 0.5632 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1943 | train 3.234193 | val 3.654962 | tok/s 2631780.86 | norm 0.5949 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1944 | train 3.078376 | val 3.654962 | tok/s 2629624.58 | norm 0.5367 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1945 | train 3.222423 | val 3.654962 | tok/s 2634014.28 | norm 0.4811 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1946 | train 3.307900 | val 3.654962 | tok/s 2624927.08 | norm 0.4845 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1947 | train 3.254454 | val 3.654962 | tok/s 2633787.47 | norm 0.5408 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1948 | train 3.209600 | val 3.654962 | tok/s 2633327.45 | norm 0.5424 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1949 | train 3.331328 | val 3.654962 | tok/s 2629848.65 | norm 0.5227 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1950 | train 3.329663 | val 3.654962 | tok/s 2632444.20 | norm 0.5903 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1951 | train 3.301470 | val 3.654962 | tok/s 2630801.56 | norm 0.5427 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1952 | train 3.285185 | val 3.654962 | tok/s 2630590.69 | norm 0.4763 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1953 | train 3.205240 | val 3.654962 | tok/s 2627463.55 | norm 0.5613 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1954 | train 3.325693 | val 3.654962 | tok/s 2624114.54 | norm 0.5695 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1955 | train 3.248080 | val 3.654962 | tok/s 2630119.70 | norm 0.5172 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1956 | train 3.222885 | val 3.654962 | tok/s 2626896.47 | norm 0.5064 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1957 | train 3.313085 | val 3.654962 | tok/s 2628086.77 | norm 0.6543 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1958 | train 3.179110 | val 3.654962 | tok/s 2617069.25 | norm 0.5424 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1959 | train 3.241354 | val 3.654962 | tok/s 2619061.67 | norm 0.4845 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1960 | train 3.269237 | val 3.654962 | tok/s 2628716.01 | norm 0.5645 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1961 | train 3.312569 | val 3.654962 | tok/s 2625421.10 | norm 0.6071 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1962 | train 3.348459 | val 3.654962 | tok/s 2631346.27 | norm 0.4872 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1963 | train 3.468630 | val 3.654962 | tok/s 2623408.87 | norm 0.5221 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1964 | train 3.280352 | val 3.654962 | tok/s 2623179.18 | norm 0.5669 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1965 | train 3.343074 | val 3.654962 | tok/s 2628775.48 | norm 0.5759 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1966 | train 3.611655 | val 3.654962 | tok/s 2628553.72 | norm 1.1879 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1967 | train 3.369038 | val 3.654962 | tok/s 2629306.74 | norm 0.6734 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1968 | train 3.257620 | val 3.654962 | tok/s 2626517.02 | norm 0.4824 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1969 | train 3.171487 | val 3.654962 | tok/s 2623339.60 | norm 0.5073 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1970 | train 3.174130 | val 3.654962 | tok/s 2628570.35 | norm 0.5057 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1971 | train 3.228178 | val 3.654962 | tok/s 2625223.91 | norm 0.5624 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1972 | train 3.334940 | val 3.654962 | tok/s 2632252.16 | norm 0.5664 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1973 | train 3.230377 | val 3.654962 | tok/s 2627836.35 | norm 0.4922 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1974 | train 3.332150 | val 3.654962 | tok/s 2627524.44 | norm 0.5206 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1975 | train 3.246772 | val 3.654962 | tok/s 2631352.33 | norm 0.5479 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1976 | train 3.191810 | val 3.654962 | tok/s 2627400.81 | norm 0.5938 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1977 | train 3.199049 | val 3.654962 | tok/s 2565796.65 | norm 0.5503 | dt 204.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1978 | train 3.218388 | val 3.654962 | tok/s 2625968.19 | norm 0.5075 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1979 | train 3.387357 | val 3.654962 | tok/s 2627646.82 | norm 0.5728 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1980 | train 3.268127 | val 3.654962 | tok/s 2627405.85 | norm 0.6152 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1981 | train 3.298557 | val 3.654962 | tok/s 2627343.79 | norm 0.5505 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1982 | train 3.277448 | val 3.654962 | tok/s 2630048.25 | norm 0.4990 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1983 | train 3.410875 | val 3.654962 | tok/s 2562758.34 | norm 0.5331 | dt 204.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1984 | train 3.271040 | val 3.654962 | tok/s 2587136.97 | norm 0.5512 | dt 202.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1985 | train 3.232995 | val 3.654962 | tok/s 2629733.02 | norm 0.5358 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1986 | train 3.090586 | val 3.654962 | tok/s 2627726.24 | norm 0.5680 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1987 | train 3.123342 | val 3.654962 | tok/s 2630097.33 | norm 0.5492 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1988 | train 3.296705 | val 3.654962 | tok/s 2622019.89 | norm 0.5964 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1989 | train 3.197576 | val 3.654962 | tok/s 2625838.62 | norm 0.5797 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1990 | train 3.181994 | val 3.654962 | tok/s 2489089.16 | norm 0.5451 | dt 210.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1991 | train 3.228989 | val 3.654962 | tok/s 2627911.11 | norm 0.5497 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1992 | train 3.119382 | val 3.654962 | tok/s 2625309.64 | norm 0.5068 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1993 | train 3.277922 | val 3.654962 | tok/s 2623306.29 | norm 0.4895 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1994 | train 3.223710 | val 3.654962 | tok/s 2624950.07 | norm 0.4853 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1995 | train 3.421286 | val 3.654962 | tok/s 2632131.75 | norm 0.5248 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1996 | train 3.306252 | val 3.654962 | tok/s 2626294.97 | norm 0.5292 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1997 | train 3.232771 | val 3.654962 | tok/s 2629006.79 | norm 0.5022 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1998 | train 3.224096 | val 3.654962 | tok/s 2627318.38 | norm 0.4941 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 1999 | train 3.234563 | val 3.654962 | tok/s 2535943.08 | norm 0.5262 | dt 206.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 2000 | val_loss 3.620728 step 2000 | train 3.312549 | val 3.620728 | tok/s 2614276.12 | norm 0.5369 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2001 | train 3.212042 | val 3.620728 | tok/s 2632176.21 | norm 0.5323 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2002 | train 3.206012 | val 3.620728 | tok/s 2626740.29 | norm 0.5865 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2003 | train 3.516300 | val 3.620728 | tok/s 2631838.47 | norm 0.5785 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2004 | train 3.229566 | val 3.620728 | tok/s 2629490.13 | norm 0.5250 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2005 | train 3.129775 | val 3.620728 | tok/s 2629139.07 | norm 0.5542 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2006 | train 3.222617 | val 3.620728 | tok/s 2634319.17 | norm 0.5292 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2007 | train 3.218764 | val 3.620728 | tok/s 2630711.91 | norm 0.5478 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2008 | train 3.255767 | val 3.620728 | tok/s 2628296.74 | norm 0.5025 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2009 | train 3.185010 | val 3.620728 | tok/s 2624859.27 | norm 0.5517 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2010 | train 3.159417 | val 3.620728 | tok/s 2628255.75 | norm 0.6088 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2011 | train 3.259422 | val 3.620728 | tok/s 2634291.35 | norm 0.5530 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2012 | train 3.255224 | val 3.620728 | tok/s 2632307.34 | norm 0.5627 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2013 | train 3.167475 | val 3.620728 | tok/s 2632434.86 | norm 0.5822 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2014 | train 3.221147 | val 3.620728 | tok/s 2626276.88 | norm 0.5655 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2015 | train 3.221975 | val 3.620728 | tok/s 2626255.86 | norm 0.5680 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2016 | train 3.275293 | val 3.620728 | tok/s 2629573.07 | norm 0.5371 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2017 | train 3.293730 | val 3.620728 | tok/s 2629583.25 | norm 0.5029 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2018 | train 3.229505 | val 3.620728 | tok/s 2631025.62 | norm 0.5105 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2019 | train 3.198327 | val 3.620728 | tok/s 2627686.19 | norm 0.5319 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2020 | train 3.261490 | val 3.620728 | tok/s 2626357.72 | norm 0.5366 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2021 | train 3.146617 | val 3.620728 | tok/s 2634992.21 | norm 0.5438 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2022 | train 3.221428 | val 3.620728 | tok/s 2629667.26 | norm 0.5577 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2023 | train 3.281921 | val 3.620728 | tok/s 2631979.99 | norm 0.5340 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2024 | train 3.267542 | val 3.620728 | tok/s 2625110.53 | norm 0.5342 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2025 | train 3.200232 | val 3.620728 | tok/s 2624928.99 | norm 0.6060 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2026 | train 3.164860 | val 3.620728 | tok/s 2628933.62 | norm 0.5734 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2027 | train 3.259825 | val 3.620728 | tok/s 2627820.62 | norm 0.5353 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2028 | train 3.195088 | val 3.620728 | tok/s 2630360.42 | norm 0.5396 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2029 | train 3.331293 | val 3.620728 | tok/s 2623716.77 | norm 0.5530 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2030 | train 3.246584 | val 3.620728 | tok/s 2625033.57 | norm 0.5109 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2031 | train 3.281148 | val 3.620728 | tok/s 2627229.95 | norm 0.4800 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2032 | train 3.175710 | val 3.620728 | tok/s 2629294.03 | norm 0.5629 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2033 | train 3.167932 | val 3.620728 | tok/s 2627691.07 | norm 0.5918 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2034 | train 3.302206 | val 3.620728 | tok/s 2626262.53 | norm 0.5719 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2035 | train 3.251951 | val 3.620728 | tok/s 2622612.85 | norm 0.5498 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2036 | train 3.206537 | val 3.620728 | tok/s 2629585.83 | norm 0.5494 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2037 | train 3.282496 | val 3.620728 | tok/s 2586026.31 | norm 0.5270 | dt 202.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2038 | train 3.225467 | val 3.620728 | tok/s 2633239.59 | norm 0.5475 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2039 | train 3.289065 | val 3.620728 | tok/s 2626088.32 | norm 0.5093 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2040 | train 3.109251 | val 3.620728 | tok/s 2628919.22 | norm 0.5648 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2041 | train 3.167370 | val 3.620728 | tok/s 2627893.59 | norm 0.5557 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2042 | train 3.197866 | val 3.620728 | tok/s 2626467.91 | norm 0.5440 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2043 | train 3.287529 | val 3.620728 | tok/s 2629263.86 | norm 0.5636 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2044 | train 3.112483 | val 3.620728 | tok/s 2620171.39 | norm 0.5745 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2045 | train 3.182410 | val 3.620728 | tok/s 2629773.92 | norm 0.5269 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2046 | train 3.126279 | val 3.620728 | tok/s 2632038.72 | norm 0.5108 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2047 | train 3.340920 | val 3.620728 | tok/s 2625465.02 | norm 0.7556 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2048 | train 3.295593 | val 3.620728 | tok/s 2580327.98 | norm 0.5427 | dt 203.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2049 | train 3.249387 | val 3.620728 | tok/s 2626854.32 | norm 0.5357 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2050 | train 3.133403 | val 3.620728 | tok/s 2626740.82 | norm 0.5432 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2051 | train 3.317620 | val 3.620728 | tok/s 2632188.32 | norm 0.5351 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2052 | train 3.294816 | val 3.620728 | tok/s 2624690.13 | norm 0.6762 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2053 | train 3.265279 | val 3.620728 | tok/s 2627412.77 | norm 0.6027 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2054 | train 3.100397 | val 3.620728 | tok/s 2629552.56 | norm 0.4992 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2055 | train 3.261255 | val 3.620728 | tok/s 2627281.58 | norm 0.6021 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2056 | train 3.203292 | val 3.620728 | tok/s 2631791.67 | norm 0.6057 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2057 | train 3.301092 | val 3.620728 | tok/s 2624985.24 | norm 0.6119 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2058 | train 3.334091 | val 3.620728 | tok/s 2629760.19 | norm 0.5992 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2059 | train 3.392899 | val 3.620728 | tok/s 2626917.19 | norm 0.5935 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2060 | train 3.252092 | val 3.620728 | tok/s 2627537.09 | norm 0.5759 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2061 | train 3.215085 | val 3.620728 | tok/s 2629870.72 | norm 0.5653 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2062 | train 3.085905 | val 3.620728 | tok/s 2629219.73 | norm 0.5975 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2063 | train 3.210980 | val 3.620728 | tok/s 2633395.02 | norm 0.5558 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2064 | train 3.188328 | val 3.620728 | tok/s 2631477.28 | norm 0.4903 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2065 | train 3.175494 | val 3.620728 | tok/s 2632203.20 | norm 0.4841 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2066 | train 3.174617 | val 3.620728 | tok/s 2634589.95 | norm 0.5540 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2067 | train 3.231351 | val 3.620728 | tok/s 2625247.45 | norm 0.5541 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2068 | train 3.233128 | val 3.620728 | tok/s 2629337.90 | norm 0.4789 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2069 | train 3.221835 | val 3.620728 | tok/s 2629383.89 | norm 0.4693 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2070 | train 3.280904 | val 3.620728 | tok/s 2630482.06 | norm 0.5098 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2071 | train 3.210613 | val 3.620728 | tok/s 2627923.57 | norm 0.5462 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2072 | train 3.282899 | val 3.620728 | tok/s 2629322.54 | norm 0.6152 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2073 | train 3.316008 | val 3.620728 | tok/s 2629802.68 | norm 0.5713 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2074 | train 3.314102 | val 3.620728 | tok/s 2625549.60 | norm 0.6061 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2075 | train 3.190904 | val 3.620728 | tok/s 2628769.47 | norm 0.5258 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2076 | train 3.287731 | val 3.620728 | tok/s 2628576.97 | norm 0.5407 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2077 | train 3.202505 | val 3.620728 | tok/s 2622134.97 | norm 0.5410 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2078 | train 3.294140 | val 3.620728 | tok/s 2627287.90 | norm 0.6006 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2079 | train 3.272179 | val 3.620728 | tok/s 2624432.34 | norm 0.6103 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2080 | train 3.235205 | val 3.620728 | tok/s 2627506.72 | norm 0.5853 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2081 | train 3.221149 | val 3.620728 | tok/s 2631412.75 | norm 0.5958 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2082 | train 3.144462 | val 3.620728 | tok/s 2624290.99 | norm 0.5797 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2083 | train 3.207816 | val 3.620728 | tok/s 2632712.47 | norm 0.5272 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2084 | train 3.186422 | val 3.620728 | tok/s 2623351.73 | norm 0.5364 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2085 | train 3.164980 | val 3.620728 | tok/s 2630010.56 | norm 0.5440 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2086 | train 3.299769 | val 3.620728 | tok/s 2627558.59 | norm 0.5144 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2087 | train 3.328132 | val 3.620728 | tok/s 2624972.31 | norm 0.4804 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2088 | train 3.163977 | val 3.620728 | tok/s 2629167.77 | norm 0.5046 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2089 | train 3.219280 | val 3.620728 | tok/s 2626847.81 | norm 0.5192 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2090 | train 3.229749 | val 3.620728 | tok/s 2633324.05 | norm 0.5373 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2091 | train 3.258133 | val 3.620728 | tok/s 2627908.62 | norm 0.5229 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2092 | train 3.250386 | val 3.620728 | tok/s 2625406.41 | norm 0.4888 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2093 | train 3.155821 | val 3.620728 | tok/s 2628377.99 | norm 0.5346 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2094 | train 3.148000 | val 3.620728 | tok/s 2626244.74 | norm 0.5490 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2095 | train 3.300946 | val 3.620728 | tok/s 2632679.78 | norm 0.5620 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2096 | train 3.142802 | val 3.620728 | tok/s 2623852.00 | norm 0.5398 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2097 | train 3.243751 | val 3.620728 | tok/s 2621354.75 | norm 0.4888 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2098 | train 3.273345 | val 3.620728 | tok/s 2625309.17 | norm 0.5216 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2099 | train 3.330254 | val 3.620728 | tok/s 2623990.19 | norm 0.5593 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2100 | train 3.423284 | val 3.620728 | tok/s 2628170.38 | norm 0.5915 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2101 | train 3.293799 | val 3.620728 | tok/s 2625037.06 | norm 0.5638 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2102 | train 3.255440 | val 3.620728 | tok/s 2622466.34 | norm 0.5479 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2103 | train 3.207914 | val 3.620728 | tok/s 2625946.60 | norm 0.6083 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2104 | train 3.286347 | val 3.620728 | tok/s 2619068.05 | norm 0.5898 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2105 | train 3.172127 | val 3.620728 | tok/s 2631156.39 | norm 0.5794 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2106 | train 3.412752 | val 3.620728 | tok/s 2627475.26 | norm 0.5930 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2107 | train 3.245040 | val 3.620728 | tok/s 2625072.22 | norm 0.5827 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2108 | train 3.190742 | val 3.620728 | tok/s 2627480.13 | norm 0.5344 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2109 | train 3.289013 | val 3.620728 | tok/s 2628566.12 | norm 0.6155 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2110 | train 3.388207 | val 3.620728 | tok/s 2630023.15 | norm 0.5760 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2111 | train 3.294665 | val 3.620728 | tok/s 2625156.25 | norm 0.4453 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2112 | train 3.298959 | val 3.620728 | tok/s 2623299.29 | norm 0.4918 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2113 | train 3.290056 | val 3.620728 | tok/s 2629257.99 | norm 0.5244 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2114 | train 3.265532 | val 3.620728 | tok/s 2629686.45 | norm 0.5630 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2115 | train 3.199324 | val 3.620728 | tok/s 2627966.66 | norm 0.5516 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2116 | train 3.090359 | val 3.620728 | tok/s 2628024.13 | norm 0.5442 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2117 | train 3.344863 | val 3.620728 | tok/s 2626232.91 | norm 0.5361 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2118 | train 3.207701 | val 3.620728 | tok/s 2633127.49 | norm 0.4962 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2119 | train 3.296192 | val 3.620728 | tok/s 2626076.80 | norm 0.5307 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2120 | train 3.313905 | val 3.620728 | tok/s 2630666.47 | norm 0.5650 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2121 | train 3.305513 | val 3.620728 | tok/s 2627745.43 | norm 0.6007 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2122 | train 3.166999 | val 3.620728 | tok/s 2625888.19 | norm 0.6296 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2123 | train 3.430582 | val 3.620728 | tok/s 2628579.91 | norm 0.6144 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2124 | train 3.250782 | val 3.620728 | tok/s 2626566.32 | norm 0.5767 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2125 | train 3.459276 | val 3.620728 | tok/s 2632309.65 | norm 0.7214 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2126 | train 3.129043 | val 3.620728 | tok/s 2625873.96 | norm 0.5167 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2127 | train 3.203911 | val 3.620728 | tok/s 2626034.15 | norm 0.5127 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2128 | train 3.264787 | val 3.620728 | tok/s 2629980.18 | norm 0.5784 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2129 | train 3.242608 | val 3.620728 | tok/s 2625880.18 | norm 0.5313 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2130 | train 3.152764 | val 3.620728 | tok/s 2627795.26 | norm 0.4937 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2131 | train 3.147788 | val 3.620728 | tok/s 2621859.23 | norm 0.4886 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2132 | train 3.282506 | val 3.620728 | tok/s 2627840.51 | norm 0.5491 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2133 | train 3.189014 | val 3.620728 | tok/s 2632858.48 | norm 0.6472 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2134 | train 3.217177 | val 3.620728 | tok/s 2628569.38 | norm 0.4476 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2135 | train 3.260291 | val 3.620728 | tok/s 2632845.28 | norm 0.4836 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2136 | train 3.275281 | val 3.620728 | tok/s 2627080.08 | norm 0.5242 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2137 | train 3.286508 | val 3.620728 | tok/s 2622628.01 | norm 0.5052 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2138 | train 3.222801 | val 3.620728 | tok/s 2633199.87 | norm 0.5490 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2139 | train 3.290890 | val 3.620728 | tok/s 2626947.28 | norm 0.5063 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2140 | train 3.208385 | val 3.620728 | tok/s 2628243.87 | norm 0.5631 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2141 | train 3.518165 | val 3.620728 | tok/s 2618451.77 | norm 0.5663 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2142 | train 3.218405 | val 3.620728 | tok/s 2623503.03 | norm 0.5209 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2143 | train 3.247245 | val 3.620728 | tok/s 2634533.05 | norm 0.4956 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2144 | train 3.190908 | val 3.620728 | tok/s 2626094.53 | norm 0.5246 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2145 | train 3.228680 | val 3.620728 | tok/s 2629105.71 | norm 0.5376 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2146 | train 3.353341 | val 3.620728 | tok/s 2620614.43 | norm 0.5190 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2147 | train 3.213673 | val 3.620728 | tok/s 2623947.33 | norm 0.4818 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2148 | train 3.151728 | val 3.620728 | tok/s 2632188.78 | norm 0.5828 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2149 | train 3.157487 | val 3.620728 | tok/s 2630401.83 | norm 0.6345 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2150 | train 3.065436 | val 3.620728 | tok/s 2631442.57 | norm 0.5990 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2151 | train 3.185179 | val 3.620728 | tok/s 2620312.37 | norm 0.5598 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2152 | train 3.263755 | val 3.620728 | tok/s 2627623.84 | norm 0.5437 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2153 | train 3.180854 | val 3.620728 | tok/s 2630298.17 | norm 0.5563 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2154 | train 3.216424 | val 3.620728 | tok/s 2624992.26 | norm 0.5603 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2155 | train 3.292382 | val 3.620728 | tok/s 2631939.57 | norm 0.5356 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2156 | train 3.174515 | val 3.620728 | tok/s 2624155.83 | norm 0.5106 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2157 | train 3.317383 | val 3.620728 | tok/s 2627609.05 | norm 0.5443 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2158 | train 3.228740 | val 3.620728 | tok/s 2631399.66 | norm 0.5194 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2159 | train 3.496353 | val 3.620728 | tok/s 2631396.85 | norm 0.5652 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2160 | train 3.112977 | val 3.620728 | tok/s 2628896.98 | norm 0.5993 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2161 | train 3.209628 | val 3.620728 | tok/s 2621215.61 | norm 0.5320 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2162 | train 3.190434 | val 3.620728 | tok/s 2627956.64 | norm 0.5381 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2163 | train 3.067317 | val 3.620728 | tok/s 2633783.16 | norm 0.4876 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2164 | train 3.230978 | val 3.620728 | tok/s 2627157.80 | norm 0.5081 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2165 | train 3.176891 | val 3.620728 | tok/s 2630942.66 | norm 0.5351 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2166 | train 3.397988 | val 3.620728 | tok/s 2624543.51 | norm 0.5594 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2167 | train 3.260611 | val 3.620728 | tok/s 2625423.54 | norm 0.5535 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2168 | train 3.171117 | val 3.620728 | tok/s 2629466.68 | norm 0.6690 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2169 | train 3.152908 | val 3.620728 | tok/s 2628042.15 | norm 0.6216 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2170 | train 3.205159 | val 3.620728 | tok/s 2631917.41 | norm 0.5185 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2171 | train 3.246243 | val 3.620728 | tok/s 2624358.25 | norm 0.5002 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2172 | train 3.363405 | val 3.620728 | tok/s 2603576.06 | norm 0.5835 | dt 201.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2173 | train 3.072324 | val 3.620728 | tok/s 2627709.98 | norm 0.5387 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2174 | train 3.312803 | val 3.620728 | tok/s 2627359.32 | norm 0.5483 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2175 | train 3.234453 | val 3.620728 | tok/s 2627413.38 | norm 0.5320 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2176 | train 3.163364 | val 3.620728 | tok/s 2622671.83 | norm 0.5253 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2177 | train 3.326624 | val 3.620728 | tok/s 2626839.39 | norm 0.5452 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2178 | train 3.243560 | val 3.620728 | tok/s 2630772.60 | norm 0.5492 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2179 | train 3.274907 | val 3.620728 | tok/s 2628478.90 | norm 0.5191 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2180 | train 3.189923 | val 3.620728 | tok/s 2629265.71 | norm 0.6078 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2181 | train 3.343890 | val 3.620728 | tok/s 2622669.14 | norm 0.5879 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2182 | train 3.208571 | val 3.620728 | tok/s 2627206.76 | norm 0.5764 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2183 | train 3.218507 | val 3.620728 | tok/s 2628551.35 | norm 0.5533 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2184 | train 3.204544 | val 3.620728 | tok/s 2630380.19 | norm 0.5748 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2185 | train 3.207716 | val 3.620728 | tok/s 2633197.16 | norm 0.5412 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2186 | train 3.223170 | val 3.620728 | tok/s 2621189.15 | norm 0.5737 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2187 | train 3.206253 | val 3.620728 | tok/s 2627412.83 | norm 0.5440 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2188 | train 3.223919 | val 3.620728 | tok/s 2630610.15 | norm 0.5039 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2189 | train 3.171825 | val 3.620728 | tok/s 2627140.34 | norm 0.4992 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2190 | train 3.128150 | val 3.620728 | tok/s 2497934.00 | norm 0.5423 | dt 209.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2191 | train 3.269952 | val 3.620728 | tok/s 2622038.98 | norm 0.5600 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2192 | train 3.328997 | val 3.620728 | tok/s 2627162.42 | norm 0.5619 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2193 | train 3.288659 | val 3.620728 | tok/s 2629709.70 | norm 0.5289 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2194 | train 3.327794 | val 3.620728 | tok/s 2625531.75 | norm 0.6530 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2195 | train 3.312379 | val 3.620728 | tok/s 2629843.74 | norm 0.5727 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2196 | train 3.211255 | val 3.620728 | tok/s 2627014.31 | norm 0.5174 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2197 | train 3.218646 | val 3.620728 | tok/s 2628179.74 | norm 0.5798 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2198 | train 3.313962 | val 3.620728 | tok/s 2633862.59 | norm 0.5641 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2199 | train 3.361457 | val 3.620728 | tok/s 2629387.31 | norm 0.5944 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2200 | train 3.309319 | val 3.620728 | tok/s 2631923.94 | norm 0.5611 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2201 | train 3.224721 | val 3.620728 | tok/s 2624223.08 | norm 0.5047 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2202 | train 3.298678 | val 3.620728 | tok/s 2626288.34 | norm 0.4858 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2203 | train 3.196774 | val 3.620728 | tok/s 2629375.09 | norm 0.5012 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2204 | train 3.228423 | val 3.620728 | tok/s 2630622.74 | norm 0.5283 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2205 | train 3.302966 | val 3.620728 | tok/s 2632127.83 | norm 0.6009 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2206 | train 3.251004 | val 3.620728 | tok/s 2630187.08 | norm 0.5237 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2207 | train 3.181222 | val 3.620728 | tok/s 2631813.50 | norm 0.5243 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2208 | train 3.290175 | val 3.620728 | tok/s 2634046.36 | norm 0.5497 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2209 | train 3.294728 | val 3.620728 | tok/s 2631253.50 | norm 0.4947 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2210 | train 3.239243 | val 3.620728 | tok/s 2631571.42 | norm 0.4988 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2211 | train 3.216759 | val 3.620728 | tok/s 2621807.91 | norm 0.5120 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2212 | train 3.165828 | val 3.620728 | tok/s 2625256.14 | norm 0.5039 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2213 | train 3.247750 | val 3.620728 | tok/s 2632856.35 | norm 0.5697 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2214 | train 3.330825 | val 3.620728 | tok/s 2624142.99 | norm 0.5475 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2215 | train 3.226709 | val 3.620728 | tok/s 2628189.47 | norm 0.4975 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2216 | train 3.292837 | val 3.620728 | tok/s 2625475.79 | norm 0.5414 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2217 | train 3.217633 | val 3.620728 | tok/s 2625319.57 | norm 0.5298 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2218 | train 3.204602 | val 3.620728 | tok/s 2634186.34 | norm 0.5088 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2219 | train 3.082589 | val 3.620728 | tok/s 2627438.90 | norm 0.4882 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2220 | train 3.161396 | val 3.620728 | tok/s 2631998.49 | norm 0.5076 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2221 | train 3.297397 | val 3.620728 | tok/s 2623134.64 | norm 0.5141 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2222 | train 3.265457 | val 3.620728 | tok/s 2627337.22 | norm 0.5114 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2223 | train 3.292093 | val 3.620728 | tok/s 2570973.85 | norm 0.5115 | dt 203.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2224 | train 3.253334 | val 3.620728 | tok/s 2630347.65 | norm 0.5241 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2225 | train 3.191353 | val 3.620728 | tok/s 2628181.48 | norm 0.5851 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2226 | train 3.178316 | val 3.620728 | tok/s 2628603.88 | norm 0.5293 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2227 | train 3.247638 | val 3.620728 | tok/s 2628110.24 | norm 0.5385 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2228 | train 3.310148 | val 3.620728 | tok/s 2633120.88 | norm 0.5323 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2229 | train 3.256069 | val 3.620728 | tok/s 2627491.48 | norm 0.4897 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2230 | train 3.173748 | val 3.620728 | tok/s 2604411.82 | norm 0.4898 | dt 201.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2231 | train 3.314075 | val 3.620728 | tok/s 2623251.39 | norm 0.4822 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2232 | train 3.123167 | val 3.620728 | tok/s 2591826.48 | norm 0.4953 | dt 202.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2233 | train 3.149350 | val 3.620728 | tok/s 2628161.31 | norm 0.5364 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2234 | train 3.356621 | val 3.620728 | tok/s 2628472.35 | norm 0.5333 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2235 | train 3.183197 | val 3.620728 | tok/s 2627879.90 | norm 0.5275 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2236 | train 3.221895 | val 3.620728 | tok/s 2622163.32 | norm 0.5396 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2237 | train 3.079222 | val 3.620728 | tok/s 2628748.39 | norm 0.5804 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2238 | train 3.311044 | val 3.620728 | tok/s 2628078.18 | norm 0.4326 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2239 | train 3.238242 | val 3.620728 | tok/s 2626653.66 | norm 0.4629 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2240 | train 3.131065 | val 3.620728 | tok/s 2628204.92 | norm 0.5639 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2241 | train 3.137866 | val 3.620728 | tok/s 2631210.25 | norm 0.5490 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2242 | train 3.193470 | val 3.620728 | tok/s 2628851.22 | norm 0.5350 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2243 | train 3.252509 | val 3.620728 | tok/s 2627529.68 | norm 0.5662 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2244 | train 3.250856 | val 3.620728 | tok/s 2626386.73 | norm 0.5918 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2245 | train 3.243178 | val 3.620728 | tok/s 2628520.88 | norm 0.5473 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2246 | train 3.163495 | val 3.620728 | tok/s 2628661.02 | norm 0.5370 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2247 | train 3.227926 | val 3.620728 | tok/s 2630296.47 | norm 0.5413 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2248 | train 3.240858 | val 3.620728 | tok/s 2635262.08 | norm 0.5714 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2249 | train 3.193353 | val 3.620728 | tok/s 2628257.40 | norm 0.5604 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 2250 | val_loss 3.585815 step 2250 | train 3.241248 | val 3.585815 | tok/s 2614923.53 | norm 0.4818 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2251 | train 3.403929 | val 3.585815 | tok/s 2628848.91 | norm 0.4856 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2252 | train 3.181127 | val 3.585815 | tok/s 2622232.03 | norm 0.5035 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2253 | train 3.295145 | val 3.585815 | tok/s 2623536.76 | norm 0.4828 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2254 | train 3.209895 | val 3.585815 | tok/s 2628581.95 | norm 0.4712 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2255 | train 3.214752 | val 3.585815 | tok/s 2627535.96 | norm 0.4675 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2256 | train 3.468055 | val 3.585815 | tok/s 2630667.38 | norm 0.5273 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2257 | train 3.269280 | val 3.585815 | tok/s 2623452.88 | norm 0.4863 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2258 | train 3.385407 | val 3.585815 | tok/s 2627467.67 | norm 0.4727 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2259 | train 3.201329 | val 3.585815 | tok/s 2629940.94 | norm 0.4626 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2260 | train 3.191061 | val 3.585815 | tok/s 2628954.16 | norm 0.5035 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2261 | train 3.174221 | val 3.585815 | tok/s 2625337.94 | norm 0.5287 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2262 | train 3.180861 | val 3.585815 | tok/s 2627349.72 | norm 0.6144 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2263 | train 3.227195 | val 3.585815 | tok/s 2629784.81 | norm 0.5726 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2264 | train 3.158900 | val 3.585815 | tok/s 2627093.44 | norm 0.5594 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2265 | train 3.337779 | val 3.585815 | tok/s 2626641.00 | norm 0.5671 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2266 | train 3.239394 | val 3.585815 | tok/s 2634315.30 | norm 0.6253 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2267 | train 3.151310 | val 3.585815 | tok/s 2628221.86 | norm 0.5950 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2268 | train 3.234263 | val 3.585815 | tok/s 2629575.59 | norm 0.5460 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2269 | train 3.336966 | val 3.585815 | tok/s 2625208.70 | norm 0.5475 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2270 | train 3.216608 | val 3.585815 | tok/s 2629389.91 | norm 0.5919 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2271 | train 3.182597 | val 3.585815 | tok/s 2625168.37 | norm 0.5340 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2272 | train 3.212960 | val 3.585815 | tok/s 2628086.21 | norm 0.5331 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2273 | train 3.281281 | val 3.585815 | tok/s 2628426.62 | norm 0.5510 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2274 | train 3.234545 | val 3.585815 | tok/s 2631326.95 | norm 0.5336 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2275 | train 3.296688 | val 3.585815 | tok/s 2632413.52 | norm 0.5822 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2276 | train 3.126920 | val 3.585815 | tok/s 2632223.80 | norm 0.5522 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2277 | train 3.011275 | val 3.585815 | tok/s 2626012.04 | norm 0.5277 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2278 | train 3.357544 | val 3.585815 | tok/s 2626755.41 | norm 0.5449 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2279 | train 3.248695 | val 3.585815 | tok/s 2631638.31 | norm 0.5183 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2280 | train 3.103000 | val 3.585815 | tok/s 2627722.85 | norm 0.5146 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2281 | train 3.237022 | val 3.585815 | tok/s 2632220.76 | norm 0.5923 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2282 | train 3.148269 | val 3.585815 | tok/s 2626535.44 | norm 0.5532 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2283 | train 3.362156 | val 3.585815 | tok/s 2624662.64 | norm 0.5838 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2284 | train 3.227123 | val 3.585815 | tok/s 2630157.25 | norm 0.4934 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2285 | train 3.267609 | val 3.585815 | tok/s 2629160.48 | norm 0.4617 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2286 | train 3.346743 | val 3.585815 | tok/s 2630398.55 | norm 0.5583 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2287 | train 3.268585 | val 3.585815 | tok/s 2629360.19 | norm 0.5631 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2288 | train 3.129298 | val 3.585815 | tok/s 2629606.77 | norm 0.4805 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2289 | train 3.233400 | val 3.585815 | tok/s 2629014.36 | norm 0.5084 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2290 | train 3.244148 | val 3.585815 | tok/s 2626016.65 | norm 0.5386 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2291 | train 3.224023 | val 3.585815 | tok/s 2631208.73 | norm 0.5794 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2292 | train 3.118388 | val 3.585815 | tok/s 2627600.96 | norm 0.4989 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2293 | train 3.200025 | val 3.585815 | tok/s 2628541.16 | norm 0.4960 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2294 | train 3.211753 | val 3.585815 | tok/s 2632133.44 | norm 0.5658 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2295 | train 3.296198 | val 3.585815 | tok/s 2625957.47 | norm 0.5320 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2296 | train 3.234968 | val 3.585815 | tok/s 2631714.94 | norm 0.5289 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2297 | train 3.115410 | val 3.585815 | tok/s 2626111.68 | norm 0.5760 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2298 | train 3.185829 | val 3.585815 | tok/s 2621242.88 | norm 0.5676 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2299 | train 3.158730 | val 3.585815 | tok/s 2629266.10 | norm 0.5567 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2300 | train 3.162380 | val 3.585815 | tok/s 2620881.50 | norm 0.5211 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2301 | train 3.354255 | val 3.585815 | tok/s 2631762.87 | norm 0.6031 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2302 | train 3.267715 | val 3.585815 | tok/s 2625692.20 | norm 0.6003 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2303 | train 3.400663 | val 3.585815 | tok/s 2626808.14 | norm 0.5455 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2304 | train 3.179188 | val 3.585815 | tok/s 2629007.85 | norm 0.5313 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2305 | train 3.222172 | val 3.585815 | tok/s 2622990.17 | norm 0.5034 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2306 | train 3.248567 | val 3.585815 | tok/s 2622121.80 | norm 0.4654 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2307 | train 3.292805 | val 3.585815 | tok/s 2624535.23 | norm 0.5385 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2308 | train 3.212009 | val 3.585815 | tok/s 2628364.89 | norm 0.5625 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2309 | train 3.300099 | val 3.585815 | tok/s 2625961.47 | norm 0.5720 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2310 | train 3.181388 | val 3.585815 | tok/s 2622494.39 | norm 0.5925 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2311 | train 3.248947 | val 3.585815 | tok/s 2629351.31 | norm 0.5356 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2312 | train 3.270319 | val 3.585815 | tok/s 2627146.57 | norm 0.5427 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2313 | train 3.230201 | val 3.585815 | tok/s 2625961.22 | norm 0.5318 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2314 | train 3.202671 | val 3.585815 | tok/s 2625038.07 | norm 0.5005 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2315 | train 3.156364 | val 3.585815 | tok/s 2625052.06 | norm 0.5098 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2316 | train 3.136417 | val 3.585815 | tok/s 2630124.68 | norm 0.5612 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2317 | train 3.295174 | val 3.585815 | tok/s 2630188.56 | norm 0.5751 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2318 | train 3.211076 | val 3.585815 | tok/s 2633140.54 | norm 0.5348 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2319 | train 3.136907 | val 3.585815 | tok/s 2631112.71 | norm 0.5876 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2320 | train 3.229027 | val 3.585815 | tok/s 2626645.20 | norm 0.5936 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2321 | train 3.089487 | val 3.585815 | tok/s 2627696.77 | norm 0.4819 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2322 | train 3.293232 | val 3.585815 | tok/s 2626598.28 | norm 0.5113 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2323 | train 3.212174 | val 3.585815 | tok/s 2628560.88 | norm 0.5255 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2324 | train 3.203339 | val 3.585815 | tok/s 2627825.80 | norm 0.4878 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2325 | train 3.297564 | val 3.585815 | tok/s 2621921.28 | norm 0.5377 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2326 | train 3.182207 | val 3.585815 | tok/s 2627343.79 | norm 0.5753 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2327 | train 3.142213 | val 3.585815 | tok/s 2625312.73 | norm 0.5628 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2328 | train 3.273941 | val 3.585815 | tok/s 2626591.85 | norm 0.5507 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2329 | train 3.259255 | val 3.585815 | tok/s 2628581.63 | norm 0.5383 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2330 | train 3.276742 | val 3.585815 | tok/s 2623596.98 | norm 0.5359 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2331 | train 3.282535 | val 3.585815 | tok/s 2628832.34 | norm 0.5440 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2332 | train 3.237733 | val 3.585815 | tok/s 2621130.72 | norm 0.5882 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2333 | train 3.333373 | val 3.585815 | tok/s 2630506.51 | norm 0.5536 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2334 | train 3.283062 | val 3.585815 | tok/s 2622835.47 | norm 0.5647 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2335 | train 3.289177 | val 3.585815 | tok/s 2628863.98 | norm 0.5277 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2336 | train 3.337441 | val 3.585815 | tok/s 2607717.97 | norm 0.4780 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2337 | train 3.177475 | val 3.585815 | tok/s 2624255.34 | norm 0.5396 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2338 | train 3.220627 | val 3.585815 | tok/s 2629513.97 | norm 0.5884 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2339 | train 3.254232 | val 3.585815 | tok/s 2620582.10 | norm 0.5530 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2340 | train 3.225518 | val 3.585815 | tok/s 2609166.82 | norm 0.5824 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2341 | train 3.350758 | val 3.585815 | tok/s 2599089.60 | norm 0.5787 | dt 201.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2342 | train 3.198514 | val 3.585815 | tok/s 2628263.48 | norm 0.5245 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2343 | train 3.147591 | val 3.585815 | tok/s 2628152.01 | norm 0.5237 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2344 | train 3.211713 | val 3.585815 | tok/s 2629923.24 | norm 0.5279 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2345 | train 3.171194 | val 3.585815 | tok/s 2629519.88 | norm 0.4931 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2346 | train 3.268867 | val 3.585815 | tok/s 2630235.75 | norm 0.5210 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2347 | train 3.109249 | val 3.585815 | tok/s 2628852.92 | norm 0.4896 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2348 | train 3.057881 | val 3.585815 | tok/s 2629645.03 | norm 0.5284 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2349 | train 3.324170 | val 3.585815 | tok/s 2624501.09 | norm 0.5802 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2350 | train 3.163913 | val 3.585815 | tok/s 2626147.34 | norm 0.4904 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2351 | train 3.136096 | val 3.585815 | tok/s 2630888.76 | norm 0.4674 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2352 | train 3.226214 | val 3.585815 | tok/s 2631180.92 | norm 0.4949 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2353 | train 3.292067 | val 3.585815 | tok/s 2487806.55 | norm 0.5399 | dt 210.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2354 | train 3.193792 | val 3.585815 | tok/s 2621116.82 | norm 0.5752 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2355 | train 3.192128 | val 3.585815 | tok/s 2626782.06 | norm 0.6024 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2356 | train 3.213907 | val 3.585815 | tok/s 2630057.57 | norm 0.5202 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2357 | train 3.179642 | val 3.585815 | tok/s 2626035.68 | norm 0.5271 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2358 | train 3.171727 | val 3.585815 | tok/s 2629187.03 | norm 0.5308 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2359 | train 3.160789 | val 3.585815 | tok/s 2622080.08 | norm 0.5682 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2360 | train 3.133683 | val 3.585815 | tok/s 2626409.52 | norm 0.5950 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2361 | train 3.320398 | val 3.585815 | tok/s 2633144.31 | norm 0.5555 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2362 | train 3.267001 | val 3.585815 | tok/s 2525454.82 | norm 0.5613 | dt 207.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2363 | train 3.295538 | val 3.585815 | tok/s 2629691.95 | norm 0.5626 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2364 | train 3.272706 | val 3.585815 | tok/s 2626534.70 | norm 0.5955 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2365 | train 3.258243 | val 3.585815 | tok/s 2627819.50 | norm 0.5873 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2366 | train 3.209045 | val 3.585815 | tok/s 2631423.30 | norm 0.6098 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2367 | train 3.160766 | val 3.585815 | tok/s 2631286.42 | norm 0.5777 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2368 | train 3.232264 | val 3.585815 | tok/s 2629782.16 | norm 0.5259 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2369 | train 3.182508 | val 3.585815 | tok/s 2626628.37 | norm 0.5451 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2370 | train 3.189143 | val 3.585815 | tok/s 2623041.51 | norm 0.4789 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2371 | train 3.167554 | val 3.585815 | tok/s 2633599.83 | norm 0.4769 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2372 | train 3.288765 | val 3.585815 | tok/s 2626590.59 | norm 0.5186 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2373 | train 3.151446 | val 3.585815 | tok/s 2630348.81 | norm 0.5567 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2374 | train 3.276596 | val 3.585815 | tok/s 2630335.11 | norm 0.5859 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2375 | train 3.158717 | val 3.585815 | tok/s 2629166.33 | norm 0.5706 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2376 | train 3.139296 | val 3.585815 | tok/s 2633119.03 | norm 0.5426 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2377 | train 3.175143 | val 3.585815 | tok/s 2625703.13 | norm 0.5136 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2378 | train 3.187197 | val 3.585815 | tok/s 2628231.28 | norm 0.4948 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2379 | train 3.107184 | val 3.585815 | tok/s 2609695.29 | norm 0.5265 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2380 | train 3.189270 | val 3.585815 | tok/s 2597530.65 | norm 0.5555 | dt 201.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2381 | train 3.152442 | val 3.585815 | tok/s 2631616.04 | norm 0.4963 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2382 | train 3.193711 | val 3.585815 | tok/s 2626414.24 | norm 0.5427 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2383 | train 3.174857 | val 3.585815 | tok/s 2631881.11 | norm 0.5837 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2384 | train 3.201298 | val 3.585815 | tok/s 2628947.13 | norm 0.4814 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2385 | train 3.192359 | val 3.585815 | tok/s 2630772.92 | norm 0.4459 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2386 | train 3.228919 | val 3.585815 | tok/s 2632814.93 | norm 0.4643 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2387 | train 3.160581 | val 3.585815 | tok/s 2616821.72 | norm 0.5019 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2388 | train 3.275572 | val 3.585815 | tok/s 2609604.46 | norm 0.5653 | dt 200.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2389 | train 3.131882 | val 3.585815 | tok/s 2625650.23 | norm 0.5529 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2390 | train 3.261980 | val 3.585815 | tok/s 2629400.73 | norm 0.5102 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2391 | train 3.180693 | val 3.585815 | tok/s 2629885.92 | norm 0.5167 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2392 | train 3.206883 | val 3.585815 | tok/s 2630100.39 | norm 0.5524 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2393 | train 3.119189 | val 3.585815 | tok/s 2631035.20 | norm 0.5493 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2394 | train 3.166419 | val 3.585815 | tok/s 2628310.13 | norm 0.5751 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2395 | train 2.970393 | val 3.585815 | tok/s 2615653.31 | norm 0.6079 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2396 | train 3.201712 | val 3.585815 | tok/s 2631970.91 | norm 0.5507 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2397 | train 3.256074 | val 3.585815 | tok/s 2631265.29 | norm 0.5279 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2398 | train 3.225253 | val 3.585815 | tok/s 2629114.82 | norm 0.5630 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2399 | train 3.101834 | val 3.585815 | tok/s 2626767.20 | norm 0.5647 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2400 | train 3.409125 | val 3.585815 | tok/s 2627488.15 | norm 0.6122 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2401 | train 3.345078 | val 3.585815 | tok/s 2632632.93 | norm 0.5327 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2402 | train 3.135574 | val 3.585815 | tok/s 2630824.31 | norm 0.4322 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2403 | train 3.204156 | val 3.585815 | tok/s 2630869.06 | norm 0.5146 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2404 | train 3.187224 | val 3.585815 | tok/s 2625833.58 | norm 0.5081 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2405 | train 3.132122 | val 3.585815 | tok/s 2628225.49 | norm 0.4909 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2406 | train 3.133399 | val 3.585815 | tok/s 2631649.21 | norm 0.5094 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2407 | train 3.253138 | val 3.585815 | tok/s 2629470.97 | norm 0.5109 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2408 | train 3.148275 | val 3.585815 | tok/s 2631606.55 | norm 0.5379 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2409 | train 3.158901 | val 3.585815 | tok/s 2634890.25 | norm 0.5443 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2410 | train 3.632923 | val 3.585815 | tok/s 2628211.92 | norm 0.6173 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2411 | train 3.219983 | val 3.585815 | tok/s 2628386.17 | norm 0.5624 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2412 | train 3.215392 | val 3.585815 | tok/s 2628446.26 | norm 0.4816 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2413 | train 3.235592 | val 3.585815 | tok/s 2628506.57 | norm 0.5362 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2414 | train 3.262288 | val 3.585815 | tok/s 2629063.42 | norm 0.5389 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2415 | train 3.290289 | val 3.585815 | tok/s 2633647.06 | norm 0.4723 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2416 | train 3.306526 | val 3.585815 | tok/s 2633197.89 | norm 0.4779 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2417 | train 3.067551 | val 3.585815 | tok/s 2629417.47 | norm 0.5040 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2418 | train 3.216692 | val 3.585815 | tok/s 2629395.09 | norm 0.4941 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2419 | train 3.096798 | val 3.585815 | tok/s 2626421.99 | norm 0.5244 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2420 | train 3.140213 | val 3.585815 | tok/s 2632250.06 | norm 0.5194 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2421 | train 3.247121 | val 3.585815 | tok/s 2631653.05 | norm 0.4854 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2422 | train 3.131778 | val 3.585815 | tok/s 2528401.50 | norm 0.4813 | dt 207.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2423 | train 3.141161 | val 3.585815 | tok/s 2628801.39 | norm 0.4658 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2424 | train 3.249275 | val 3.585815 | tok/s 2626182.66 | norm 0.5112 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2425 | train 3.176822 | val 3.585815 | tok/s 2628310.50 | norm 0.5334 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2426 | train 3.257409 | val 3.585815 | tok/s 2627580.31 | norm 0.5187 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2427 | train 3.324651 | val 3.585815 | tok/s 2626291.96 | norm 0.5517 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2428 | train 3.258030 | val 3.585815 | tok/s 2625009.46 | norm 0.4675 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2429 | train 3.175281 | val 3.585815 | tok/s 2630587.35 | norm 0.4834 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2430 | train 3.171118 | val 3.585815 | tok/s 2628469.22 | norm 0.4943 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2431 | train 3.097505 | val 3.585815 | tok/s 2625142.90 | norm 0.5005 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2432 | train 3.169324 | val 3.585815 | tok/s 2626663.32 | norm 0.4954 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2433 | train 3.200933 | val 3.585815 | tok/s 2627526.01 | norm 0.5202 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2434 | train 3.016284 | val 3.585815 | tok/s 2626146.63 | norm 0.5324 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2435 | train 3.282556 | val 3.585815 | tok/s 2627398.08 | norm 0.5604 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2436 | train 3.511328 | val 3.585815 | tok/s 2623596.91 | norm 0.6135 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2437 | train 3.281372 | val 3.585815 | tok/s 2623912.25 | norm 0.4941 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2438 | train 3.289057 | val 3.585815 | tok/s 2632262.57 | norm 0.4525 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2439 | train 3.150422 | val 3.585815 | tok/s 2629624.81 | norm 0.5142 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2440 | train 3.186656 | val 3.585815 | tok/s 2631913.19 | norm 0.5211 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2441 | train 3.247823 | val 3.585815 | tok/s 2625174.28 | norm 0.5109 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2442 | train 3.293143 | val 3.585815 | tok/s 2621600.01 | norm 0.5256 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2443 | train 3.128683 | val 3.585815 | tok/s 2628064.34 | norm 0.5248 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2444 | train 3.159353 | val 3.585815 | tok/s 2626661.42 | norm 0.5547 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2445 | train 3.179186 | val 3.585815 | tok/s 2631380.51 | norm 0.5861 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2446 | train 3.465313 | val 3.585815 | tok/s 2627517.49 | norm 0.6976 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2447 | train 3.234394 | val 3.585815 | tok/s 2630240.60 | norm 0.5068 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2448 | train 3.291019 | val 3.585815 | tok/s 2632747.10 | norm 0.5465 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2449 | train 3.187035 | val 3.585815 | tok/s 2625026.48 | norm 0.5452 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2450 | train 3.197182 | val 3.585815 | tok/s 2628333.16 | norm 0.5153 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2451 | train 3.216023 | val 3.585815 | tok/s 2628557.60 | norm 0.5689 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2452 | train 3.125728 | val 3.585815 | tok/s 2622251.42 | norm 0.5620 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2453 | train 3.267138 | val 3.585815 | tok/s 2625582.91 | norm 0.5279 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2454 | train 3.200686 | val 3.585815 | tok/s 2625200.13 | norm 0.5322 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2455 | train 3.183078 | val 3.585815 | tok/s 2630014.43 | norm 0.5488 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2456 | train 3.274428 | val 3.585815 | tok/s 2628727.25 | norm 0.5532 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2457 | train 3.119829 | val 3.585815 | tok/s 2624740.32 | norm 0.5321 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2458 | train 3.112398 | val 3.585815 | tok/s 2630954.39 | norm 0.5562 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2459 | train 3.123708 | val 3.585815 | tok/s 2627331.16 | norm 0.5156 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2460 | train 3.253806 | val 3.585815 | tok/s 2629328.59 | norm 0.4614 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2461 | train 3.210621 | val 3.585815 | tok/s 2625952.37 | norm 0.4728 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2462 | train 3.198918 | val 3.585815 | tok/s 2627050.14 | norm 0.4888 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2463 | train 3.269807 | val 3.585815 | tok/s 2628799.05 | norm 0.5901 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2464 | train 3.183376 | val 3.585815 | tok/s 2628017.73 | norm 0.5891 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2465 | train 3.251824 | val 3.585815 | tok/s 2630967.65 | norm 0.4895 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2466 | train 3.147518 | val 3.585815 | tok/s 2619195.40 | norm 0.4827 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2467 | train 3.250396 | val 3.585815 | tok/s 2621490.15 | norm 0.5186 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2468 | train 3.117088 | val 3.585815 | tok/s 2629537.09 | norm 0.5167 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2469 | train 3.138080 | val 3.585815 | tok/s 2628886.19 | norm 0.4702 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2470 | train 3.206240 | val 3.585815 | tok/s 2632511.32 | norm 0.4928 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2471 | train 3.151456 | val 3.585815 | tok/s 2621681.57 | norm 0.4911 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2472 | train 3.241504 | val 3.585815 | tok/s 2627566.59 | norm 0.4830 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2473 | train 3.181976 | val 3.585815 | tok/s 2628579.42 | norm 0.4902 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2474 | train 3.282471 | val 3.585815 | tok/s 2623455.80 | norm 0.4965 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2475 | train 3.212776 | val 3.585815 | tok/s 2613929.81 | norm 0.4889 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2476 | train 3.291378 | val 3.585815 | tok/s 2620505.65 | norm 0.4793 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2477 | train 3.157203 | val 3.585815 | tok/s 2625846.59 | norm 0.5378 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2478 | train 3.065800 | val 3.585815 | tok/s 2625875.47 | norm 0.5875 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2479 | train 3.346622 | val 3.585815 | tok/s 2628474.05 | norm 0.6345 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2480 | train 3.199947 | val 3.585815 | tok/s 2626716.50 | norm 0.5134 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2481 | train 3.310361 | val 3.585815 | tok/s 2621376.06 | norm 0.4956 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2482 | train 3.245987 | val 3.585815 | tok/s 2537871.58 | norm 0.5161 | dt 206.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2483 | train 3.168657 | val 3.585815 | tok/s 2629176.17 | norm 0.5691 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2484 | train 3.186293 | val 3.585815 | tok/s 2628549.62 | norm 0.5611 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2485 | train 3.194468 | val 3.585815 | tok/s 2629075.12 | norm 0.4893 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2486 | train 3.221074 | val 3.585815 | tok/s 2628423.76 | norm 0.4803 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2487 | train 3.169269 | val 3.585815 | tok/s 2628483.55 | norm 0.5002 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2488 | train 3.116730 | val 3.585815 | tok/s 2627491.72 | norm 0.4971 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2489 | train 3.190526 | val 3.585815 | tok/s 2627145.86 | norm 0.5664 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2490 | train 3.147366 | val 3.585815 | tok/s 2626368.22 | norm 0.5322 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2491 | train 3.238244 | val 3.585815 | tok/s 2619097.61 | norm 0.5002 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2492 | train 3.147220 | val 3.585815 | tok/s 2628223.30 | norm 0.5555 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2493 | train 3.237304 | val 3.585815 | tok/s 2631155.08 | norm 0.5504 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2494 | train 3.097027 | val 3.585815 | tok/s 2627611.26 | norm 0.4936 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2495 | train 3.245176 | val 3.585815 | tok/s 2630183.11 | norm 0.4750 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2496 | train 3.112186 | val 3.585815 | tok/s 2626622.69 | norm 0.4572 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2497 | train 3.076014 | val 3.585815 | tok/s 2627020.58 | norm 0.5125 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2498 | train 3.143716 | val 3.585815 | tok/s 2632116.11 | norm 0.5476 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2499 | train 3.153612 | val 3.585815 | tok/s 2503988.30 | norm 0.5109 | dt 209.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 2500 | val_loss 3.569917 step 2500 | train 3.195743 | val 3.569917 | tok/s 2618532.22 | norm 0.6000 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2501 | train 3.127514 | val 3.569917 | tok/s 2631218.82 | norm 0.5944 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2502 | train 3.084671 | val 3.569917 | tok/s 2626633.81 | norm 0.5926 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2503 | train 3.162188 | val 3.569917 | tok/s 2628840.62 | norm 0.5445 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2504 | train 3.170629 | val 3.569917 | tok/s 2631957.39 | norm 0.5237 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2505 | train 3.141085 | val 3.569917 | tok/s 2624769.26 | norm 0.5148 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2506 | train 3.248222 | val 3.569917 | tok/s 2626942.79 | norm 0.5263 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2507 | train 3.121600 | val 3.569917 | tok/s 2525539.08 | norm 0.5123 | dt 207.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2508 | train 3.235024 | val 3.569917 | tok/s 2606130.04 | norm 0.5520 | dt 201.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2509 | train 3.156633 | val 3.569917 | tok/s 2630990.36 | norm 0.5364 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2510 | train 3.263680 | val 3.569917 | tok/s 2627864.27 | norm 0.5423 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2511 | train 3.203640 | val 3.569917 | tok/s 2629132.13 | norm 0.5236 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2512 | train 3.164607 | val 3.569917 | tok/s 2626677.00 | norm 0.5543 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2513 | train 3.238346 | val 3.569917 | tok/s 2624552.30 | norm 0.4942 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2514 | train 3.189326 | val 3.569917 | tok/s 2630160.68 | norm 0.4777 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2515 | train 3.157243 | val 3.569917 | tok/s 2625205.32 | norm 0.5003 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2516 | train 3.236801 | val 3.569917 | tok/s 2626267.13 | norm 0.4670 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2517 | train 3.224494 | val 3.569917 | tok/s 2623095.34 | norm 0.4799 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2518 | train 3.276606 | val 3.569917 | tok/s 2611210.06 | norm 0.4838 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2519 | train 3.284843 | val 3.569917 | tok/s 2630293.96 | norm 0.5085 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2520 | train 3.227541 | val 3.569917 | tok/s 2624663.13 | norm 0.5589 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2521 | train 3.194129 | val 3.569917 | tok/s 2625691.41 | norm 0.5563 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2522 | train 3.136186 | val 3.569917 | tok/s 2625833.22 | norm 0.5726 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2523 | train 3.292135 | val 3.569917 | tok/s 2626884.73 | norm 0.5302 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2524 | train 3.152016 | val 3.569917 | tok/s 2630605.54 | norm 0.5355 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2525 | train 3.249409 | val 3.569917 | tok/s 2633163.59 | norm 0.4874 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2526 | train 3.156073 | val 3.569917 | tok/s 2605955.58 | norm 0.4431 | dt 201.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2527 | train 3.253308 | val 3.569917 | tok/s 2628840.03 | norm 0.5182 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2528 | train 3.271195 | val 3.569917 | tok/s 2629693.64 | norm 0.4699 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2529 | train 3.232243 | val 3.569917 | tok/s 2622045.75 | norm 0.4365 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2530 | train 3.148573 | val 3.569917 | tok/s 2624763.96 | norm 0.4648 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2531 | train 3.107017 | val 3.569917 | tok/s 2621083.49 | norm 0.4726 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2532 | train 3.221808 | val 3.569917 | tok/s 2622233.75 | norm 0.4878 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2533 | train 3.478704 | val 3.569917 | tok/s 2624938.08 | norm 0.5023 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2534 | train 3.246639 | val 3.569917 | tok/s 2628740.49 | norm 0.6024 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2535 | train 3.536132 | val 3.569917 | tok/s 2626506.78 | norm 0.6709 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2536 | train 3.249649 | val 3.569917 | tok/s 2628309.49 | norm 0.5639 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2537 | train 3.086222 | val 3.569917 | tok/s 2623128.99 | norm 0.4665 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2538 | train 3.203783 | val 3.569917 | tok/s 2629568.67 | norm 0.7060 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2539 | train 3.315336 | val 3.569917 | tok/s 2630818.00 | norm 0.6175 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2540 | train 3.313538 | val 3.569917 | tok/s 2626586.02 | norm 0.5216 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2541 | train 3.125889 | val 3.569917 | tok/s 2630986.15 | norm 0.4894 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2542 | train 3.260263 | val 3.569917 | tok/s 2626780.81 | norm 0.5005 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2543 | train 3.047251 | val 3.569917 | tok/s 2619633.47 | norm 0.4984 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2544 | train 3.180824 | val 3.569917 | tok/s 2627129.51 | norm 0.5114 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2545 | train 3.183879 | val 3.569917 | tok/s 2626986.63 | norm 0.5517 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2546 | train 3.093660 | val 3.569917 | tok/s 2628549.07 | norm 0.5351 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2547 | train 3.230218 | val 3.569917 | tok/s 2627012.00 | norm 0.5647 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2548 | train 3.247581 | val 3.569917 | tok/s 2624396.46 | norm 0.5410 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2549 | train 3.186755 | val 3.569917 | tok/s 2623809.27 | norm 0.5539 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2550 | train 3.178300 | val 3.569917 | tok/s 2627934.70 | norm 0.5377 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2551 | train 3.293165 | val 3.569917 | tok/s 2629325.76 | norm 0.6006 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2552 | train 3.243958 | val 3.569917 | tok/s 2628325.78 | norm 0.5390 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2553 | train 3.137675 | val 3.569917 | tok/s 2628913.45 | norm 0.5036 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2554 | train 3.236925 | val 3.569917 | tok/s 2631761.20 | norm 0.5524 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2555 | train 3.195699 | val 3.569917 | tok/s 2631221.66 | norm 0.5380 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2556 | train 3.227319 | val 3.569917 | tok/s 2629661.48 | norm 0.5447 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2557 | train 3.098527 | val 3.569917 | tok/s 2605308.05 | norm 0.5436 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2558 | train 3.217382 | val 3.569917 | tok/s 2629292.50 | norm 0.5203 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2559 | train 3.179214 | val 3.569917 | tok/s 2628254.55 | norm 0.5781 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2560 | train 3.227236 | val 3.569917 | tok/s 2628435.65 | norm 0.6019 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2561 | train 3.079980 | val 3.569917 | tok/s 2627643.70 | norm 0.6379 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2562 | train 3.146942 | val 3.569917 | tok/s 2627178.55 | norm 0.5687 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2563 | train 3.148101 | val 3.569917 | tok/s 2623798.65 | norm 0.7865 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2564 | train 3.178022 | val 3.569917 | tok/s 2628975.72 | norm 0.4803 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2565 | train 3.274878 | val 3.569917 | tok/s 2621997.53 | norm 0.5242 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2566 | train 3.154366 | val 3.569917 | tok/s 2615275.40 | norm 0.5511 | dt 200.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2567 | train 3.408052 | val 3.569917 | tok/s 2625067.84 | norm 0.5027 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2568 | train 3.173895 | val 3.569917 | tok/s 2624681.00 | norm 0.5217 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2569 | train 3.271847 | val 3.569917 | tok/s 2627204.61 | norm 0.5100 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2570 | train 3.159166 | val 3.569917 | tok/s 2628203.54 | norm 0.4962 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2571 | train 3.247955 | val 3.569917 | tok/s 2630416.12 | norm 0.5217 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2572 | train 3.590073 | val 3.569917 | tok/s 2624030.02 | norm 0.5037 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2573 | train 3.067270 | val 3.569917 | tok/s 2627596.89 | norm 0.4369 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2574 | train 3.225068 | val 3.569917 | tok/s 2628296.58 | norm 0.5048 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2575 | train 3.186438 | val 3.569917 | tok/s 2622586.75 | norm 0.5291 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2576 | train 3.111803 | val 3.569917 | tok/s 2630126.35 | norm 0.5080 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2577 | train 3.247652 | val 3.569917 | tok/s 2629841.82 | norm 0.4907 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2578 | train 3.203708 | val 3.569917 | tok/s 2629931.98 | norm 0.4939 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2579 | train 3.055076 | val 3.569917 | tok/s 2631459.73 | norm 0.5406 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2580 | train 3.164531 | val 3.569917 | tok/s 2634474.05 | norm 0.5251 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2581 | train 3.161129 | val 3.569917 | tok/s 2634464.53 | norm 0.5402 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2582 | train 3.064801 | val 3.569917 | tok/s 2628784.71 | norm 0.5111 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2583 | train 3.207603 | val 3.569917 | tok/s 2629927.55 | norm 0.5385 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2584 | train 3.332490 | val 3.569917 | tok/s 2629292.63 | norm 0.5037 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2585 | train 3.243260 | val 3.569917 | tok/s 2625427.71 | norm 0.4960 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2586 | train 3.110869 | val 3.569917 | tok/s 2629507.90 | norm 0.4298 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2587 | train 3.144339 | val 3.569917 | tok/s 2624577.60 | norm 0.4619 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2588 | train 3.185859 | val 3.569917 | tok/s 2629342.95 | norm 0.5002 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2589 | train 3.173541 | val 3.569917 | tok/s 2627439.84 | norm 0.5118 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2590 | train 3.200044 | val 3.569917 | tok/s 2626256.58 | norm 0.5545 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2591 | train 3.229841 | val 3.569917 | tok/s 2627618.74 | norm 0.5286 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2592 | train 3.194403 | val 3.569917 | tok/s 2630971.50 | norm 0.4931 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2593 | train 3.159847 | val 3.569917 | tok/s 2630570.41 | norm 0.5163 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2594 | train 3.264261 | val 3.569917 | tok/s 2627118.56 | norm 0.5405 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2595 | train 3.185494 | val 3.569917 | tok/s 2623391.92 | norm 0.4605 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2596 | train 3.183598 | val 3.569917 | tok/s 2634136.75 | norm 0.4584 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2597 | train 3.264116 | val 3.569917 | tok/s 2627027.00 | norm 0.5114 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2598 | train 3.098108 | val 3.569917 | tok/s 2632938.40 | norm 0.4901 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2599 | train 3.243091 | val 3.569917 | tok/s 2629092.72 | norm 0.5093 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2600 | train 3.222440 | val 3.569917 | tok/s 2627665.68 | norm 0.5288 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2601 | train 3.373996 | val 3.569917 | tok/s 2632882.31 | norm 0.5426 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2602 | train 3.157118 | val 3.569917 | tok/s 2630752.52 | norm 0.5241 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2603 | train 3.194980 | val 3.569917 | tok/s 2626237.82 | norm 0.5100 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2604 | train 3.110198 | val 3.569917 | tok/s 2627414.20 | norm 0.5431 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2605 | train 3.216275 | val 3.569917 | tok/s 2627742.77 | norm 0.5043 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2606 | train 3.182465 | val 3.569917 | tok/s 2630779.83 | norm 0.5560 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2607 | train 3.107649 | val 3.569917 | tok/s 2624185.51 | norm 0.5037 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2608 | train 3.250480 | val 3.569917 | tok/s 2629178.01 | norm 0.4753 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2609 | train 3.356465 | val 3.569917 | tok/s 2627852.38 | norm 0.5038 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2610 | train 3.167816 | val 3.569917 | tok/s 2624694.15 | norm 0.4886 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2611 | train 3.234889 | val 3.569917 | tok/s 2630398.14 | norm 0.4682 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2612 | train 3.240803 | val 3.569917 | tok/s 2627203.66 | norm 0.4693 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2613 | train 3.242028 | val 3.569917 | tok/s 2614617.44 | norm 0.4789 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2614 | train 3.118979 | val 3.569917 | tok/s 2626931.96 | norm 0.5369 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2615 | train 3.209766 | val 3.569917 | tok/s 2627835.89 | norm 0.5423 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2616 | train 3.133847 | val 3.569917 | tok/s 2633755.73 | norm 0.5001 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2617 | train 3.167824 | val 3.569917 | tok/s 2629219.84 | norm 0.5349 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2618 | train 3.125529 | val 3.569917 | tok/s 2628477.42 | norm 0.5663 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2619 | train 3.308424 | val 3.569917 | tok/s 2616833.44 | norm 0.6177 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2620 | train 3.301337 | val 3.569917 | tok/s 2627891.12 | norm 0.5432 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2621 | train 3.202893 | val 3.569917 | tok/s 2633015.99 | norm 0.4489 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2622 | train 3.243474 | val 3.569917 | tok/s 2624952.53 | norm 0.5087 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2623 | train 3.284168 | val 3.569917 | tok/s 2629620.53 | norm 0.5316 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2624 | train 3.293854 | val 3.569917 | tok/s 2626226.10 | norm 0.4866 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2625 | train 3.085402 | val 3.569917 | tok/s 2624390.51 | norm 0.4880 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2626 | train 3.282691 | val 3.569917 | tok/s 2630625.34 | norm 0.4657 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2627 | train 3.167794 | val 3.569917 | tok/s 2630454.20 | norm 0.4971 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2628 | train 3.149655 | val 3.569917 | tok/s 2631474.44 | norm 0.4794 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2629 | train 3.125822 | val 3.569917 | tok/s 2629337.72 | norm 0.5238 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2630 | train 3.115160 | val 3.569917 | tok/s 2620783.01 | norm 0.5821 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2631 | train 3.123329 | val 3.569917 | tok/s 2627393.53 | norm 0.5311 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2632 | train 3.235331 | val 3.569917 | tok/s 2627206.64 | norm 0.5492 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2633 | train 3.288775 | val 3.569917 | tok/s 2630839.41 | norm 0.5956 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2634 | train 3.165259 | val 3.569917 | tok/s 2621822.60 | norm 0.4884 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2635 | train 3.106394 | val 3.569917 | tok/s 2625185.80 | norm 0.5188 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2636 | train 3.411449 | val 3.569917 | tok/s 2633576.03 | norm 0.5484 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2637 | train 3.169829 | val 3.569917 | tok/s 2631252.86 | norm 0.5799 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2638 | train 3.225748 | val 3.569917 | tok/s 2632347.00 | norm 0.5617 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2639 | train 3.206469 | val 3.569917 | tok/s 2623604.36 | norm 0.5586 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2640 | train 3.126334 | val 3.569917 | tok/s 2628170.11 | norm 0.5192 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2641 | train 3.235319 | val 3.569917 | tok/s 2629506.01 | norm 0.5042 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2642 | train 3.228580 | val 3.569917 | tok/s 2631096.14 | norm 0.4825 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2643 | train 3.144258 | val 3.569917 | tok/s 2632364.23 | norm 0.5008 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2644 | train 3.301323 | val 3.569917 | tok/s 2628348.52 | norm 0.4868 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2645 | train 3.188212 | val 3.569917 | tok/s 2626851.09 | norm 0.4746 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2646 | train 3.364433 | val 3.569917 | tok/s 2631922.72 | norm 0.5022 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2647 | train 3.175933 | val 3.569917 | tok/s 2628106.84 | norm 0.4751 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2648 | train 3.138632 | val 3.569917 | tok/s 2631327.30 | norm 0.4742 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2649 | train 3.071989 | val 3.569917 | tok/s 2624318.65 | norm 0.5310 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2650 | train 3.173131 | val 3.569917 | tok/s 2624415.01 | norm 0.4954 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2651 | train 3.228686 | val 3.569917 | tok/s 2627932.89 | norm 0.5295 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2652 | train 3.142894 | val 3.569917 | tok/s 2626058.51 | norm 0.5013 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2653 | train 3.262153 | val 3.569917 | tok/s 2630797.11 | norm 0.4814 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2654 | train 3.120898 | val 3.569917 | tok/s 2629063.94 | norm 0.5064 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2655 | train 3.115222 | val 3.569917 | tok/s 2625838.10 | norm 0.5498 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2656 | train 3.165929 | val 3.569917 | tok/s 2628770.75 | norm 0.5260 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2657 | train 3.119131 | val 3.569917 | tok/s 2628060.74 | norm 0.5340 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2658 | train 3.143542 | val 3.569917 | tok/s 2626721.04 | norm 0.5266 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2659 | train 3.227606 | val 3.569917 | tok/s 2560954.29 | norm 0.4851 | dt 204.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2660 | train 3.204715 | val 3.569917 | tok/s 2626170.35 | norm 0.4853 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2661 | train 3.022415 | val 3.569917 | tok/s 2630759.40 | norm 0.5562 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2662 | train 3.081397 | val 3.569917 | tok/s 2628292.09 | norm 0.5601 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2663 | train 3.174206 | val 3.569917 | tok/s 2621148.28 | norm 0.5023 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2664 | train 3.165097 | val 3.569917 | tok/s 2625486.77 | norm 0.4920 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2665 | train 3.209359 | val 3.569917 | tok/s 2626552.02 | norm 0.5568 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2666 | train 3.309346 | val 3.569917 | tok/s 2631003.71 | norm 0.5602 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2667 | train 3.042268 | val 3.569917 | tok/s 2578396.44 | norm 0.5782 | dt 203.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2668 | train 3.242756 | val 3.569917 | tok/s 2629198.77 | norm 0.5354 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2669 | train 3.125669 | val 3.569917 | tok/s 2623804.87 | norm 0.5123 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2670 | train 3.169145 | val 3.569917 | tok/s 2626129.73 | norm 0.5535 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2671 | train 3.236244 | val 3.569917 | tok/s 2629591.34 | norm 0.5417 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2672 | train 3.206269 | val 3.569917 | tok/s 2606855.81 | norm 0.4898 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2673 | train 3.213425 | val 3.569917 | tok/s 2625839.87 | norm 0.5085 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2674 | train 3.157220 | val 3.569917 | tok/s 2623484.35 | norm 0.5271 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2675 | train 3.160032 | val 3.569917 | tok/s 2632624.86 | norm 0.5713 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2676 | train 3.037100 | val 3.569917 | tok/s 2627511.73 | norm 0.5788 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2677 | train 3.182917 | val 3.569917 | tok/s 2630011.47 | norm 0.5280 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2678 | train 3.226424 | val 3.569917 | tok/s 2628351.70 | norm 0.5428 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2679 | train 3.251348 | val 3.569917 | tok/s 2622137.46 | norm 0.5019 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2680 | train 3.138233 | val 3.569917 | tok/s 2629634.82 | norm 0.4763 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2681 | train 3.070162 | val 3.569917 | tok/s 2624456.92 | norm 0.4874 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2682 | train 3.253377 | val 3.569917 | tok/s 2626254.44 | norm 0.5143 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2683 | train 3.298200 | val 3.569917 | tok/s 2625462.93 | norm 0.5363 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2684 | train 3.246488 | val 3.569917 | tok/s 2621980.91 | norm 0.5359 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2685 | train 3.171130 | val 3.569917 | tok/s 2627901.72 | norm 0.5329 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2686 | train 3.138288 | val 3.569917 | tok/s 2626103.20 | norm 0.5624 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2687 | train 3.212571 | val 3.569917 | tok/s 2632860.14 | norm 0.5399 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2688 | train 3.234051 | val 3.569917 | tok/s 2630250.87 | norm 0.5403 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2689 | train 3.149058 | val 3.569917 | tok/s 2626078.23 | norm 0.5927 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2690 | train 3.232217 | val 3.569917 | tok/s 2630044.51 | norm 0.5609 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2691 | train 3.171988 | val 3.569917 | tok/s 2629872.54 | norm 0.5239 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2692 | train 3.143657 | val 3.569917 | tok/s 2625150.81 | norm 0.4777 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2693 | train 3.169742 | val 3.569917 | tok/s 2623000.50 | norm 0.5254 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2694 | train 3.158921 | val 3.569917 | tok/s 2617226.14 | norm 0.5039 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2695 | train 3.234095 | val 3.569917 | tok/s 2625583.01 | norm 0.5239 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2696 | train 3.223242 | val 3.569917 | tok/s 2625186.44 | norm 0.5719 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2697 | train 3.190457 | val 3.569917 | tok/s 2623867.77 | norm 0.5420 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2698 | train 3.198564 | val 3.569917 | tok/s 2622810.67 | norm 0.5424 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2699 | train 3.274074 | val 3.569917 | tok/s 2621356.47 | norm 0.5370 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2700 | train 3.208528 | val 3.569917 | tok/s 2624657.57 | norm 0.5382 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2701 | train 3.228967 | val 3.569917 | tok/s 2623872.30 | norm 0.5440 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2702 | train 3.183010 | val 3.569917 | tok/s 2630142.50 | norm 0.4898 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2703 | train 3.131874 | val 3.569917 | tok/s 2631409.11 | norm 0.4559 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2704 | train 3.273954 | val 3.569917 | tok/s 2628260.19 | norm 0.4678 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2705 | train 3.205396 | val 3.569917 | tok/s 2625589.07 | norm 0.5057 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2706 | train 3.201941 | val 3.569917 | tok/s 2621505.47 | norm 0.4978 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2707 | train 3.156853 | val 3.569917 | tok/s 2632233.39 | norm 0.5208 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2708 | train 3.176765 | val 3.569917 | tok/s 2628261.17 | norm 0.5349 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2709 | train 3.072620 | val 3.569917 | tok/s 2627695.06 | norm 0.5035 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2710 | train 3.212621 | val 3.569917 | tok/s 2624771.14 | norm 0.5255 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2711 | train 3.167250 | val 3.569917 | tok/s 2629312.17 | norm 0.5008 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2712 | train 3.247731 | val 3.569917 | tok/s 2627941.43 | norm 0.4919 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2713 | train 3.153875 | val 3.569917 | tok/s 2617128.45 | norm 0.5055 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2714 | train 3.179309 | val 3.569917 | tok/s 2623097.59 | norm 0.5811 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2715 | train 3.173622 | val 3.569917 | tok/s 2629950.94 | norm 0.5243 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2716 | train 3.178420 | val 3.569917 | tok/s 2624821.03 | norm 0.5251 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2717 | train 3.200148 | val 3.569917 | tok/s 2629586.45 | norm 0.5190 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2718 | train 3.224370 | val 3.569917 | tok/s 2626029.91 | norm 0.4746 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2719 | train 3.148928 | val 3.569917 | tok/s 2627869.16 | norm 0.5136 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2720 | train 3.150931 | val 3.569917 | tok/s 2631679.15 | norm 0.4984 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2721 | train 3.252048 | val 3.569917 | tok/s 2625606.25 | norm 0.5068 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2722 | train 3.173846 | val 3.569917 | tok/s 2627712.07 | norm 0.5683 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2723 | train 3.098780 | val 3.569917 | tok/s 2619365.83 | norm 0.5882 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2724 | train 3.292472 | val 3.569917 | tok/s 2627163.20 | norm 0.6414 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2725 | train 3.175192 | val 3.569917 | tok/s 2629091.87 | norm 0.5466 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2726 | train 3.152316 | val 3.569917 | tok/s 2628320.44 | norm 0.5248 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2727 | train 3.282875 | val 3.569917 | tok/s 2632821.65 | norm 0.5046 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2728 | train 3.158251 | val 3.569917 | tok/s 2625090.29 | norm 0.5007 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2729 | train 3.219020 | val 3.569917 | tok/s 2627544.63 | norm 0.4834 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2730 | train 3.262882 | val 3.569917 | tok/s 2633832.30 | norm 0.4736 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2731 | train 3.146960 | val 3.569917 | tok/s 2624619.61 | norm 0.4795 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2732 | train 3.086074 | val 3.569917 | tok/s 2626065.87 | norm 0.4915 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2733 | train 3.247432 | val 3.569917 | tok/s 2623577.22 | norm 0.5181 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2734 | train 3.289474 | val 3.569917 | tok/s 2625068.05 | norm 0.5362 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2735 | train 3.077926 | val 3.569917 | tok/s 2627730.38 | norm 0.4618 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2736 | train 3.229785 | val 3.569917 | tok/s 2626786.67 | norm 0.4763 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2737 | train 3.216789 | val 3.569917 | tok/s 2629118.26 | norm 0.5193 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2738 | train 3.112003 | val 3.569917 | tok/s 2619896.06 | norm 0.4888 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2739 | train 3.118350 | val 3.569917 | tok/s 2624660.69 | norm 0.5218 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2740 | train 3.145665 | val 3.569917 | tok/s 2626622.50 | norm 0.4966 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2741 | train 3.118197 | val 3.569917 | tok/s 2625284.23 | norm 0.5073 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2742 | train 3.158321 | val 3.569917 | tok/s 2625693.57 | norm 0.5582 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2743 | train 3.250075 | val 3.569917 | tok/s 2621343.77 | norm 0.5016 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2744 | train 3.130513 | val 3.569917 | tok/s 2628285.03 | norm 0.5001 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2745 | train 3.237043 | val 3.569917 | tok/s 2631110.65 | norm 0.5410 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2746 | train 3.126149 | val 3.569917 | tok/s 2622641.17 | norm 0.5413 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2747 | train 3.189451 | val 3.569917 | tok/s 2628714.81 | norm 0.5432 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2748 | train 3.129326 | val 3.569917 | tok/s 2627235.41 | norm 0.5302 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2749 | train 3.218754 | val 3.569917 | tok/s 2625533.60 | norm 0.5202 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 2750 | val_loss 3.545726 step 2750 | train 3.169813 | val 3.545726 | tok/s 2547605.24 | norm 0.4620 | dt 205.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2751 | train 3.126315 | val 3.545726 | tok/s 2624601.41 | norm 0.4865 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2752 | train 3.267027 | val 3.545726 | tok/s 2621180.15 | norm 0.5330 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2753 | train 3.229198 | val 3.545726 | tok/s 2592250.41 | norm 0.5115 | dt 202.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2754 | train 3.202015 | val 3.545726 | tok/s 2624534.56 | norm 0.5938 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2755 | train 3.161016 | val 3.545726 | tok/s 2624259.99 | norm 0.6178 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2756 | train 3.206169 | val 3.545726 | tok/s 2625559.56 | norm 0.5140 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2757 | train 3.093484 | val 3.545726 | tok/s 2621997.12 | norm 0.5374 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2758 | train 3.119363 | val 3.545726 | tok/s 2631851.68 | norm 0.5539 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2759 | train 3.207309 | val 3.545726 | tok/s 2623270.92 | norm 0.5306 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2760 | train 3.123381 | val 3.545726 | tok/s 2625208.05 | norm 0.5236 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2761 | train 3.242188 | val 3.545726 | tok/s 2628812.45 | norm 0.4855 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2762 | train 3.214575 | val 3.545726 | tok/s 2624957.46 | norm 0.4739 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2763 | train 3.274914 | val 3.545726 | tok/s 2631586.26 | norm 0.5304 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2764 | train 3.161489 | val 3.545726 | tok/s 2625218.13 | norm 0.4710 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2765 | train 3.268782 | val 3.545726 | tok/s 2627947.22 | norm 0.4972 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2766 | train 3.218191 | val 3.545726 | tok/s 2631199.08 | norm 0.5530 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2767 | train 3.072468 | val 3.545726 | tok/s 2627052.33 | norm 0.6439 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2768 | train 3.119563 | val 3.545726 | tok/s 2626967.97 | norm 0.4914 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2769 | train 3.206714 | val 3.545726 | tok/s 2621602.75 | norm 0.4479 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2770 | train 3.248145 | val 3.545726 | tok/s 2626461.47 | norm 0.5522 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2771 | train 3.283089 | val 3.545726 | tok/s 2630033.11 | norm 0.5796 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2772 | train 3.134185 | val 3.545726 | tok/s 2629068.22 | norm 0.5584 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2773 | train 3.226302 | val 3.545726 | tok/s 2628959.53 | norm 0.5358 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2774 | train 3.192780 | val 3.545726 | tok/s 2619366.48 | norm 0.5199 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2775 | train 3.093575 | val 3.545726 | tok/s 2624980.92 | norm 0.5071 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2776 | train 3.272056 | val 3.545726 | tok/s 2631968.94 | norm 0.4953 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2777 | train 3.102762 | val 3.545726 | tok/s 2628333.78 | norm 0.4914 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2778 | train 3.221615 | val 3.545726 | tok/s 2630142.68 | norm 0.5465 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2779 | train 3.036367 | val 3.545726 | tok/s 2625937.06 | norm 0.4857 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2780 | train 3.319275 | val 3.545726 | tok/s 2629364.64 | norm 0.5720 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2781 | train 3.105240 | val 3.545726 | tok/s 2631700.99 | norm 0.5426 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2782 | train 3.154345 | val 3.545726 | tok/s 2627353.87 | norm 0.4777 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2783 | train 3.043083 | val 3.545726 | tok/s 2629809.33 | norm 0.5208 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2784 | train 3.163471 | val 3.545726 | tok/s 2623251.97 | norm 0.5806 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2785 | train 3.069910 | val 3.545726 | tok/s 2625686.65 | norm 0.5670 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2786 | train 3.284660 | val 3.545726 | tok/s 2631273.52 | norm 0.5490 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2787 | train 2.985165 | val 3.545726 | tok/s 2627940.68 | norm 0.5528 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2788 | train 2.971514 | val 3.545726 | tok/s 2626593.39 | norm 0.5027 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2789 | train 3.373278 | val 3.545726 | tok/s 2625977.44 | norm 0.5485 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2790 | train 3.181599 | val 3.545726 | tok/s 2624162.95 | norm 0.4774 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2791 | train 3.269009 | val 3.545726 | tok/s 2628282.67 | norm 0.4679 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2792 | train 3.225629 | val 3.545726 | tok/s 2628139.69 | norm 0.5096 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2793 | train 3.196688 | val 3.545726 | tok/s 2628848.92 | norm 0.4838 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2794 | train 3.084196 | val 3.545726 | tok/s 2626576.76 | norm 0.5043 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2795 | train 3.241891 | val 3.545726 | tok/s 2629136.27 | norm 0.5469 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2796 | train 3.188034 | val 3.545726 | tok/s 2629381.15 | norm 0.5075 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2797 | train 3.037092 | val 3.545726 | tok/s 2626150.20 | norm 0.5108 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2798 | train 3.208363 | val 3.545726 | tok/s 2628928.40 | norm 0.5499 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2799 | train 3.199050 | val 3.545726 | tok/s 2622832.08 | norm 0.5047 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2800 | train 3.102894 | val 3.545726 | tok/s 2627838.43 | norm 0.5325 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2801 | train 3.273867 | val 3.545726 | tok/s 2628406.76 | norm 0.5359 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2802 | train 3.221773 | val 3.545726 | tok/s 2627816.59 | norm 0.4661 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2803 | train 3.200863 | val 3.545726 | tok/s 2629286.12 | norm 0.4766 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2804 | train 3.132166 | val 3.545726 | tok/s 2620219.78 | norm 0.4890 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2805 | train 3.231527 | val 3.545726 | tok/s 2623491.75 | norm 0.4550 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2806 | train 3.133927 | val 3.545726 | tok/s 2625902.92 | norm 0.4697 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2807 | train 3.213433 | val 3.545726 | tok/s 2628497.25 | norm 0.5242 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2808 | train 3.183712 | val 3.545726 | tok/s 2628819.16 | norm 0.5534 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2809 | train 3.262417 | val 3.545726 | tok/s 2624869.47 | norm 0.4986 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2810 | train 3.109761 | val 3.545726 | tok/s 2628160.19 | norm 0.4631 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2811 | train 3.240380 | val 3.545726 | tok/s 2632431.52 | norm 0.5052 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2812 | train 3.090577 | val 3.545726 | tok/s 2625255.28 | norm 0.4628 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2813 | train 3.116209 | val 3.545726 | tok/s 2630719.55 | norm 0.4742 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2814 | train 3.186281 | val 3.545726 | tok/s 2625532.93 | norm 0.5013 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2815 | train 3.102988 | val 3.545726 | tok/s 2630851.03 | norm 0.4741 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2816 | train 3.157158 | val 3.545726 | tok/s 2633535.14 | norm 0.4822 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2817 | train 3.080379 | val 3.545726 | tok/s 2628427.70 | norm 0.5201 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2818 | train 3.151445 | val 3.545726 | tok/s 2631839.01 | norm 0.5259 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2819 | train 3.143935 | val 3.545726 | tok/s 2624209.59 | norm 0.5376 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2820 | train 3.251827 | val 3.545726 | tok/s 2627536.61 | norm 0.5747 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2821 | train 3.135360 | val 3.545726 | tok/s 2627306.13 | norm 0.5755 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2822 | train 3.173030 | val 3.545726 | tok/s 2633127.72 | norm 0.5754 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2823 | train 3.221034 | val 3.545726 | tok/s 2625983.00 | norm 0.5407 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2824 | train 3.203369 | val 3.545726 | tok/s 2622280.30 | norm 0.4595 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2825 | train 3.053506 | val 3.545726 | tok/s 2627435.76 | norm 0.5048 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2826 | train 3.137367 | val 3.545726 | tok/s 2631222.58 | norm 0.5473 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2827 | train 3.183396 | val 3.545726 | tok/s 2626762.43 | norm 0.4983 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2828 | train 3.114860 | val 3.545726 | tok/s 2626868.79 | norm 0.5148 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2829 | train 3.189620 | val 3.545726 | tok/s 2621569.72 | norm 0.5861 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2830 | train 3.193840 | val 3.545726 | tok/s 2625990.59 | norm 0.5554 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2831 | train 3.146194 | val 3.545726 | tok/s 2630970.73 | norm 0.5243 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2832 | train 3.044606 | val 3.545726 | tok/s 2589190.61 | norm 0.5860 | dt 202.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2833 | train 3.135964 | val 3.545726 | tok/s 2628355.19 | norm 0.5594 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2834 | train 3.048927 | val 3.545726 | tok/s 2627521.77 | norm 0.5044 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2835 | train 3.189274 | val 3.545726 | tok/s 2627197.35 | norm 0.5019 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2836 | train 3.360996 | val 3.545726 | tok/s 2626713.93 | norm 0.4698 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2837 | train 3.189837 | val 3.545726 | tok/s 2628826.00 | norm 0.4556 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2838 | train 3.293294 | val 3.545726 | tok/s 2632350.08 | norm 0.4868 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2839 | train 3.169514 | val 3.545726 | tok/s 2626617.08 | norm 0.5144 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2840 | train 3.158570 | val 3.545726 | tok/s 2591875.63 | norm 0.4812 | dt 202.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2841 | train 3.097965 | val 3.545726 | tok/s 2626029.92 | norm 0.4414 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2842 | train 3.032878 | val 3.545726 | tok/s 2551452.00 | norm 0.4513 | dt 205.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2843 | train 3.059193 | val 3.545726 | tok/s 2632289.10 | norm 0.4411 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2844 | train 3.032593 | val 3.545726 | tok/s 2631443.42 | norm 0.4743 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2845 | train 3.285480 | val 3.545726 | tok/s 2630823.17 | norm 0.4734 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2846 | train 3.227164 | val 3.545726 | tok/s 2631051.03 | norm 0.5573 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2847 | train 3.191331 | val 3.545726 | tok/s 2629782.08 | norm 0.5804 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2848 | train 3.065626 | val 3.545726 | tok/s 2630398.61 | norm 0.5386 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2849 | train 3.132441 | val 3.545726 | tok/s 2623605.81 | norm 0.5057 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2850 | train 3.256108 | val 3.545726 | tok/s 2629516.18 | norm 0.5131 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2851 | train 3.174680 | val 3.545726 | tok/s 2629481.07 | norm 0.5046 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2852 | train 3.096250 | val 3.545726 | tok/s 2625663.02 | norm 0.5894 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2853 | train 3.211541 | val 3.545726 | tok/s 2632141.84 | norm 0.5323 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2854 | train 3.167364 | val 3.545726 | tok/s 2629088.48 | norm 0.4963 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2855 | train 3.215403 | val 3.545726 | tok/s 2630433.35 | norm 0.5312 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2856 | train 3.133877 | val 3.545726 | tok/s 2634340.80 | norm 0.5072 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2857 | train 3.157823 | val 3.545726 | tok/s 2629992.54 | norm 0.5392 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2858 | train 3.182010 | val 3.545726 | tok/s 2629134.56 | norm 0.4899 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2859 | train 3.166232 | val 3.545726 | tok/s 2625573.90 | norm 0.4586 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2860 | train 3.157429 | val 3.545726 | tok/s 2628874.11 | norm 0.4806 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2861 | train 3.224708 | val 3.545726 | tok/s 2632305.59 | norm 0.4926 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2862 | train 3.165664 | val 3.545726 | tok/s 2628682.81 | norm 0.4805 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2863 | train 3.240221 | val 3.545726 | tok/s 2630900.89 | norm 0.4806 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2864 | train 3.206968 | val 3.545726 | tok/s 2629939.73 | norm 0.4961 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2865 | train 3.312814 | val 3.545726 | tok/s 2627478.68 | norm 0.4683 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2866 | train 3.162374 | val 3.545726 | tok/s 2631804.14 | norm 0.4974 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2867 | train 3.163710 | val 3.545726 | tok/s 2627883.86 | norm 0.5041 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2868 | train 3.227079 | val 3.545726 | tok/s 2630300.32 | norm 0.5060 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2869 | train 3.116335 | val 3.545726 | tok/s 2622533.05 | norm 0.5403 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2870 | train 3.026729 | val 3.545726 | tok/s 2630923.36 | norm 0.4748 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2871 | train 3.212347 | val 3.545726 | tok/s 2630463.33 | norm 0.4560 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2872 | train 3.205893 | val 3.545726 | tok/s 2533812.63 | norm 0.4549 | dt 206.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2873 | train 3.200640 | val 3.545726 | tok/s 2625912.34 | norm 0.4554 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2874 | train 3.143265 | val 3.545726 | tok/s 2626121.93 | norm 0.4509 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2875 | train 3.227155 | val 3.545726 | tok/s 2627992.57 | norm 0.5140 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2876 | train 3.157105 | val 3.545726 | tok/s 2629360.08 | norm 0.5311 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2877 | train 3.090704 | val 3.545726 | tok/s 2624749.56 | norm 0.5045 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2878 | train 3.087430 | val 3.545726 | tok/s 2628762.96 | norm 0.4565 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2879 | train 3.066096 | val 3.545726 | tok/s 2623730.37 | norm 0.4902 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2880 | train 3.288275 | val 3.545726 | tok/s 2627712.26 | norm 0.5489 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2881 | train 3.028410 | val 3.545726 | tok/s 2544339.37 | norm 0.4914 | dt 206.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2882 | train 3.172267 | val 3.545726 | tok/s 2587246.97 | norm 0.5035 | dt 202.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2883 | train 3.287198 | val 3.545726 | tok/s 2629174.58 | norm 0.5083 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2884 | train 3.228428 | val 3.545726 | tok/s 2630152.68 | norm 0.4976 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2885 | train 3.348349 | val 3.545726 | tok/s 2627852.39 | norm 0.5000 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2886 | train 3.101964 | val 3.545726 | tok/s 2615904.41 | norm 0.4847 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2887 | train 3.166806 | val 3.545726 | tok/s 2622192.78 | norm 0.5272 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2888 | train 3.096976 | val 3.545726 | tok/s 2627376.65 | norm 0.5378 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2889 | train 3.205708 | val 3.545726 | tok/s 2600004.74 | norm 0.4951 | dt 201.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2890 | train 3.145450 | val 3.545726 | tok/s 2626965.43 | norm 0.5746 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2891 | train 3.171544 | val 3.545726 | tok/s 2603396.34 | norm 0.5867 | dt 201.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2892 | train 3.172773 | val 3.545726 | tok/s 2625666.56 | norm 0.5493 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2893 | train 3.173604 | val 3.545726 | tok/s 2624550.51 | norm 0.6032 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2894 | train 3.129765 | val 3.545726 | tok/s 2622339.44 | norm 0.5618 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2895 | train 3.119821 | val 3.545726 | tok/s 2626400.52 | norm 0.5580 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2896 | train 3.270726 | val 3.545726 | tok/s 2619698.92 | norm 0.5242 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2897 | train 3.060050 | val 3.545726 | tok/s 2625394.88 | norm 0.4881 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2898 | train 3.246845 | val 3.545726 | tok/s 2627262.93 | norm 0.4782 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2899 | train 3.155382 | val 3.545726 | tok/s 2615546.68 | norm 0.4653 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2900 | train 3.333997 | val 3.545726 | tok/s 2618235.36 | norm 0.4947 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2901 | train 3.185739 | val 3.545726 | tok/s 2622251.89 | norm 0.5365 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2902 | train 3.174857 | val 3.545726 | tok/s 2618494.08 | norm 0.5335 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2903 | train 3.099429 | val 3.545726 | tok/s 2629044.75 | norm 0.5080 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2904 | train 3.226905 | val 3.545726 | tok/s 2625686.36 | norm 0.5473 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2905 | train 3.139225 | val 3.545726 | tok/s 2628599.27 | norm 0.5732 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2906 | train 3.054774 | val 3.545726 | tok/s 2615913.65 | norm 0.5327 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2907 | train 3.150711 | val 3.545726 | tok/s 2621724.30 | norm 0.5111 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2908 | train 3.204724 | val 3.545726 | tok/s 2626740.17 | norm 0.5341 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2909 | train 3.257720 | val 3.545726 | tok/s 2628870.65 | norm 0.4897 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2910 | train 3.196673 | val 3.545726 | tok/s 2627183.81 | norm 0.4886 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2911 | train 3.112962 | val 3.545726 | tok/s 2625953.69 | norm 0.4679 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2912 | train 3.210751 | val 3.545726 | tok/s 2625457.28 | norm 0.5004 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2913 | train 3.201583 | val 3.545726 | tok/s 2623507.55 | norm 0.4836 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2914 | train 3.147168 | val 3.545726 | tok/s 2623537.98 | norm 0.4755 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2915 | train 3.210031 | val 3.545726 | tok/s 2628161.09 | norm 0.4913 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2916 | train 3.181574 | val 3.545726 | tok/s 2623997.20 | norm 0.4596 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2917 | train 3.227024 | val 3.545726 | tok/s 2624127.11 | norm 0.5115 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2918 | train 3.182213 | val 3.545726 | tok/s 2633279.18 | norm 0.5499 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2919 | train 3.265334 | val 3.545726 | tok/s 2629926.54 | norm 0.5635 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2920 | train 3.153750 | val 3.545726 | tok/s 2630706.16 | norm 0.5668 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2921 | train 3.160470 | val 3.545726 | tok/s 2611531.41 | norm 0.4547 | dt 200.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2922 | train 3.222258 | val 3.545726 | tok/s 2623842.50 | norm 0.4480 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2923 | train 3.186280 | val 3.545726 | tok/s 2627467.75 | norm 0.5517 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2924 | train 3.200465 | val 3.545726 | tok/s 2626186.25 | norm 0.5350 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2925 | train 3.182524 | val 3.545726 | tok/s 2628287.83 | norm 0.4896 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2926 | train 3.150391 | val 3.545726 | tok/s 2618818.24 | norm 0.5035 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2927 | train 3.111771 | val 3.545726 | tok/s 2622144.33 | norm 0.5774 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2928 | train 3.195515 | val 3.545726 | tok/s 2622020.11 | norm 0.5994 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2929 | train 3.206545 | val 3.545726 | tok/s 2625323.87 | norm 0.5023 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2930 | train 3.192608 | val 3.545726 | tok/s 2627608.13 | norm 0.4667 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2931 | train 3.279398 | val 3.545726 | tok/s 2623454.01 | norm 0.5042 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2932 | train 3.092889 | val 3.545726 | tok/s 2626181.68 | norm 0.5116 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2933 | train 3.015737 | val 3.545726 | tok/s 2628367.94 | norm 0.5202 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2934 | train 3.321936 | val 3.545726 | tok/s 2621974.34 | norm 0.5807 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2935 | train 3.227596 | val 3.545726 | tok/s 2624245.21 | norm 0.4529 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2936 | train 3.057181 | val 3.545726 | tok/s 2623918.54 | norm 0.4237 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2937 | train 3.100396 | val 3.545726 | tok/s 2621947.03 | norm 0.4926 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2938 | train 3.109555 | val 3.545726 | tok/s 2627745.22 | norm 0.5388 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2939 | train 3.225461 | val 3.545726 | tok/s 2625700.15 | norm 0.5607 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2940 | train 3.073105 | val 3.545726 | tok/s 2627717.53 | norm 0.5454 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2941 | train 3.063297 | val 3.545726 | tok/s 2621594.60 | norm 0.5335 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2942 | train 3.140878 | val 3.545726 | tok/s 2622659.60 | norm 0.5555 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2943 | train 3.046021 | val 3.545726 | tok/s 2628877.75 | norm 0.5209 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2944 | train 3.189635 | val 3.545726 | tok/s 2626203.16 | norm 0.4740 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2945 | train 3.234862 | val 3.545726 | tok/s 2626935.43 | norm 0.4466 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2946 | train 3.156417 | val 3.545726 | tok/s 2620959.38 | norm 0.4456 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2947 | train 3.171573 | val 3.545726 | tok/s 2620223.22 | norm 0.4655 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2948 | train 3.097062 | val 3.545726 | tok/s 2626370.43 | norm 0.4826 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2949 | train 3.075580 | val 3.545726 | tok/s 2626796.51 | norm 0.5011 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2950 | train 3.141905 | val 3.545726 | tok/s 2627397.64 | norm 0.5491 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2951 | train 3.166334 | val 3.545726 | tok/s 2626773.87 | norm 0.5592 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2952 | train 3.085772 | val 3.545726 | tok/s 2625461.94 | norm 0.5041 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2953 | train 3.016385 | val 3.545726 | tok/s 2629508.32 | norm 0.5219 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2954 | train 3.109636 | val 3.545726 | tok/s 2628159.56 | norm 0.5573 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2955 | train 3.153399 | val 3.545726 | tok/s 2630506.13 | norm 0.4982 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2956 | train 3.120980 | val 3.545726 | tok/s 2624068.48 | norm 0.5184 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2957 | train 3.257799 | val 3.545726 | tok/s 2626802.68 | norm 0.5316 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2958 | train 3.218205 | val 3.545726 | tok/s 2628178.24 | norm 0.5069 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2959 | train 3.058082 | val 3.545726 | tok/s 2627996.38 | norm 0.4574 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2960 | train 3.201823 | val 3.545726 | tok/s 2630386.44 | norm 0.4423 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2961 | train 3.193903 | val 3.545726 | tok/s 2628660.58 | norm 0.4411 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2962 | train 3.059139 | val 3.545726 | tok/s 2628655.30 | norm 0.5448 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2963 | train 3.239704 | val 3.545726 | tok/s 2627197.78 | norm 0.5611 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2964 | train 3.122601 | val 3.545726 | tok/s 2624181.71 | norm 0.5231 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2965 | train 3.250337 | val 3.545726 | tok/s 2626639.23 | norm 0.5025 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2966 | train 3.189559 | val 3.545726 | tok/s 2623862.63 | norm 0.5523 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2967 | train 3.207917 | val 3.545726 | tok/s 2626501.28 | norm 0.5388 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2968 | train 3.168205 | val 3.545726 | tok/s 2630606.22 | norm 0.4922 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2969 | train 3.078609 | val 3.545726 | tok/s 2625524.07 | norm 0.4842 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2970 | train 3.250487 | val 3.545726 | tok/s 2629981.43 | norm 0.5122 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2971 | train 3.137032 | val 3.545726 | tok/s 2623637.36 | norm 0.5842 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2972 | train 3.141899 | val 3.545726 | tok/s 2624650.07 | norm 0.5824 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2973 | train 2.988401 | val 3.545726 | tok/s 2633401.96 | norm 0.5962 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2974 | train 3.362120 | val 3.545726 | tok/s 2625626.18 | norm 0.6366 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2975 | train 3.155617 | val 3.545726 | tok/s 2632097.42 | norm 0.5525 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2976 | train 3.125432 | val 3.545726 | tok/s 2628303.59 | norm 0.4898 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2977 | train 3.251993 | val 3.545726 | tok/s 2628020.31 | norm 0.4521 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2978 | train 3.272621 | val 3.545726 | tok/s 2629582.67 | norm 0.4970 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2979 | train 3.188996 | val 3.545726 | tok/s 2631313.98 | norm 0.4973 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2980 | train 3.118495 | val 3.545726 | tok/s 2628526.86 | norm 0.5148 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2981 | train 3.200984 | val 3.545726 | tok/s 2624392.98 | norm 0.5195 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2982 | train 3.013401 | val 3.545726 | tok/s 2627814.24 | norm 0.4426 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2983 | train 3.199507 | val 3.545726 | tok/s 2632353.57 | norm 0.5065 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2984 | train 3.221339 | val 3.545726 | tok/s 2628911.67 | norm 0.5158 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2985 | train 3.202364 | val 3.545726 | tok/s 2626023.45 | norm 0.4620 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2986 | train 3.242521 | val 3.545726 | tok/s 2624404.33 | norm 0.4483 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2987 | train 3.244921 | val 3.545726 | tok/s 2627551.70 | norm 0.4662 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2988 | train 3.158385 | val 3.545726 | tok/s 2630984.34 | norm 0.4907 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2989 | train 3.181387 | val 3.545726 | tok/s 2630646.83 | norm 0.4666 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2990 | train 3.138670 | val 3.545726 | tok/s 2632221.17 | norm 0.4520 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2991 | train 3.095538 | val 3.545726 | tok/s 2625912.82 | norm 0.4227 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2992 | train 3.290133 | val 3.545726 | tok/s 2627007.11 | norm 0.4572 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2993 | train 3.111021 | val 3.545726 | tok/s 2629658.37 | norm 0.5032 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2994 | train 3.195685 | val 3.545726 | tok/s 2626074.52 | norm 0.4790 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2995 | train 3.170018 | val 3.545726 | tok/s 2629517.67 | norm 0.4859 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2996 | train 3.235790 | val 3.545726 | tok/s 2618429.89 | norm 0.4868 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2997 | train 3.163278 | val 3.545726 | tok/s 2625472.09 | norm 0.5197 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2998 | train 3.199605 | val 3.545726 | tok/s 2631731.54 | norm 0.5600 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 2999 | train 3.258646 | val 3.545726 | tok/s 2628580.84 | norm 0.5244 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 3000 | val_loss 3.528444 step 3000 | train 3.227739 | val 3.528444 | tok/s 2608916.16 | norm 0.4816 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3001 | train 3.294506 | val 3.528444 | tok/s 2629272.22 | norm 0.5098 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3002 | train 3.026258 | val 3.528444 | tok/s 2624598.88 | norm 0.5028 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3003 | train 3.186750 | val 3.528444 | tok/s 2627058.61 | norm 0.4871 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3004 | train 3.073205 | val 3.528444 | tok/s 2631335.27 | norm 0.4858 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3005 | train 3.114311 | val 3.528444 | tok/s 2627759.34 | norm 0.5518 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3006 | train 3.225008 | val 3.528444 | tok/s 2628115.94 | norm 0.5499 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3007 | train 3.072589 | val 3.528444 | tok/s 2623932.05 | norm 0.5161 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3008 | train 3.066893 | val 3.528444 | tok/s 2622216.53 | norm 0.5061 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3009 | train 3.048472 | val 3.528444 | tok/s 2631633.29 | norm 0.5135 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3010 | train 3.155399 | val 3.528444 | tok/s 2629320.56 | norm 0.4324 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3011 | train 3.015678 | val 3.528444 | tok/s 2627012.38 | norm 0.4380 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3012 | train 3.205341 | val 3.528444 | tok/s 2625050.81 | norm 0.4629 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3013 | train 3.259704 | val 3.528444 | tok/s 2623704.97 | norm 0.5034 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3014 | train 3.017412 | val 3.528444 | tok/s 2627745.45 | norm 0.4949 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3015 | train 3.220206 | val 3.528444 | tok/s 2628484.98 | norm 0.4566 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3016 | train 3.173625 | val 3.528444 | tok/s 2630373.56 | norm 0.4519 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3017 | train 3.279222 | val 3.528444 | tok/s 2626256.07 | norm 0.4598 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3018 | train 3.143549 | val 3.528444 | tok/s 2625112.52 | norm 0.4950 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3019 | train 3.140670 | val 3.528444 | tok/s 2631026.78 | norm 0.4770 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3020 | train 3.215615 | val 3.528444 | tok/s 2630623.01 | norm 0.4414 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3021 | train 3.271962 | val 3.528444 | tok/s 2624589.99 | norm 0.4548 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3022 | train 3.190944 | val 3.528444 | tok/s 2630250.58 | norm 0.5378 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3023 | train 3.191120 | val 3.528444 | tok/s 2630358.22 | norm 0.5305 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3024 | train 3.071824 | val 3.528444 | tok/s 2631702.97 | norm 0.5358 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3025 | train 3.039347 | val 3.528444 | tok/s 2629266.49 | norm 0.5125 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3026 | train 3.164949 | val 3.528444 | tok/s 2632577.79 | norm 0.4633 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3027 | train 3.061080 | val 3.528444 | tok/s 2628845.02 | norm 0.5397 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3028 | train 3.156689 | val 3.528444 | tok/s 2627936.61 | norm 0.5474 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3029 | train 3.287357 | val 3.528444 | tok/s 2628723.17 | norm 0.5670 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3030 | train 3.302162 | val 3.528444 | tok/s 2629145.43 | norm 0.5059 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3031 | train 3.097245 | val 3.528444 | tok/s 2631982.93 | norm 0.5092 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3032 | train 3.217967 | val 3.528444 | tok/s 2625583.87 | norm 0.5267 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3033 | train 3.221030 | val 3.528444 | tok/s 2626085.98 | norm 0.5494 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3034 | train 3.202450 | val 3.528444 | tok/s 2630649.88 | norm 0.5931 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3035 | train 3.210288 | val 3.528444 | tok/s 2625564.49 | norm 0.5088 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3036 | train 3.178488 | val 3.528444 | tok/s 2629958.69 | norm 0.5119 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3037 | train 3.195834 | val 3.528444 | tok/s 2626687.77 | norm 0.5646 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3038 | train 3.083486 | val 3.528444 | tok/s 2628888.80 | norm 0.4766 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3039 | train 3.318836 | val 3.528444 | tok/s 2630462.99 | norm 0.5097 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3040 | train 3.136756 | val 3.528444 | tok/s 2631347.88 | norm 0.5535 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3041 | train 3.366338 | val 3.528444 | tok/s 2632869.13 | norm 0.5157 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3042 | train 3.144135 | val 3.528444 | tok/s 2631325.85 | norm 0.4371 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3043 | train 3.227726 | val 3.528444 | tok/s 2628242.33 | norm 0.4879 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3044 | train 3.048950 | val 3.528444 | tok/s 2630602.39 | norm 0.5276 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3045 | train 3.163774 | val 3.528444 | tok/s 2628172.55 | norm 0.5022 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3046 | train 3.464734 | val 3.528444 | tok/s 2629708.42 | norm 0.6690 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3047 | train 3.306386 | val 3.528444 | tok/s 2609979.23 | norm 0.4520 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3048 | train 3.194467 | val 3.528444 | tok/s 2627432.78 | norm 0.4447 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3049 | train 3.059818 | val 3.528444 | tok/s 2631114.65 | norm 0.5209 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3050 | train 3.115782 | val 3.528444 | tok/s 2626879.70 | norm 0.5091 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3051 | train 3.167981 | val 3.528444 | tok/s 2629780.39 | norm 0.5357 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3052 | train 3.169509 | val 3.528444 | tok/s 2627869.81 | norm 0.5311 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3053 | train 3.041906 | val 3.528444 | tok/s 2629343.29 | norm 0.5321 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3054 | train 3.133753 | val 3.528444 | tok/s 2621155.83 | norm 0.5575 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3055 | train 3.125031 | val 3.528444 | tok/s 2624326.39 | norm 0.5404 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3056 | train 3.148591 | val 3.528444 | tok/s 2631141.04 | norm 0.5226 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3057 | train 3.091175 | val 3.528444 | tok/s 2627287.13 | norm 0.5362 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3058 | train 3.325242 | val 3.528444 | tok/s 2628068.09 | norm 0.6018 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3059 | train 3.184218 | val 3.528444 | tok/s 2622846.24 | norm 0.5744 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3060 | train 3.118666 | val 3.528444 | tok/s 2624187.43 | norm 0.4608 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3061 | train 3.056450 | val 3.528444 | tok/s 2623880.44 | norm 0.4425 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3062 | train 3.196944 | val 3.528444 | tok/s 2619158.39 | norm 0.4943 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3063 | train 3.082315 | val 3.528444 | tok/s 2621692.59 | norm 0.4549 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3064 | train 3.190202 | val 3.528444 | tok/s 2621004.07 | norm 0.4620 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3065 | train 3.158982 | val 3.528444 | tok/s 2625798.81 | norm 0.4961 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3066 | train 3.126673 | val 3.528444 | tok/s 2629383.80 | norm 0.4472 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3067 | train 3.026362 | val 3.528444 | tok/s 2628233.17 | norm 0.4813 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3068 | train 3.156990 | val 3.528444 | tok/s 2629111.02 | norm 0.5276 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3069 | train 3.097092 | val 3.528444 | tok/s 2628437.12 | norm 0.4815 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3070 | train 3.147761 | val 3.528444 | tok/s 2624718.75 | norm 0.4606 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3071 | train 3.131557 | val 3.528444 | tok/s 2625260.98 | norm 0.5104 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3072 | train 3.256135 | val 3.528444 | tok/s 2625320.20 | norm 0.5185 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3073 | train 3.098189 | val 3.528444 | tok/s 2631488.89 | norm 0.4814 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3074 | train 3.080593 | val 3.528444 | tok/s 2631362.98 | norm 0.5419 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3075 | train 3.144619 | val 3.528444 | tok/s 2628277.25 | norm 0.5401 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3076 | train 3.200724 | val 3.528444 | tok/s 2633699.57 | norm 0.5187 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3077 | train 3.147340 | val 3.528444 | tok/s 2631014.55 | norm 0.5295 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3078 | train 3.264488 | val 3.528444 | tok/s 2625981.75 | norm 0.5253 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3079 | train 3.089778 | val 3.528444 | tok/s 2626712.57 | norm 0.5029 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3080 | train 3.180256 | val 3.528444 | tok/s 2630243.15 | norm 0.5205 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3081 | train 3.177998 | val 3.528444 | tok/s 2627320.11 | norm 0.5532 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3082 | train 3.141623 | val 3.528444 | tok/s 2465596.19 | norm 0.5198 | dt 212.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3083 | train 3.192019 | val 3.528444 | tok/s 2630515.85 | norm 0.5263 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3084 | train 3.167247 | val 3.528444 | tok/s 2623471.25 | norm 0.5663 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3085 | train 3.256385 | val 3.528444 | tok/s 2627819.04 | norm 0.5805 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3086 | train 3.250959 | val 3.528444 | tok/s 2629991.67 | norm 0.5568 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3087 | train 3.161088 | val 3.528444 | tok/s 2630135.16 | norm 0.5000 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3088 | train 3.232571 | val 3.528444 | tok/s 2631782.89 | norm 0.5613 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3089 | train 3.224277 | val 3.528444 | tok/s 2628106.52 | norm 0.5576 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3090 | train 3.140548 | val 3.528444 | tok/s 2628018.00 | norm 0.5127 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3091 | train 3.202907 | val 3.528444 | tok/s 2628958.11 | norm 0.4910 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3092 | train 3.276548 | val 3.528444 | tok/s 2629151.80 | norm 0.4693 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3093 | train 3.362894 | val 3.528444 | tok/s 2627417.70 | norm 0.5236 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3094 | train 3.191883 | val 3.528444 | tok/s 2623629.38 | norm 0.5714 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3095 | train 3.170170 | val 3.528444 | tok/s 2626589.69 | norm 0.5868 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3096 | train 3.104045 | val 3.528444 | tok/s 2631821.23 | norm 0.5595 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3097 | train 3.100833 | val 3.528444 | tok/s 2629158.10 | norm 0.5421 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3098 | train 3.214285 | val 3.528444 | tok/s 2627387.95 | norm 0.5334 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3099 | train 3.175349 | val 3.528444 | tok/s 2618855.36 | norm 0.4929 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3100 | train 3.247170 | val 3.528444 | tok/s 2624420.21 | norm 0.4617 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3101 | train 3.153843 | val 3.528444 | tok/s 2631482.32 | norm 0.4524 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3102 | train 3.100903 | val 3.528444 | tok/s 2627516.17 | norm 0.4924 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3103 | train 3.256310 | val 3.528444 | tok/s 2629905.36 | norm 0.5036 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3104 | train 3.051390 | val 3.528444 | tok/s 2619357.43 | norm 0.4577 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3105 | train 3.127507 | val 3.528444 | tok/s 2623223.37 | norm 0.4795 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3106 | train 3.219315 | val 3.528444 | tok/s 2631395.37 | norm 0.4901 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3107 | train 3.172206 | val 3.528444 | tok/s 2583466.16 | norm 0.4658 | dt 202.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3108 | train 2.974726 | val 3.528444 | tok/s 2630591.01 | norm 0.4921 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3109 | train 3.153765 | val 3.528444 | tok/s 2629267.38 | norm 0.5815 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3110 | train 3.274985 | val 3.528444 | tok/s 2628677.83 | norm 0.5953 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3111 | train 3.214439 | val 3.528444 | tok/s 2632644.69 | norm 0.5757 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3112 | train 3.179770 | val 3.528444 | tok/s 2629657.30 | norm 0.5418 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3113 | train 3.180311 | val 3.528444 | tok/s 2629608.02 | norm 0.4830 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3114 | train 3.192458 | val 3.528444 | tok/s 2627569.80 | norm 0.5095 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3115 | train 3.197157 | val 3.528444 | tok/s 2622296.69 | norm 0.4846 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3116 | train 3.185229 | val 3.528444 | tok/s 2630399.87 | norm 0.4682 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3117 | train 3.030598 | val 3.528444 | tok/s 2629818.93 | norm 0.5518 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3118 | train 3.237584 | val 3.528444 | tok/s 2633066.46 | norm 0.5180 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3119 | train 3.100530 | val 3.528444 | tok/s 2629591.60 | norm 0.5009 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3120 | train 3.172631 | val 3.528444 | tok/s 2629553.57 | norm 0.4747 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3121 | train 3.169140 | val 3.528444 | tok/s 2634048.81 | norm 0.4897 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3122 | train 3.139269 | val 3.528444 | tok/s 2629378.04 | norm 0.5344 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3123 | train 3.190629 | val 3.528444 | tok/s 2630493.33 | norm 0.5138 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3124 | train 3.114393 | val 3.528444 | tok/s 2629008.39 | norm 0.5099 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3125 | train 3.118771 | val 3.528444 | tok/s 2626109.00 | norm 0.4530 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3126 | train 3.140693 | val 3.528444 | tok/s 2621087.75 | norm 0.4227 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3127 | train 3.189743 | val 3.528444 | tok/s 2624558.55 | norm 0.4734 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3128 | train 3.080790 | val 3.528444 | tok/s 2623380.20 | norm 0.4684 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3129 | train 3.268361 | val 3.528444 | tok/s 2624620.45 | norm 0.4915 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3130 | train 3.187851 | val 3.528444 | tok/s 2628388.25 | norm 0.4661 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3131 | train 3.122112 | val 3.528444 | tok/s 2632911.08 | norm 0.4877 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3132 | train 3.014711 | val 3.528444 | tok/s 2623475.36 | norm 0.5148 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3133 | train 3.058328 | val 3.528444 | tok/s 2626626.98 | norm 0.5363 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3134 | train 3.129204 | val 3.528444 | tok/s 2625722.09 | norm 0.5376 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3135 | train 3.293145 | val 3.528444 | tok/s 2628737.47 | norm 0.4872 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3136 | train 3.266561 | val 3.528444 | tok/s 2631234.75 | norm 0.4458 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3137 | train 3.078957 | val 3.528444 | tok/s 2629948.82 | norm 0.4581 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3138 | train 3.167172 | val 3.528444 | tok/s 2631166.65 | norm 0.4908 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3139 | train 3.179981 | val 3.528444 | tok/s 2627351.20 | norm 0.5435 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3140 | train 3.025145 | val 3.528444 | tok/s 2627848.48 | norm 0.5386 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3141 | train 3.044735 | val 3.528444 | tok/s 2626823.91 | norm 0.5029 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3142 | train 3.187355 | val 3.528444 | tok/s 2624919.65 | norm 0.4854 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3143 | train 3.102433 | val 3.528444 | tok/s 2629430.02 | norm 0.4602 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3144 | train 3.131655 | val 3.528444 | tok/s 2623914.00 | norm 0.4598 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3145 | train 3.214598 | val 3.528444 | tok/s 2625719.32 | norm 0.4981 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3146 | train 3.325422 | val 3.528444 | tok/s 2629664.87 | norm 0.5319 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3147 | train 3.077494 | val 3.528444 | tok/s 2631470.60 | norm 0.4869 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3148 | train 3.104759 | val 3.528444 | tok/s 2633480.34 | norm 0.4572 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3149 | train 3.167736 | val 3.528444 | tok/s 2629812.35 | norm 0.4902 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3150 | train 3.200559 | val 3.528444 | tok/s 2590612.21 | norm 0.4766 | dt 202.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3151 | train 3.118944 | val 3.528444 | tok/s 2630644.12 | norm 0.4850 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3152 | train 3.260875 | val 3.528444 | tok/s 2628751.58 | norm 0.5373 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3153 | train 3.136354 | val 3.528444 | tok/s 2626790.17 | norm 0.5374 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3154 | train 3.200508 | val 3.528444 | tok/s 2627342.85 | norm 0.5360 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3155 | train 3.187532 | val 3.528444 | tok/s 2624912.94 | norm 0.5971 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3156 | train 3.124806 | val 3.528444 | tok/s 2631723.01 | norm 0.5537 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3157 | train 3.119730 | val 3.528444 | tok/s 2626793.98 | norm 0.4991 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3158 | train 3.022636 | val 3.528444 | tok/s 2630762.25 | norm 0.5206 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3159 | train 3.125041 | val 3.528444 | tok/s 2628936.50 | norm 0.4887 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3160 | train 3.109220 | val 3.528444 | tok/s 2627757.51 | norm 0.5009 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3161 | train 3.243829 | val 3.528444 | tok/s 2627494.75 | norm 0.5212 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3162 | train 3.132561 | val 3.528444 | tok/s 2628179.23 | norm 0.5132 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3163 | train 3.073591 | val 3.528444 | tok/s 2630699.05 | norm 0.4696 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3164 | train 3.242148 | val 3.528444 | tok/s 2624833.69 | norm 0.4779 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3165 | train 3.068766 | val 3.528444 | tok/s 2626266.51 | norm 0.4686 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3166 | train 3.101279 | val 3.528444 | tok/s 2632023.84 | norm 0.4777 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3167 | train 3.089328 | val 3.528444 | tok/s 2631052.18 | norm 0.4448 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3168 | train 3.095622 | val 3.528444 | tok/s 2628963.20 | norm 0.4269 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3169 | train 3.145950 | val 3.528444 | tok/s 2625237.13 | norm 0.4789 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3170 | train 3.161598 | val 3.528444 | tok/s 2628340.73 | norm 0.5104 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3171 | train 3.205132 | val 3.528444 | tok/s 2628116.02 | norm 0.5533 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3172 | train 3.148025 | val 3.528444 | tok/s 2624945.37 | norm 0.5729 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3173 | train 3.048165 | val 3.528444 | tok/s 2627829.95 | norm 0.5719 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3174 | train 3.104877 | val 3.528444 | tok/s 2627699.51 | norm 0.4186 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3175 | train 3.192458 | val 3.528444 | tok/s 2627085.71 | norm 0.4561 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3176 | train 3.158247 | val 3.528444 | tok/s 2629012.07 | norm 0.5247 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3177 | train 3.260960 | val 3.528444 | tok/s 2625979.12 | norm 0.5446 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3178 | train 3.194654 | val 3.528444 | tok/s 2630659.90 | norm 0.5456 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3179 | train 3.179596 | val 3.528444 | tok/s 2629780.95 | norm 0.5001 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3180 | train 3.081681 | val 3.528444 | tok/s 2629010.26 | norm 0.5183 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3181 | train 3.156535 | val 3.528444 | tok/s 2628678.54 | norm 0.4686 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3182 | train 3.184998 | val 3.528444 | tok/s 2626877.16 | norm 0.5055 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3183 | train 3.126010 | val 3.528444 | tok/s 2629865.30 | norm 0.5587 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3184 | train 3.154626 | val 3.528444 | tok/s 2627682.07 | norm 0.5351 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3185 | train 3.131878 | val 3.528444 | tok/s 2625522.36 | norm 0.4901 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3186 | train 3.051152 | val 3.528444 | tok/s 2627664.65 | norm 0.4920 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3187 | train 3.192450 | val 3.528444 | tok/s 2629027.82 | norm 0.5469 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3188 | train 3.155623 | val 3.528444 | tok/s 2629971.93 | norm 0.5106 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3189 | train 3.134583 | val 3.528444 | tok/s 2626178.22 | norm 0.6179 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3190 | train 3.176611 | val 3.528444 | tok/s 2627153.56 | norm 0.5860 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3191 | train 3.241130 | val 3.528444 | tok/s 2626355.92 | norm 0.4958 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3192 | train 3.190134 | val 3.528444 | tok/s 2623804.66 | norm 0.5130 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3193 | train 3.108360 | val 3.528444 | tok/s 2631418.15 | norm 0.4256 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3194 | train 3.169724 | val 3.528444 | tok/s 2628996.96 | norm 0.4803 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3195 | train 3.139433 | val 3.528444 | tok/s 2626641.70 | norm 0.5382 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3196 | train 3.194433 | val 3.528444 | tok/s 2630241.12 | norm 0.4823 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3197 | train 3.230820 | val 3.528444 | tok/s 2627052.16 | norm 0.5300 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3198 | train 3.114749 | val 3.528444 | tok/s 2627844.21 | norm 0.5302 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3199 | train 3.101595 | val 3.528444 | tok/s 2625934.44 | norm 0.4814 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3200 | train 3.054330 | val 3.528444 | tok/s 2625116.42 | norm 0.5301 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3201 | train 3.232913 | val 3.528444 | tok/s 2631017.19 | norm 0.4869 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3202 | train 3.116542 | val 3.528444 | tok/s 2622308.50 | norm 0.4432 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3203 | train 3.074293 | val 3.528444 | tok/s 2628649.20 | norm 0.5099 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3204 | train 3.048748 | val 3.528444 | tok/s 2621502.67 | norm 0.5183 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3205 | train 3.112373 | val 3.528444 | tok/s 2629055.62 | norm 0.5062 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3206 | train 3.212516 | val 3.528444 | tok/s 2612489.57 | norm 0.5115 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3207 | train 3.192138 | val 3.528444 | tok/s 2622098.19 | norm 0.4880 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3208 | train 3.095989 | val 3.528444 | tok/s 2624644.32 | norm 0.5104 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3209 | train 3.039689 | val 3.528444 | tok/s 2624238.76 | norm 0.4928 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3210 | train 3.222946 | val 3.528444 | tok/s 2609408.11 | norm 0.5413 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3211 | train 3.127369 | val 3.528444 | tok/s 2624023.93 | norm 0.5352 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3212 | train 3.113805 | val 3.528444 | tok/s 2621600.08 | norm 0.4777 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3213 | train 3.217025 | val 3.528444 | tok/s 2626779.11 | norm 0.4473 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3214 | train 3.113865 | val 3.528444 | tok/s 2616652.89 | norm 0.5011 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3215 | train 3.061441 | val 3.528444 | tok/s 2626843.63 | norm 0.5436 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3216 | train 3.112909 | val 3.528444 | tok/s 2626834.03 | norm 0.4691 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3217 | train 3.087572 | val 3.528444 | tok/s 2621276.38 | norm 0.5557 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3218 | train 3.121628 | val 3.528444 | tok/s 2625667.48 | norm 0.5395 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3219 | train 3.080513 | val 3.528444 | tok/s 2628063.85 | norm 0.4821 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3220 | train 3.152596 | val 3.528444 | tok/s 2626168.01 | norm 0.5053 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3221 | train 3.060524 | val 3.528444 | tok/s 2625931.26 | norm 0.5374 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3222 | train 3.094505 | val 3.528444 | tok/s 2622979.90 | norm 0.5349 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3223 | train 3.356085 | val 3.528444 | tok/s 2625524.30 | norm 0.5161 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3224 | train 3.192357 | val 3.528444 | tok/s 2624421.92 | norm 0.4902 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3225 | train 3.217746 | val 3.528444 | tok/s 2630442.78 | norm 0.5235 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3226 | train 3.119011 | val 3.528444 | tok/s 2626960.26 | norm 0.6298 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3227 | train 3.109255 | val 3.528444 | tok/s 2625820.76 | norm 0.5524 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3228 | train 3.149122 | val 3.528444 | tok/s 2635074.68 | norm 0.5329 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3229 | train 3.117857 | val 3.528444 | tok/s 2627887.90 | norm 0.4995 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3230 | train 3.180959 | val 3.528444 | tok/s 2629051.21 | norm 0.5147 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3231 | train 3.204456 | val 3.528444 | tok/s 2607944.50 | norm 0.5124 | dt 201.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3232 | train 3.023849 | val 3.528444 | tok/s 2625280.19 | norm 0.5051 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3233 | train 3.089752 | val 3.528444 | tok/s 2629182.85 | norm 0.5187 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3234 | train 3.160148 | val 3.528444 | tok/s 2623896.81 | norm 0.5605 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3235 | train 3.113782 | val 3.528444 | tok/s 2631758.61 | norm 0.5668 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3236 | train 3.182792 | val 3.528444 | tok/s 2626808.84 | norm 0.5222 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3237 | train 3.086977 | val 3.528444 | tok/s 2628544.43 | norm 0.5108 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3238 | train 3.083064 | val 3.528444 | tok/s 2631026.62 | norm 0.5221 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3239 | train 3.048377 | val 3.528444 | tok/s 2626763.78 | norm 0.4812 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3240 | train 3.188179 | val 3.528444 | tok/s 2626016.66 | norm 0.5184 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3241 | train 3.098440 | val 3.528444 | tok/s 2628354.99 | norm 0.5240 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3242 | train 3.236934 | val 3.528444 | tok/s 2628095.89 | norm 0.4658 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3243 | train 3.242032 | val 3.528444 | tok/s 2630058.45 | norm 0.4839 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3244 | train 3.267370 | val 3.528444 | tok/s 2625478.80 | norm 0.5563 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3245 | train 3.019505 | val 3.528444 | tok/s 2626327.79 | norm 0.4776 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3246 | train 3.233168 | val 3.528444 | tok/s 2623375.42 | norm 0.4525 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3247 | train 3.108846 | val 3.528444 | tok/s 2560730.27 | norm 0.5041 | dt 204.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3248 | train 3.170814 | val 3.528444 | tok/s 2627856.05 | norm 0.5239 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3249 | train 3.190389 | val 3.528444 | tok/s 2524482.21 | norm 0.5193 | dt 207.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 3250 | val_loss 3.510341 step 3250 | train 3.182691 | val 3.510341 | tok/s 2610675.53 | norm 0.5066 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3251 | train 3.045713 | val 3.510341 | tok/s 2628606.69 | norm 0.5121 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3252 | train 3.164148 | val 3.510341 | tok/s 2623247.85 | norm 0.4788 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3253 | train 3.178535 | val 3.510341 | tok/s 2622649.84 | norm 0.4609 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3254 | train 3.230125 | val 3.510341 | tok/s 2632515.70 | norm 0.5041 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3255 | train 3.081990 | val 3.510341 | tok/s 2624410.36 | norm 0.5327 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3256 | train 3.037260 | val 3.510341 | tok/s 2626937.01 | norm 0.5370 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3257 | train 3.116072 | val 3.510341 | tok/s 2623449.44 | norm 0.5667 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3258 | train 3.156947 | val 3.510341 | tok/s 2625606.93 | norm 0.5524 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3259 | train 3.040404 | val 3.510341 | tok/s 2627160.38 | norm 0.5010 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3260 | train 3.082457 | val 3.510341 | tok/s 2627204.03 | norm 0.4843 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3261 | train 3.087190 | val 3.510341 | tok/s 2629652.70 | norm 0.4981 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3262 | train 3.168066 | val 3.510341 | tok/s 2621892.34 | norm 0.4924 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3263 | train 3.054216 | val 3.510341 | tok/s 2625864.69 | norm 0.4650 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3264 | train 3.262575 | val 3.510341 | tok/s 2629326.77 | norm 0.4803 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3265 | train 3.230054 | val 3.510341 | tok/s 2555709.09 | norm 0.4975 | dt 205.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3266 | train 3.044456 | val 3.510341 | tok/s 2627608.77 | norm 0.4304 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3267 | train 2.990135 | val 3.510341 | tok/s 2625505.85 | norm 0.4614 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3268 | train 3.055101 | val 3.510341 | tok/s 2396134.59 | norm 0.4702 | dt 218.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3269 | train 3.062146 | val 3.510341 | tok/s 2626235.26 | norm 0.4807 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3270 | train 3.324205 | val 3.510341 | tok/s 2628096.36 | norm 0.5068 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3271 | train 3.083340 | val 3.510341 | tok/s 2631027.90 | norm 0.4967 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3272 | train 3.155438 | val 3.510341 | tok/s 2621616.15 | norm 0.5236 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3273 | train 3.061013 | val 3.510341 | tok/s 2625055.32 | norm 0.4823 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3274 | train 3.167888 | val 3.510341 | tok/s 2630774.58 | norm 0.5056 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3275 | train 3.104824 | val 3.510341 | tok/s 2627237.05 | norm 0.5268 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3276 | train 3.202407 | val 3.510341 | tok/s 2631066.34 | norm 0.5043 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3277 | train 3.109308 | val 3.510341 | tok/s 2628562.81 | norm 0.4636 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3278 | train 3.222041 | val 3.510341 | tok/s 2628801.33 | norm 0.4299 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3279 | train 3.089787 | val 3.510341 | tok/s 2633358.21 | norm 0.4634 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3280 | train 3.179904 | val 3.510341 | tok/s 2627518.26 | norm 0.4599 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3281 | train 2.985875 | val 3.510341 | tok/s 2628251.18 | norm 0.4514 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3282 | train 3.158934 | val 3.510341 | tok/s 2622765.77 | norm 0.4517 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3283 | train 3.118627 | val 3.510341 | tok/s 2622972.51 | norm 0.5080 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3284 | train 3.204047 | val 3.510341 | tok/s 2631911.96 | norm 0.4939 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3285 | train 3.153964 | val 3.510341 | tok/s 2629243.79 | norm 0.5198 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3286 | train 3.113541 | val 3.510341 | tok/s 2630385.90 | norm 0.4896 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3287 | train 3.032118 | val 3.510341 | tok/s 2627498.51 | norm 0.4533 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3288 | train 3.055279 | val 3.510341 | tok/s 2625994.80 | norm 0.4833 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3289 | train 3.168831 | val 3.510341 | tok/s 2628807.05 | norm 0.4611 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3290 | train 3.109371 | val 3.510341 | tok/s 2630054.30 | norm 0.4560 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3291 | train 3.123954 | val 3.510341 | tok/s 2627260.50 | norm 0.4861 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3292 | train 3.152492 | val 3.510341 | tok/s 2626997.93 | norm 0.5259 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3293 | train 3.136001 | val 3.510341 | tok/s 2625524.53 | norm 0.4766 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3294 | train 3.115758 | val 3.510341 | tok/s 2627337.72 | norm 0.5370 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3295 | train 3.242069 | val 3.510341 | tok/s 2625122.10 | norm 0.5809 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3296 | train 3.221176 | val 3.510341 | tok/s 2629937.72 | norm 0.4536 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3297 | train 3.086938 | val 3.510341 | tok/s 2606289.21 | norm 0.4216 | dt 201.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3298 | train 3.176517 | val 3.510341 | tok/s 2628994.71 | norm 0.5308 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3299 | train 3.221628 | val 3.510341 | tok/s 2629906.63 | norm 0.5491 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3300 | train 3.251454 | val 3.510341 | tok/s 2630255.43 | norm 0.5620 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3301 | train 3.285464 | val 3.510341 | tok/s 2631506.82 | norm 0.5612 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3302 | train 3.101513 | val 3.510341 | tok/s 2628442.43 | norm 0.5496 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3303 | train 3.071558 | val 3.510341 | tok/s 2624781.69 | norm 0.4991 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3304 | train 3.148815 | val 3.510341 | tok/s 2632598.53 | norm 0.5550 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3305 | train 3.100855 | val 3.510341 | tok/s 2629165.71 | norm 0.4738 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3306 | train 3.062538 | val 3.510341 | tok/s 2629592.53 | norm 0.4229 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3307 | train 3.178203 | val 3.510341 | tok/s 2628880.76 | norm 0.4351 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3308 | train 3.178355 | val 3.510341 | tok/s 2626932.13 | norm 0.4495 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3309 | train 3.092141 | val 3.510341 | tok/s 2633661.31 | norm 0.4493 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3310 | train 3.131487 | val 3.510341 | tok/s 2567409.57 | norm 0.4640 | dt 204.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3311 | train 3.201267 | val 3.510341 | tok/s 2615555.74 | norm 0.4905 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3312 | train 3.159478 | val 3.510341 | tok/s 2629259.01 | norm 0.4700 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3313 | train 3.049306 | val 3.510341 | tok/s 2611843.30 | norm 0.4847 | dt 200.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3314 | train 3.180170 | val 3.510341 | tok/s 2630671.24 | norm 0.4954 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3315 | train 3.099358 | val 3.510341 | tok/s 2629252.22 | norm 0.5104 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3316 | train 3.149271 | val 3.510341 | tok/s 2633837.09 | norm 0.5392 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3317 | train 3.185794 | val 3.510341 | tok/s 2623974.80 | norm 0.5186 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3318 | train 3.242104 | val 3.510341 | tok/s 2623301.00 | norm 0.4817 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3319 | train 2.953458 | val 3.510341 | tok/s 2634449.15 | norm 0.4737 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3320 | train 3.154654 | val 3.510341 | tok/s 2626290.14 | norm 0.5180 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3321 | train 3.181414 | val 3.510341 | tok/s 2628312.45 | norm 0.5382 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3322 | train 3.111421 | val 3.510341 | tok/s 2628524.87 | norm 0.4672 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3323 | train 3.166632 | val 3.510341 | tok/s 2630479.15 | norm 0.4426 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3324 | train 2.968640 | val 3.510341 | tok/s 2632146.84 | norm 0.4664 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3325 | train 3.236770 | val 3.510341 | tok/s 2625338.80 | norm 0.4434 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3326 | train 3.174868 | val 3.510341 | tok/s 2631155.67 | norm 0.4578 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3327 | train 3.175976 | val 3.510341 | tok/s 2619390.86 | norm 0.4587 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3328 | train 3.093419 | val 3.510341 | tok/s 2593718.68 | norm 0.5070 | dt 202.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3329 | train 3.085068 | val 3.510341 | tok/s 2631917.56 | norm 0.5396 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3330 | train 3.155586 | val 3.510341 | tok/s 2628116.72 | norm 0.5554 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3331 | train 3.195810 | val 3.510341 | tok/s 2627203.65 | norm 0.5794 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3332 | train 3.163119 | val 3.510341 | tok/s 2621488.56 | norm 0.5350 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3333 | train 3.169855 | val 3.510341 | tok/s 2628074.96 | norm 0.5061 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3334 | train 3.101947 | val 3.510341 | tok/s 2578894.76 | norm 0.5412 | dt 203.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3335 | train 3.190537 | val 3.510341 | tok/s 2579753.29 | norm 0.5183 | dt 203.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3336 | train 3.120331 | val 3.510341 | tok/s 2628692.55 | norm 0.4699 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3337 | train 3.132735 | val 3.510341 | tok/s 2628932.22 | norm 0.4861 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3338 | train 2.919721 | val 3.510341 | tok/s 2630181.04 | norm 0.5897 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3339 | train 3.160727 | val 3.510341 | tok/s 2631679.50 | norm 0.4620 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3340 | train 3.039442 | val 3.510341 | tok/s 2624685.61 | norm 0.4909 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3341 | train 3.170715 | val 3.510341 | tok/s 2625094.83 | norm 0.4838 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3342 | train 3.189436 | val 3.510341 | tok/s 2627411.34 | norm 0.4885 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3343 | train 3.143561 | val 3.510341 | tok/s 2628350.79 | norm 0.5051 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3344 | train 3.018021 | val 3.510341 | tok/s 2629449.18 | norm 0.5001 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3345 | train 3.135489 | val 3.510341 | tok/s 2624226.19 | norm 0.4926 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3346 | train 3.109135 | val 3.510341 | tok/s 2627760.30 | norm 0.4628 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3347 | train 3.110380 | val 3.510341 | tok/s 2626227.80 | norm 0.5294 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3348 | train 3.151458 | val 3.510341 | tok/s 2632366.04 | norm 0.5806 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3349 | train 3.208867 | val 3.510341 | tok/s 2628945.61 | norm 0.4971 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3350 | train 3.117402 | val 3.510341 | tok/s 2608367.83 | norm 0.4337 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3351 | train 3.246435 | val 3.510341 | tok/s 2634255.71 | norm 0.4578 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3352 | train 3.229988 | val 3.510341 | tok/s 2629377.10 | norm 0.5070 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3353 | train 3.144598 | val 3.510341 | tok/s 2629369.85 | norm 0.4712 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3354 | train 3.129252 | val 3.510341 | tok/s 2629716.32 | norm 0.5085 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3355 | train 3.048698 | val 3.510341 | tok/s 2626800.38 | norm 0.4878 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3356 | train 3.161636 | val 3.510341 | tok/s 2630313.44 | norm 0.5077 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3357 | train 3.112774 | val 3.510341 | tok/s 2623852.45 | norm 0.5298 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3358 | train 3.224758 | val 3.510341 | tok/s 2627993.18 | norm 0.4857 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3359 | train 3.172946 | val 3.510341 | tok/s 2619086.16 | norm 0.4824 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3360 | train 3.116990 | val 3.510341 | tok/s 2624607.30 | norm 0.4542 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3361 | train 3.108186 | val 3.510341 | tok/s 2625275.82 | norm 0.4274 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3362 | train 3.192695 | val 3.510341 | tok/s 2619791.25 | norm 0.4146 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3363 | train 3.157200 | val 3.510341 | tok/s 2491831.34 | norm 0.4684 | dt 210.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3364 | train 3.124870 | val 3.510341 | tok/s 2621299.04 | norm 0.5083 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3365 | train 3.176881 | val 3.510341 | tok/s 2618104.63 | norm 0.4902 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3366 | train 3.132444 | val 3.510341 | tok/s 2627681.33 | norm 0.5410 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3367 | train 3.067477 | val 3.510341 | tok/s 2498405.23 | norm 0.5520 | dt 209.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3368 | train 3.157188 | val 3.510341 | tok/s 2627388.58 | norm 0.5038 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3369 | train 3.153624 | val 3.510341 | tok/s 2623106.77 | norm 0.4829 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3370 | train 3.128972 | val 3.510341 | tok/s 2624295.48 | norm 0.5133 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3371 | train 3.120159 | val 3.510341 | tok/s 2629474.99 | norm 0.4867 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3372 | train 3.167080 | val 3.510341 | tok/s 2627714.04 | norm 0.4964 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3373 | train 3.169954 | val 3.510341 | tok/s 2633134.34 | norm 0.4815 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3374 | train 3.043490 | val 3.510341 | tok/s 2632942.75 | norm 0.4615 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3375 | train 3.103522 | val 3.510341 | tok/s 2629711.09 | norm 0.4998 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3376 | train 3.166707 | val 3.510341 | tok/s 2629977.83 | norm 0.4627 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3377 | train 3.387730 | val 3.510341 | tok/s 2626413.10 | norm 0.5673 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3378 | train 3.074227 | val 3.510341 | tok/s 2629684.89 | norm 0.4803 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3379 | train 3.258818 | val 3.510341 | tok/s 2627724.39 | norm 0.4406 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3380 | train 3.195391 | val 3.510341 | tok/s 2628026.04 | norm 0.5095 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3381 | train 3.170689 | val 3.510341 | tok/s 2627681.12 | norm 0.4658 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3382 | train 3.110256 | val 3.510341 | tok/s 2622708.67 | norm 0.4828 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3383 | train 3.099210 | val 3.510341 | tok/s 2631283.76 | norm 0.4579 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3384 | train 3.169564 | val 3.510341 | tok/s 2630405.22 | norm 0.4690 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3385 | train 3.168285 | val 3.510341 | tok/s 2627579.88 | norm 0.5434 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3386 | train 3.206009 | val 3.510341 | tok/s 2631564.76 | norm 0.5313 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3387 | train 3.291534 | val 3.510341 | tok/s 2625716.33 | norm 0.5119 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3388 | train 3.193429 | val 3.510341 | tok/s 2630143.21 | norm 0.5148 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3389 | train 3.039683 | val 3.510341 | tok/s 2627295.48 | norm 0.4839 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3390 | train 3.178067 | val 3.510341 | tok/s 2627690.65 | norm 0.4684 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3391 | train 3.090938 | val 3.510341 | tok/s 2623854.71 | norm 0.4917 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3392 | train 3.143805 | val 3.510341 | tok/s 2623633.86 | norm 0.4842 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3393 | train 3.199279 | val 3.510341 | tok/s 2631318.87 | norm 0.4574 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3394 | train 3.202301 | val 3.510341 | tok/s 2625076.64 | norm 0.4253 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3395 | train 3.121116 | val 3.510341 | tok/s 2629808.62 | norm 0.4659 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3396 | train 3.163131 | val 3.510341 | tok/s 2624657.48 | norm 0.4750 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3397 | train 3.140115 | val 3.510341 | tok/s 2626783.14 | norm 0.5195 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3398 | train 3.142868 | val 3.510341 | tok/s 2630869.22 | norm 0.5292 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3399 | train 2.954667 | val 3.510341 | tok/s 2624880.75 | norm 0.4895 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3400 | train 3.260875 | val 3.510341 | tok/s 2628352.33 | norm 0.4537 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3401 | train 3.179204 | val 3.510341 | tok/s 2619660.91 | norm 0.4434 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3402 | train 3.212841 | val 3.510341 | tok/s 2626088.64 | norm 0.4430 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3403 | train 3.138666 | val 3.510341 | tok/s 2633572.51 | norm 0.4745 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3404 | train 3.007497 | val 3.510341 | tok/s 2630378.30 | norm 0.5463 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3405 | train 3.133283 | val 3.510341 | tok/s 2632028.86 | norm 0.4943 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3406 | train 3.054096 | val 3.510341 | tok/s 2628170.17 | norm 0.4325 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3407 | train 3.055903 | val 3.510341 | tok/s 2627148.08 | norm 0.4601 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3408 | train 3.153016 | val 3.510341 | tok/s 2631356.54 | norm 0.5150 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3409 | train 3.143501 | val 3.510341 | tok/s 2628910.52 | norm 0.4766 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3410 | train 3.094605 | val 3.510341 | tok/s 2625498.97 | norm 0.4772 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3411 | train 3.212036 | val 3.510341 | tok/s 2626948.83 | norm 0.4776 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3412 | train 3.137284 | val 3.510341 | tok/s 2627088.79 | norm 0.5425 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3413 | train 3.098277 | val 3.510341 | tok/s 2630019.00 | norm 0.4960 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3414 | train 3.200728 | val 3.510341 | tok/s 2628289.82 | norm 0.4932 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3415 | train 3.062458 | val 3.510341 | tok/s 2621648.68 | norm 0.5094 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3416 | train 3.274452 | val 3.510341 | tok/s 2625341.68 | norm 0.4685 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3417 | train 3.122048 | val 3.510341 | tok/s 2627014.54 | norm 0.5105 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3418 | train 3.025142 | val 3.510341 | tok/s 2632718.01 | norm 0.5158 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3419 | train 3.172877 | val 3.510341 | tok/s 2630278.72 | norm 0.5235 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3420 | train 3.124490 | val 3.510341 | tok/s 2632490.67 | norm 0.5364 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3421 | train 3.083241 | val 3.510341 | tok/s 2621040.54 | norm 0.5721 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3422 | train 3.087904 | val 3.510341 | tok/s 2625701.13 | norm 0.5282 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3423 | train 3.059604 | val 3.510341 | tok/s 2630366.62 | norm 0.4819 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3424 | train 3.121994 | val 3.510341 | tok/s 2630617.40 | norm 0.5157 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3425 | train 3.113155 | val 3.510341 | tok/s 2631677.44 | norm 0.5110 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3426 | train 3.051920 | val 3.510341 | tok/s 2623871.17 | norm 0.5032 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3427 | train 3.013827 | val 3.510341 | tok/s 2625870.17 | norm 0.5187 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3428 | train 3.185447 | val 3.510341 | tok/s 2628664.70 | norm 0.5540 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3429 | train 3.126171 | val 3.510341 | tok/s 2627837.98 | norm 0.4791 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3430 | train 3.153770 | val 3.510341 | tok/s 2629861.33 | norm 0.4908 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3431 | train 3.290169 | val 3.510341 | tok/s 2627961.17 | norm 0.5148 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3432 | train 3.139642 | val 3.510341 | tok/s 2631089.60 | norm 0.4684 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3433 | train 3.168011 | val 3.510341 | tok/s 2629999.07 | norm 0.4212 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3434 | train 3.123548 | val 3.510341 | tok/s 2540246.38 | norm 0.4906 | dt 206.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3435 | train 3.087554 | val 3.510341 | tok/s 2612299.20 | norm 0.4806 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3436 | train 3.189129 | val 3.510341 | tok/s 2625400.01 | norm 0.4952 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3437 | train 3.040392 | val 3.510341 | tok/s 2628815.46 | norm 0.5076 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3438 | train 3.082957 | val 3.510341 | tok/s 2633518.32 | norm 0.5172 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3439 | train 3.112719 | val 3.510341 | tok/s 2625264.72 | norm 0.5648 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3440 | train 3.072198 | val 3.510341 | tok/s 2628831.74 | norm 0.5344 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3441 | train 3.134865 | val 3.510341 | tok/s 2630578.38 | norm 0.4984 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3442 | train 3.221283 | val 3.510341 | tok/s 2626678.74 | norm 0.4691 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3443 | train 3.168910 | val 3.510341 | tok/s 2629902.05 | norm 0.4846 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3444 | train 3.096990 | val 3.510341 | tok/s 2627517.32 | norm 0.5266 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3445 | train 3.051282 | val 3.510341 | tok/s 2631337.41 | norm 0.5243 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3446 | train 3.124893 | val 3.510341 | tok/s 2630077.08 | norm 0.4588 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3447 | train 3.045293 | val 3.510341 | tok/s 2625981.28 | norm 0.4875 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3448 | train 3.153781 | val 3.510341 | tok/s 2629078.35 | norm 0.4906 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3449 | train 3.203905 | val 3.510341 | tok/s 2626985.32 | norm 0.4755 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3450 | train 3.154874 | val 3.510341 | tok/s 2631577.24 | norm 0.4749 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3451 | train 3.159740 | val 3.510341 | tok/s 2479982.79 | norm 0.4615 | dt 211.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3452 | train 3.189180 | val 3.510341 | tok/s 2626734.39 | norm 0.4395 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3453 | train 3.213011 | val 3.510341 | tok/s 2628864.62 | norm 0.4688 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3454 | train 3.154465 | val 3.510341 | tok/s 2628808.38 | norm 0.4407 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3455 | train 3.049700 | val 3.510341 | tok/s 2628306.15 | norm 0.4192 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3456 | train 3.152480 | val 3.510341 | tok/s 2625686.93 | norm 0.4550 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3457 | train 3.109770 | val 3.510341 | tok/s 2627553.38 | norm 0.4866 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3458 | train 3.125291 | val 3.510341 | tok/s 2627094.22 | norm 0.5196 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3459 | train 3.323681 | val 3.510341 | tok/s 2628106.13 | norm 0.5432 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3460 | train 3.123910 | val 3.510341 | tok/s 2626722.36 | norm 0.5176 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3461 | train 3.128166 | val 3.510341 | tok/s 2628165.90 | norm 0.4816 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3462 | train 3.110869 | val 3.510341 | tok/s 2623008.09 | norm 0.5170 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3463 | train 3.156838 | val 3.510341 | tok/s 2631865.36 | norm 0.4766 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3464 | train 3.216335 | val 3.510341 | tok/s 2628407.20 | norm 0.4392 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3465 | train 3.099941 | val 3.510341 | tok/s 2628200.90 | norm 0.4555 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3466 | train 3.089361 | val 3.510341 | tok/s 2623696.46 | norm 0.4620 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3467 | train 3.144050 | val 3.510341 | tok/s 2626289.89 | norm 0.5314 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3468 | train 3.046826 | val 3.510341 | tok/s 2630779.11 | norm 0.4870 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3469 | train 3.091506 | val 3.510341 | tok/s 2627377.04 | norm 0.4651 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3470 | train 3.040401 | val 3.510341 | tok/s 2630703.71 | norm 0.4279 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3471 | train 3.124704 | val 3.510341 | tok/s 2629361.40 | norm 0.4315 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3472 | train 3.226573 | val 3.510341 | tok/s 2628745.22 | norm 0.4494 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3473 | train 3.280488 | val 3.510341 | tok/s 2627184.56 | norm 0.5351 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3474 | train 3.161722 | val 3.510341 | tok/s 2627750.15 | norm 0.5537 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3475 | train 3.156831 | val 3.510341 | tok/s 2631347.80 | norm 0.4781 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3476 | train 3.153770 | val 3.510341 | tok/s 2627079.38 | norm 0.4887 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3477 | train 3.222252 | val 3.510341 | tok/s 2629592.06 | norm 0.4737 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3478 | train 3.113986 | val 3.510341 | tok/s 2626526.70 | norm 0.4276 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3479 | train 3.158445 | val 3.510341 | tok/s 2623485.58 | norm 0.4568 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3480 | train 3.170970 | val 3.510341 | tok/s 2628791.14 | norm 0.5086 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3481 | train 3.124952 | val 3.510341 | tok/s 2626895.01 | norm 0.5040 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3482 | train 3.155780 | val 3.510341 | tok/s 2631340.22 | norm 0.4912 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3483 | train 3.150840 | val 3.510341 | tok/s 2629464.98 | norm 0.4694 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3484 | train 3.087815 | val 3.510341 | tok/s 2625328.06 | norm 0.4763 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3485 | train 3.152033 | val 3.510341 | tok/s 2630965.37 | norm 0.4958 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3486 | train 3.156042 | val 3.510341 | tok/s 2630060.13 | norm 0.4712 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3487 | train 3.156995 | val 3.510341 | tok/s 2630830.08 | norm 0.4666 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3488 | train 3.099923 | val 3.510341 | tok/s 2630691.04 | norm 0.5263 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3489 | train 3.204309 | val 3.510341 | tok/s 2630067.01 | norm 0.4907 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3490 | train 3.155305 | val 3.510341 | tok/s 2632623.78 | norm 0.4969 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3491 | train 3.113156 | val 3.510341 | tok/s 2624249.10 | norm 0.5044 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3492 | train 3.055234 | val 3.510341 | tok/s 2630238.32 | norm 0.4554 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3493 | train 3.166965 | val 3.510341 | tok/s 2628747.54 | norm 0.4542 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3494 | train 3.139601 | val 3.510341 | tok/s 2628651.85 | norm 0.4992 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3495 | train 3.132060 | val 3.510341 | tok/s 2630588.46 | norm 0.4813 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3496 | train 3.085569 | val 3.510341 | tok/s 2627749.85 | norm 0.4841 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3497 | train 3.107747 | val 3.510341 | tok/s 2631575.34 | norm 0.4607 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3498 | train 3.092891 | val 3.510341 | tok/s 2631432.34 | norm 0.4493 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3499 | train 3.115816 | val 3.510341 | tok/s 2627709.12 | norm 0.5098 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 3500 | val_loss 3.499398 step 3500 | train 3.230617 | val 3.499398 | tok/s 2620711.48 | norm 0.5360 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3501 | train 3.198026 | val 3.499398 | tok/s 2629159.90 | norm 0.4979 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3502 | train 3.264713 | val 3.499398 | tok/s 2625278.50 | norm 0.5457 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3503 | train 3.075437 | val 3.499398 | tok/s 2629353.06 | norm 0.4870 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3504 | train 3.249568 | val 3.499398 | tok/s 2626003.90 | norm 0.6408 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3505 | train 3.181210 | val 3.499398 | tok/s 2627438.59 | norm 0.4480 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3506 | train 3.149470 | val 3.499398 | tok/s 2630424.27 | norm 0.3969 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3507 | train 3.262410 | val 3.499398 | tok/s 2627312.75 | norm 0.4762 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3508 | train 3.138424 | val 3.499398 | tok/s 2625042.91 | norm 0.5372 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3509 | train 3.043461 | val 3.499398 | tok/s 2629538.81 | norm 0.4673 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3510 | train 3.088236 | val 3.499398 | tok/s 2626579.64 | norm 0.4705 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3511 | train 3.034730 | val 3.499398 | tok/s 2629101.86 | norm 0.4952 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3512 | train 3.310190 | val 3.499398 | tok/s 2627151.91 | norm 0.4909 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3513 | train 3.149060 | val 3.499398 | tok/s 2625411.83 | norm 0.4776 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3514 | train 3.047380 | val 3.499398 | tok/s 2627460.13 | norm 0.4953 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3515 | train 3.201339 | val 3.499398 | tok/s 2577829.66 | norm 0.4968 | dt 203.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3516 | train 3.140475 | val 3.499398 | tok/s 2630299.49 | norm 0.4945 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3517 | train 3.173858 | val 3.499398 | tok/s 2624854.52 | norm 0.4387 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3518 | train 3.153930 | val 3.499398 | tok/s 2621502.69 | norm 0.4074 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3519 | train 3.073007 | val 3.499398 | tok/s 2612696.96 | norm 0.4639 | dt 200.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3520 | train 3.097046 | val 3.499398 | tok/s 2629285.18 | norm 0.5406 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3521 | train 3.059276 | val 3.499398 | tok/s 2629387.96 | norm 0.5511 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3522 | train 3.042542 | val 3.499398 | tok/s 2628791.73 | norm 0.5429 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3523 | train 3.010112 | val 3.499398 | tok/s 2627558.98 | norm 0.5579 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3524 | train 3.210107 | val 3.499398 | tok/s 2627869.92 | norm 0.5542 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3525 | train 3.192256 | val 3.499398 | tok/s 2630554.33 | norm 0.4978 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3526 | train 3.202682 | val 3.499398 | tok/s 2631803.93 | norm 0.4366 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3527 | train 3.219620 | val 3.499398 | tok/s 2625213.05 | norm 0.4688 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3528 | train 3.168470 | val 3.499398 | tok/s 2626836.52 | norm 0.5131 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3529 | train 3.181318 | val 3.499398 | tok/s 2628697.91 | norm 0.5027 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3530 | train 3.002873 | val 3.499398 | tok/s 2626991.90 | norm 0.4495 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3531 | train 3.164463 | val 3.499398 | tok/s 2629911.64 | norm 0.4314 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3532 | train 3.138631 | val 3.499398 | tok/s 2626121.13 | norm 0.4455 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3533 | train 3.139361 | val 3.499398 | tok/s 2630770.82 | norm 0.5034 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3534 | train 2.995231 | val 3.499398 | tok/s 2629166.59 | norm 0.4693 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3535 | train 3.066583 | val 3.499398 | tok/s 2629918.04 | norm 0.4985 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3536 | train 3.366369 | val 3.499398 | tok/s 2632304.03 | norm 0.5556 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3537 | train 3.202154 | val 3.499398 | tok/s 2625604.29 | norm 0.5403 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3538 | train 3.114791 | val 3.499398 | tok/s 2630668.95 | norm 0.4518 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3539 | train 2.985111 | val 3.499398 | tok/s 2628760.15 | norm 0.4628 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3540 | train 3.111894 | val 3.499398 | tok/s 2627551.22 | norm 0.7927 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3541 | train 3.225337 | val 3.499398 | tok/s 2632669.45 | norm 0.5087 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3542 | train 3.159721 | val 3.499398 | tok/s 2629661.84 | norm 0.4827 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3543 | train 3.076711 | val 3.499398 | tok/s 2625486.01 | norm 0.5669 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3544 | train 3.205663 | val 3.499398 | tok/s 2624413.34 | norm 0.5380 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3545 | train 3.148688 | val 3.499398 | tok/s 2627282.54 | norm 0.4679 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3546 | train 3.286141 | val 3.499398 | tok/s 2630018.36 | norm 0.5926 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3547 | train 3.304303 | val 3.499398 | tok/s 2626293.50 | norm 0.5381 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3548 | train 3.187400 | val 3.499398 | tok/s 2634232.36 | norm 0.5180 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3549 | train 3.178698 | val 3.499398 | tok/s 2627974.66 | norm 0.4630 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3550 | train 3.200153 | val 3.499398 | tok/s 2621891.03 | norm 0.4565 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3551 | train 3.194664 | val 3.499398 | tok/s 2630738.37 | norm 0.4582 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3552 | train 3.263705 | val 3.499398 | tok/s 2629809.75 | norm 0.4985 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3553 | train 3.171588 | val 3.499398 | tok/s 2627768.13 | norm 0.5060 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3554 | train 3.099561 | val 3.499398 | tok/s 2630799.23 | norm 0.4925 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3555 | train 3.181312 | val 3.499398 | tok/s 2629057.81 | norm 0.5416 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3556 | train 3.280299 | val 3.499398 | tok/s 2628944.59 | norm 0.5437 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3557 | train 3.209012 | val 3.499398 | tok/s 2623925.05 | norm 0.5299 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3558 | train 3.091752 | val 3.499398 | tok/s 2629929.78 | norm 0.4878 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3559 | train 3.136718 | val 3.499398 | tok/s 2630157.47 | norm 0.4644 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3560 | train 3.074707 | val 3.499398 | tok/s 2627734.49 | norm 0.4334 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3561 | train 3.082512 | val 3.499398 | tok/s 2630250.11 | norm 0.4557 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3562 | train 3.082770 | val 3.499398 | tok/s 2629726.89 | norm 0.4698 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3563 | train 3.124492 | val 3.499398 | tok/s 2632612.37 | norm 0.4685 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3564 | train 3.206693 | val 3.499398 | tok/s 2627324.97 | norm 0.4527 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3565 | train 3.271348 | val 3.499398 | tok/s 2623632.24 | norm 0.4612 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3566 | train 3.160492 | val 3.499398 | tok/s 2627784.52 | norm 0.4415 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3567 | train 3.082738 | val 3.499398 | tok/s 2624247.70 | norm 0.5183 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3568 | train 3.184155 | val 3.499398 | tok/s 2630600.21 | norm 0.4743 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3569 | train 3.134106 | val 3.499398 | tok/s 2623512.81 | norm 0.4567 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3570 | train 3.090858 | val 3.499398 | tok/s 2622992.42 | norm 0.4892 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3571 | train 3.230573 | val 3.499398 | tok/s 2628295.03 | norm 0.4383 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3572 | train 3.111128 | val 3.499398 | tok/s 2628231.14 | norm 0.4664 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3573 | train 3.120954 | val 3.499398 | tok/s 2629221.84 | norm 0.4515 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3574 | train 3.148468 | val 3.499398 | tok/s 2626928.09 | norm 0.4465 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3575 | train 3.095038 | val 3.499398 | tok/s 2625352.54 | norm 0.4595 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3576 | train 3.086558 | val 3.499398 | tok/s 2631228.62 | norm 0.4809 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3577 | train 3.166294 | val 3.499398 | tok/s 2627713.85 | norm 0.4588 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3578 | train 3.169530 | val 3.499398 | tok/s 2628673.19 | norm 0.4978 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3579 | train 3.192371 | val 3.499398 | tok/s 2625968.94 | norm 0.5282 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3580 | train 3.149574 | val 3.499398 | tok/s 2624813.55 | norm 0.4723 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3581 | train 3.215556 | val 3.499398 | tok/s 2627352.02 | norm 0.4705 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3582 | train 2.896691 | val 3.499398 | tok/s 2626215.29 | norm 0.5523 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3583 | train 3.141191 | val 3.499398 | tok/s 2627781.03 | norm 0.5208 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3584 | train 3.180530 | val 3.499398 | tok/s 2626389.46 | norm 0.4610 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3585 | train 3.225351 | val 3.499398 | tok/s 2625004.98 | norm 0.4944 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3586 | train 3.140038 | val 3.499398 | tok/s 2630264.76 | norm 0.4752 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3587 | train 3.131029 | val 3.499398 | tok/s 2627828.62 | norm 0.4531 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3588 | train 3.292055 | val 3.499398 | tok/s 2600339.58 | norm 0.4652 | dt 201.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3589 | train 3.155951 | val 3.499398 | tok/s 2628360.78 | norm 0.4501 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3590 | train 3.091260 | val 3.499398 | tok/s 2628858.22 | norm 0.4429 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3591 | train 3.189001 | val 3.499398 | tok/s 2632046.36 | norm 0.5405 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3592 | train 3.157118 | val 3.499398 | tok/s 2627795.70 | norm 0.5329 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3593 | train 3.246239 | val 3.499398 | tok/s 2627368.06 | norm 0.5018 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3594 | train 3.136374 | val 3.499398 | tok/s 2628417.00 | norm 0.5010 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3595 | train 3.084128 | val 3.499398 | tok/s 2626276.31 | norm 0.4741 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3596 | train 3.113291 | val 3.499398 | tok/s 2628055.67 | norm 0.4691 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3597 | train 3.127301 | val 3.499398 | tok/s 2625649.81 | norm 0.4970 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3598 | train 3.026047 | val 3.499398 | tok/s 2630980.83 | norm 0.4949 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3599 | train 3.087934 | val 3.499398 | tok/s 2623460.35 | norm 0.4715 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3600 | train 3.118431 | val 3.499398 | tok/s 2627282.95 | norm 0.4953 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3601 | train 3.138350 | val 3.499398 | tok/s 2629815.09 | norm 0.4926 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3602 | train 3.246905 | val 3.499398 | tok/s 2628057.24 | norm 0.4990 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3603 | train 3.111603 | val 3.499398 | tok/s 2630315.70 | norm 0.5248 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3604 | train 3.160648 | val 3.499398 | tok/s 2620119.08 | norm 0.5136 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3605 | train 3.235340 | val 3.499398 | tok/s 2621558.11 | norm 0.5337 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3606 | train 3.099880 | val 3.499398 | tok/s 2628469.57 | norm 0.5637 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3607 | train 3.148730 | val 3.499398 | tok/s 2627284.84 | norm 0.4955 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3608 | train 3.112285 | val 3.499398 | tok/s 2629475.70 | norm 0.4841 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3609 | train 2.971526 | val 3.499398 | tok/s 2623956.47 | norm 0.5154 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3610 | train 3.053002 | val 3.499398 | tok/s 2619972.83 | norm 0.5401 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3611 | train 3.145271 | val 3.499398 | tok/s 2629442.52 | norm 0.5211 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3612 | train 3.085325 | val 3.499398 | tok/s 2625823.94 | norm 0.4658 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3613 | train 3.124655 | val 3.499398 | tok/s 2631150.06 | norm 0.4750 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3614 | train 3.021451 | val 3.499398 | tok/s 2622943.75 | norm 0.4965 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3615 | train 3.181395 | val 3.499398 | tok/s 2623732.26 | norm 0.4958 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3616 | train 3.163846 | val 3.499398 | tok/s 2630948.93 | norm 0.5481 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3617 | train 3.125367 | val 3.499398 | tok/s 2627495.09 | norm 0.5248 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3618 | train 3.050476 | val 3.499398 | tok/s 2629627.83 | norm 0.4488 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3619 | train 3.202790 | val 3.499398 | tok/s 2624794.83 | norm 0.4853 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3620 | train 3.063576 | val 3.499398 | tok/s 2628311.48 | norm 0.4728 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3621 | train 3.128478 | val 3.499398 | tok/s 2521155.36 | norm 0.4569 | dt 207.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3622 | train 3.275436 | val 3.499398 | tok/s 2627834.95 | norm 0.5119 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3623 | train 3.099210 | val 3.499398 | tok/s 2626439.34 | norm 0.5383 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3624 | train 3.108481 | val 3.499398 | tok/s 2627610.94 | norm 0.4460 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3625 | train 3.288832 | val 3.499398 | tok/s 2627040.75 | norm 0.4529 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3626 | train 3.075095 | val 3.499398 | tok/s 2626690.49 | norm 0.4763 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3627 | train 3.098062 | val 3.499398 | tok/s 2631491.84 | norm 0.4448 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3628 | train 3.276809 | val 3.499398 | tok/s 2629796.74 | norm 0.5716 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3629 | train 3.237176 | val 3.499398 | tok/s 2624110.40 | norm 0.4639 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3630 | train 3.246844 | val 3.499398 | tok/s 2628421.93 | norm 0.4896 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3631 | train 3.247371 | val 3.499398 | tok/s 2630486.37 | norm 0.4910 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3632 | train 3.170574 | val 3.499398 | tok/s 2621331.69 | norm 0.5070 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3633 | train 3.089397 | val 3.499398 | tok/s 2626492.56 | norm 0.4505 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3634 | train 3.175278 | val 3.499398 | tok/s 2629241.89 | norm 0.4458 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3635 | train 3.070582 | val 3.499398 | tok/s 2629605.16 | norm 0.5594 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3636 | train 3.092415 | val 3.499398 | tok/s 2625917.69 | norm 0.4888 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3637 | train 3.050510 | val 3.499398 | tok/s 2624893.95 | norm 0.4726 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3638 | train 3.224980 | val 3.499398 | tok/s 2630679.43 | norm 0.4827 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3639 | train 3.048709 | val 3.499398 | tok/s 2626964.95 | norm 0.5217 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3640 | train 3.083003 | val 3.499398 | tok/s 2629214.68 | norm 0.4996 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3641 | train 3.231595 | val 3.499398 | tok/s 2629231.64 | norm 0.4819 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3642 | train 3.177654 | val 3.499398 | tok/s 2630921.91 | norm 0.4653 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3643 | train 3.198668 | val 3.499398 | tok/s 2628897.83 | norm 0.5265 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3644 | train 3.138923 | val 3.499398 | tok/s 2625931.47 | norm 0.5423 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3645 | train 3.127288 | val 3.499398 | tok/s 2628991.09 | norm 0.4591 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3646 | train 3.152704 | val 3.499398 | tok/s 2628637.58 | norm 0.5351 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3647 | train 3.118500 | val 3.499398 | tok/s 2627700.54 | norm 0.4725 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3648 | train 3.188725 | val 3.499398 | tok/s 2629521.10 | norm 0.4454 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3649 | train 3.012239 | val 3.499398 | tok/s 2628726.87 | norm 0.4538 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3650 | train 3.153296 | val 3.499398 | tok/s 2629150.85 | norm 0.4862 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3651 | train 3.241330 | val 3.499398 | tok/s 2619538.20 | norm 0.4718 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3652 | train 3.106631 | val 3.499398 | tok/s 2621767.10 | norm 0.4745 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3653 | train 3.156615 | val 3.499398 | tok/s 2612820.01 | norm 0.4827 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3654 | train 3.141227 | val 3.499398 | tok/s 2629522.81 | norm 0.5062 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3655 | train 3.110734 | val 3.499398 | tok/s 2628634.61 | norm 0.4945 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3656 | train 3.216856 | val 3.499398 | tok/s 2624263.77 | norm 0.5178 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3657 | train 3.177737 | val 3.499398 | tok/s 2623991.82 | norm 0.4781 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3658 | train 3.152924 | val 3.499398 | tok/s 2614268.47 | norm 0.4314 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3659 | train 3.102381 | val 3.499398 | tok/s 2620173.51 | norm 0.4549 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3660 | train 3.157098 | val 3.499398 | tok/s 2626104.89 | norm 0.4783 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3661 | train 3.193258 | val 3.499398 | tok/s 2625573.28 | norm 0.4780 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3662 | train 3.235873 | val 3.499398 | tok/s 2627113.50 | norm 0.4772 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3663 | train 3.097914 | val 3.499398 | tok/s 2630958.57 | norm 0.4845 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3664 | train 2.966327 | val 3.499398 | tok/s 2626517.99 | norm 0.5080 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3665 | train 3.116342 | val 3.499398 | tok/s 2628760.23 | norm 0.4776 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3666 | train 3.439403 | val 3.499398 | tok/s 2542911.31 | norm 0.5325 | dt 206.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3667 | train 3.141701 | val 3.499398 | tok/s 2619163.89 | norm 0.4957 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3668 | train 3.033847 | val 3.499398 | tok/s 2628257.04 | norm 0.4578 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3669 | train 3.112255 | val 3.499398 | tok/s 2621675.35 | norm 0.4978 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3670 | train 3.143500 | val 3.499398 | tok/s 2626624.26 | norm 0.4544 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3671 | train 3.149032 | val 3.499398 | tok/s 2620026.99 | norm 0.4138 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3672 | train 3.019098 | val 3.499398 | tok/s 2627353.69 | norm 0.4224 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3673 | train 3.074401 | val 3.499398 | tok/s 2623993.53 | norm 0.4426 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3674 | train 3.248822 | val 3.499398 | tok/s 2621242.88 | norm 0.4462 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3675 | train 3.086026 | val 3.499398 | tok/s 2626978.75 | norm 0.4497 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3676 | train 3.084715 | val 3.499398 | tok/s 2623682.16 | norm 0.4288 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3677 | train 3.144996 | val 3.499398 | tok/s 2628531.25 | norm 0.4432 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3678 | train 3.034038 | val 3.499398 | tok/s 2624891.94 | norm 0.4994 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3679 | train 3.134340 | val 3.499398 | tok/s 2624172.97 | norm 0.5277 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3680 | train 3.391574 | val 3.499398 | tok/s 2625672.02 | norm 0.5354 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3681 | train 3.037052 | val 3.499398 | tok/s 2625326.45 | norm 0.5308 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3682 | train 3.165580 | val 3.499398 | tok/s 2629461.73 | norm 0.4468 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3683 | train 3.238360 | val 3.499398 | tok/s 2621616.07 | norm 0.4690 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3684 | train 3.148874 | val 3.499398 | tok/s 2626239.44 | norm 0.5146 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3685 | train 3.171824 | val 3.499398 | tok/s 2625392.82 | norm 0.4705 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3686 | train 3.144576 | val 3.499398 | tok/s 2627801.76 | norm 0.4769 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3687 | train 3.036970 | val 3.499398 | tok/s 2629482.61 | norm 0.4936 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3688 | train 3.136846 | val 3.499398 | tok/s 2622019.76 | norm 0.4836 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3689 | train 3.147889 | val 3.499398 | tok/s 2624862.12 | norm 0.5209 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3690 | train 3.276219 | val 3.499398 | tok/s 2631340.74 | norm 0.5400 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3691 | train 3.292193 | val 3.499398 | tok/s 2628277.44 | norm 0.5381 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3692 | train 3.069651 | val 3.499398 | tok/s 2626853.60 | norm 0.5061 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3693 | train 3.102964 | val 3.499398 | tok/s 2623763.16 | norm 0.5346 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3694 | train 3.103970 | val 3.499398 | tok/s 2626306.16 | norm 0.4898 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3695 | train 3.056550 | val 3.499398 | tok/s 2629285.10 | norm 0.4956 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3696 | train 3.081966 | val 3.499398 | tok/s 2559662.78 | norm 0.5233 | dt 204.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3697 | train 3.024451 | val 3.499398 | tok/s 2601570.56 | norm 0.4833 | dt 201.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3698 | train 3.063905 | val 3.499398 | tok/s 2623027.63 | norm 0.4906 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3699 | train 3.370899 | val 3.499398 | tok/s 2598231.44 | norm 0.5831 | dt 201.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3700 | train 3.108500 | val 3.499398 | tok/s 2627765.46 | norm 0.4937 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3701 | train 3.252410 | val 3.499398 | tok/s 2626744.14 | norm 0.4505 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3702 | train 3.096889 | val 3.499398 | tok/s 2630354.68 | norm 0.4602 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3703 | train 3.066983 | val 3.499398 | tok/s 2627969.55 | norm 0.4472 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3704 | train 3.131383 | val 3.499398 | tok/s 2629357.16 | norm 0.4966 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3705 | train 3.060640 | val 3.499398 | tok/s 2629955.80 | norm 0.4452 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3706 | train 3.307468 | val 3.499398 | tok/s 2547597.97 | norm 0.4081 | dt 205.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3707 | train 3.074618 | val 3.499398 | tok/s 2630215.42 | norm 0.4075 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3708 | train 3.210339 | val 3.499398 | tok/s 2628126.61 | norm 0.4461 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3709 | train 3.339406 | val 3.499398 | tok/s 2629893.54 | norm 0.5663 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3710 | train 3.148607 | val 3.499398 | tok/s 2625538.27 | norm 0.4829 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3711 | train 3.037324 | val 3.499398 | tok/s 2618395.50 | norm 0.4748 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3712 | train 3.169602 | val 3.499398 | tok/s 2630007.34 | norm 0.5066 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3713 | train 3.089716 | val 3.499398 | tok/s 2624049.84 | norm 0.4586 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3714 | train 3.251073 | val 3.499398 | tok/s 2630465.40 | norm 0.4420 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3715 | train 3.208957 | val 3.499398 | tok/s 2632940.12 | norm 0.4919 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3716 | train 3.162270 | val 3.499398 | tok/s 2624255.47 | norm 0.4788 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3717 | train 3.163915 | val 3.499398 | tok/s 2630458.55 | norm 0.4497 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3718 | train 3.173114 | val 3.499398 | tok/s 2628123.83 | norm 0.4553 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3719 | train 3.186348 | val 3.499398 | tok/s 2629594.64 | norm 0.4964 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3720 | train 3.116907 | val 3.499398 | tok/s 2620483.55 | norm 0.4794 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3721 | train 3.145330 | val 3.499398 | tok/s 2621292.16 | norm 0.4805 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3722 | train 3.183339 | val 3.499398 | tok/s 2627476.60 | norm 0.4804 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3723 | train 3.041266 | val 3.499398 | tok/s 2624856.97 | norm 0.4861 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3724 | train 3.242662 | val 3.499398 | tok/s 2626225.27 | norm 0.4953 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3725 | train 3.074731 | val 3.499398 | tok/s 2624675.45 | norm 0.4651 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3726 | train 3.212017 | val 3.499398 | tok/s 2627448.54 | norm 0.5095 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3727 | train 3.057962 | val 3.499398 | tok/s 2632283.86 | norm 0.5047 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3728 | train 3.234426 | val 3.499398 | tok/s 2627641.54 | norm 0.4987 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3729 | train 3.297987 | val 3.499398 | tok/s 2630566.50 | norm 0.5270 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3730 | train 3.157434 | val 3.499398 | tok/s 2625808.93 | norm 0.4926 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3731 | train 3.196133 | val 3.499398 | tok/s 2626575.31 | norm 0.4780 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3732 | train 3.215369 | val 3.499398 | tok/s 2629124.04 | norm 0.4820 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3733 | train 3.042788 | val 3.499398 | tok/s 2625683.92 | norm 0.5328 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3734 | train 3.107685 | val 3.499398 | tok/s 2630087.61 | norm 0.5332 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3735 | train 3.222497 | val 3.499398 | tok/s 2625977.86 | norm 0.4986 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3736 | train 3.062411 | val 3.499398 | tok/s 2504858.46 | norm 0.4513 | dt 209.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3737 | train 3.165757 | val 3.499398 | tok/s 2624146.43 | norm 0.4268 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3738 | train 3.034589 | val 3.499398 | tok/s 2628579.79 | norm 0.4641 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3739 | train 3.007503 | val 3.499398 | tok/s 2630794.42 | norm 0.4608 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3740 | train 3.043046 | val 3.499398 | tok/s 2625029.16 | norm 0.4357 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3741 | train 3.181041 | val 3.499398 | tok/s 2628638.58 | norm 0.6334 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3742 | train 3.130058 | val 3.499398 | tok/s 2628334.40 | norm 0.4444 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3743 | train 3.082230 | val 3.499398 | tok/s 2536348.57 | norm 0.4712 | dt 206.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3744 | train 3.241940 | val 3.499398 | tok/s 2632357.60 | norm 0.4462 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3745 | train 3.424220 | val 3.499398 | tok/s 2627704.18 | norm 0.6256 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3746 | train 3.195757 | val 3.499398 | tok/s 2630169.76 | norm 0.4673 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3747 | train 3.105937 | val 3.499398 | tok/s 2625685.35 | norm 0.4294 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3748 | train 3.146508 | val 3.499398 | tok/s 2626292.34 | norm 0.5006 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3749 | train 3.119779 | val 3.499398 | tok/s 2628559.57 | norm 0.4380 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 3750 | val_loss 3.495013 step 3750 | train 3.293768 | val 3.495013 | tok/s 2614216.81 | norm 0.4454 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3751 | train 3.022685 | val 3.495013 | tok/s 2625943.12 | norm 0.4629 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3752 | train 3.160407 | val 3.495013 | tok/s 2628077.45 | norm 0.4214 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3753 | train 3.260841 | val 3.495013 | tok/s 2624460.43 | norm 0.4824 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3754 | train 3.083126 | val 3.495013 | tok/s 2624184.43 | norm 0.4945 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3755 | train 3.073256 | val 3.495013 | tok/s 2630183.37 | norm 0.4793 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3756 | train 3.135721 | val 3.495013 | tok/s 2627899.63 | norm 0.4460 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3757 | train 3.228307 | val 3.495013 | tok/s 2630456.63 | norm 0.4547 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3758 | train 3.110988 | val 3.495013 | tok/s 2628560.39 | norm 0.4411 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3759 | train 3.069471 | val 3.495013 | tok/s 2625004.78 | norm 0.4359 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3760 | train 3.142053 | val 3.495013 | tok/s 2627744.59 | norm 0.4281 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3761 | train 3.164568 | val 3.495013 | tok/s 2624976.66 | norm 0.4162 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3762 | train 3.056435 | val 3.495013 | tok/s 2630294.52 | norm 0.4629 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3763 | train 3.107870 | val 3.495013 | tok/s 2629000.84 | norm 0.5099 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3764 | train 3.137971 | val 3.495013 | tok/s 2621196.74 | norm 0.5211 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3765 | train 3.216102 | val 3.495013 | tok/s 2632734.50 | norm 0.4730 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3766 | train 3.138438 | val 3.495013 | tok/s 2626841.81 | norm 0.4453 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3767 | train 3.236375 | val 3.495013 | tok/s 2627876.30 | norm 0.4699 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3768 | train 3.143879 | val 3.495013 | tok/s 2612028.18 | norm 0.5058 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3769 | train 3.151213 | val 3.495013 | tok/s 2624161.93 | norm 0.4810 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3770 | train 3.088487 | val 3.495013 | tok/s 2629328.91 | norm 0.4504 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3771 | train 3.245146 | val 3.495013 | tok/s 2628280.99 | norm 0.4437 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3772 | train 3.096675 | val 3.495013 | tok/s 2628154.62 | norm 0.4505 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3773 | train 3.084552 | val 3.495013 | tok/s 2623087.01 | norm 0.4666 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3774 | train 3.048705 | val 3.495013 | tok/s 2628676.71 | norm 0.4615 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3775 | train 3.140412 | val 3.495013 | tok/s 2630345.94 | norm 0.4368 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3776 | train 3.278225 | val 3.495013 | tok/s 2628684.31 | norm 0.4933 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3777 | train 3.219394 | val 3.495013 | tok/s 2626346.04 | norm 0.4585 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3778 | train 3.065764 | val 3.495013 | tok/s 2626905.16 | norm 0.4621 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3779 | train 3.115261 | val 3.495013 | tok/s 2627334.79 | norm 0.4721 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3780 | train 3.061233 | val 3.495013 | tok/s 2633007.57 | norm 0.4366 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3781 | train 3.085722 | val 3.495013 | tok/s 2629234.46 | norm 0.4905 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3782 | train 3.079294 | val 3.495013 | tok/s 2628974.09 | norm 0.4855 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3783 | train 3.173177 | val 3.495013 | tok/s 2621938.05 | norm 0.4934 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3784 | train 3.083760 | val 3.495013 | tok/s 2528623.89 | norm 0.4672 | dt 207.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3785 | train 3.107449 | val 3.495013 | tok/s 2628554.99 | norm 0.4693 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3786 | train 3.093703 | val 3.495013 | tok/s 2629652.99 | norm 0.4556 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3787 | train 3.117892 | val 3.495013 | tok/s 2628991.19 | norm 0.4508 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3788 | train 3.118124 | val 3.495013 | tok/s 2629498.94 | norm 0.4708 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3789 | train 3.179876 | val 3.495013 | tok/s 2623982.10 | norm 0.4371 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3790 | train 3.137104 | val 3.495013 | tok/s 2627405.39 | norm 0.4492 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3791 | train 3.152492 | val 3.495013 | tok/s 2474157.85 | norm 0.5075 | dt 211.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3792 | train 2.931802 | val 3.495013 | tok/s 2630568.51 | norm 0.5190 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3793 | train 2.987676 | val 3.495013 | tok/s 2581601.27 | norm 0.4877 | dt 203.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3794 | train 3.178795 | val 3.495013 | tok/s 2629021.43 | norm 0.4953 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3795 | train 3.079748 | val 3.495013 | tok/s 2622725.63 | norm 0.4664 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3796 | train 3.097751 | val 3.495013 | tok/s 2621493.24 | norm 0.4549 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3797 | train 3.182567 | val 3.495013 | tok/s 2628344.91 | norm 0.4572 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3798 | train 3.111662 | val 3.495013 | tok/s 2627267.62 | norm 0.4534 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3799 | train 3.131558 | val 3.495013 | tok/s 2624614.41 | norm 0.4707 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3800 | train 3.119357 | val 3.495013 | tok/s 2624683.12 | norm 0.5411 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3801 | train 3.092597 | val 3.495013 | tok/s 2626303.03 | norm 0.5106 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3802 | train 3.060722 | val 3.495013 | tok/s 2632052.81 | norm 0.4569 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3803 | train 3.142252 | val 3.495013 | tok/s 2627334.91 | norm 0.5024 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3804 | train 3.167606 | val 3.495013 | tok/s 2630000.84 | norm 0.4795 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3805 | train 3.123017 | val 3.495013 | tok/s 2627102.42 | norm 0.4155 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3806 | train 3.149245 | val 3.495013 | tok/s 2628794.55 | norm 0.4453 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3807 | train 3.079911 | val 3.495013 | tok/s 2628271.27 | norm 0.4592 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3808 | train 3.100245 | val 3.495013 | tok/s 2629504.52 | norm 0.4746 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3809 | train 3.118805 | val 3.495013 | tok/s 2626212.87 | norm 0.4948 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3810 | train 2.979586 | val 3.495013 | tok/s 2626261.26 | norm 0.4858 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3811 | train 3.173518 | val 3.495013 | tok/s 2628555.79 | norm 0.5030 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3812 | train 3.174329 | val 3.495013 | tok/s 2633381.19 | norm 0.4516 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3813 | train 3.147338 | val 3.495013 | tok/s 2623306.65 | norm 0.4891 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3814 | train 3.129754 | val 3.495013 | tok/s 2629795.14 | norm 0.5243 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3815 | train 3.129892 | val 3.495013 | tok/s 2620012.54 | norm 0.5034 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3816 | train 3.030054 | val 3.495013 | tok/s 2626292.22 | norm 0.4785 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3817 | train 3.226426 | val 3.495013 | tok/s 2629954.33 | norm 0.4901 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3818 | train 3.159469 | val 3.495013 | tok/s 2626574.06 | norm 0.4707 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3819 | train 3.165417 | val 3.495013 | tok/s 2627227.95 | norm 0.4669 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3820 | train 3.160202 | val 3.495013 | tok/s 2626862.17 | norm 0.4864 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3821 | train 3.206291 | val 3.495013 | tok/s 2630427.32 | norm 0.4783 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3822 | train 3.131208 | val 3.495013 | tok/s 2630291.14 | norm 0.4739 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3823 | train 3.138905 | val 3.495013 | tok/s 2628922.32 | norm 0.5073 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3824 | train 3.104775 | val 3.495013 | tok/s 2630778.28 | norm 0.4627 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3825 | train 3.163560 | val 3.495013 | tok/s 2627809.89 | norm 0.4877 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3826 | train 3.217960 | val 3.495013 | tok/s 2622370.61 | norm 0.5099 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3827 | train 3.112050 | val 3.495013 | tok/s 2627822.28 | norm 0.5260 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3828 | train 3.202720 | val 3.495013 | tok/s 2627000.78 | norm 0.5165 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3829 | train 3.055805 | val 3.495013 | tok/s 2630289.99 | norm 0.4997 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3830 | train 3.287440 | val 3.495013 | tok/s 2625231.54 | norm 0.5599 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3831 | train 3.089117 | val 3.495013 | tok/s 2628096.75 | norm 0.5518 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3832 | train 3.120270 | val 3.495013 | tok/s 2629238.15 | norm 0.4816 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3833 | train 3.059277 | val 3.495013 | tok/s 2629493.77 | norm 0.5201 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3834 | train 3.252521 | val 3.495013 | tok/s 2629725.23 | norm 0.5244 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3835 | train 3.075839 | val 3.495013 | tok/s 2473569.64 | norm 0.4363 | dt 211.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3836 | train 3.127014 | val 3.495013 | tok/s 2626789.60 | norm 0.4295 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3837 | train 3.247891 | val 3.495013 | tok/s 2559734.10 | norm 0.4873 | dt 204.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3838 | train 3.070788 | val 3.495013 | tok/s 2623873.71 | norm 0.4361 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3839 | train 3.029281 | val 3.495013 | tok/s 2623279.75 | norm 0.5051 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3840 | train 3.301319 | val 3.495013 | tok/s 2626175.45 | norm 0.6744 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3841 | train 3.055133 | val 3.495013 | tok/s 2628047.36 | norm 0.4303 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3842 | train 3.044946 | val 3.495013 | tok/s 2624117.33 | norm 0.3939 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3843 | train 3.065089 | val 3.495013 | tok/s 2624694.95 | norm 0.4991 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3844 | train 3.009058 | val 3.495013 | tok/s 2626380.00 | norm 0.4316 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3845 | train 3.118265 | val 3.495013 | tok/s 2627217.54 | norm 0.4079 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3846 | train 3.253740 | val 3.495013 | tok/s 2627191.97 | norm 0.4721 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3847 | train 3.199981 | val 3.495013 | tok/s 2594235.06 | norm 0.5253 | dt 202.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3848 | train 3.194291 | val 3.495013 | tok/s 2623523.90 | norm 0.4980 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3849 | train 3.050829 | val 3.495013 | tok/s 2631782.51 | norm 0.4710 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3850 | train 3.109838 | val 3.495013 | tok/s 2613002.13 | norm 0.4986 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3851 | train 3.114935 | val 3.495013 | tok/s 2627147.28 | norm 0.5186 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3852 | train 3.029075 | val 3.495013 | tok/s 2622817.66 | norm 0.5209 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3853 | train 2.999780 | val 3.495013 | tok/s 2625409.71 | norm 0.4677 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3854 | train 3.198109 | val 3.495013 | tok/s 2630561.81 | norm 0.4418 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3855 | train 3.093903 | val 3.495013 | tok/s 2602555.26 | norm 0.4551 | dt 201.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3856 | train 3.043894 | val 3.495013 | tok/s 2629273.16 | norm 0.4712 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3857 | train 3.070652 | val 3.495013 | tok/s 2622736.08 | norm 0.4622 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3858 | train 3.193814 | val 3.495013 | tok/s 2625911.05 | norm 0.4195 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3859 | train 3.062934 | val 3.495013 | tok/s 2629520.94 | norm 0.4390 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3860 | train 3.153069 | val 3.495013 | tok/s 2625273.50 | norm 0.4970 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3861 | train 3.141464 | val 3.495013 | tok/s 2626941.16 | norm 0.5301 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3862 | train 3.115268 | val 3.495013 | tok/s 2621488.42 | norm 0.5031 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3863 | train 3.101629 | val 3.495013 | tok/s 2624771.66 | norm 0.4932 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3864 | train 3.059879 | val 3.495013 | tok/s 2629603.17 | norm 0.4753 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3865 | train 3.021585 | val 3.495013 | tok/s 2624796.47 | norm 0.4801 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3866 | train 3.110434 | val 3.495013 | tok/s 2548816.11 | norm 0.4789 | dt 205.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3867 | train 3.198466 | val 3.495013 | tok/s 2602785.81 | norm 0.4873 | dt 201.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3868 | train 3.016385 | val 3.495013 | tok/s 2488389.22 | norm 0.4688 | dt 210.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3869 | train 3.142670 | val 3.495013 | tok/s 2628659.27 | norm 0.4826 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3870 | train 3.146613 | val 3.495013 | tok/s 2625663.02 | norm 0.5184 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3871 | train 3.180070 | val 3.495013 | tok/s 2628899.90 | norm 0.5152 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3872 | train 3.112808 | val 3.495013 | tok/s 2622730.55 | norm 0.4638 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3873 | train 3.088408 | val 3.495013 | tok/s 2629653.00 | norm 0.4691 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3874 | train 3.010489 | val 3.495013 | tok/s 2622727.48 | norm 0.4971 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3875 | train 3.192085 | val 3.495013 | tok/s 2624661.22 | norm 0.4943 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3876 | train 3.022084 | val 3.495013 | tok/s 2626924.60 | norm 0.4731 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3877 | train 3.189712 | val 3.495013 | tok/s 2605329.33 | norm 0.5100 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3878 | train 3.120393 | val 3.495013 | tok/s 2627738.01 | norm 0.4707 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3879 | train 3.238127 | val 3.495013 | tok/s 2627031.37 | norm 0.4647 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3880 | train 3.166805 | val 3.495013 | tok/s 2628880.74 | norm 0.5218 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3881 | train 3.126542 | val 3.495013 | tok/s 2631315.05 | norm 0.4565 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3882 | train 3.154274 | val 3.495013 | tok/s 2626160.96 | norm 0.4621 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3883 | train 3.198705 | val 3.495013 | tok/s 2625079.92 | norm 0.5379 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3884 | train 3.198439 | val 3.495013 | tok/s 2618644.32 | norm 0.4930 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3885 | train 3.115197 | val 3.495013 | tok/s 2623356.93 | norm 0.4667 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3886 | train 3.098222 | val 3.495013 | tok/s 2626418.10 | norm 0.4992 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3887 | train 3.144811 | val 3.495013 | tok/s 2629291.74 | norm 0.5184 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3888 | train 3.140850 | val 3.495013 | tok/s 2628966.88 | norm 0.5342 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3889 | train 3.060933 | val 3.495013 | tok/s 2632203.73 | norm 0.4567 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3890 | train 3.095612 | val 3.495013 | tok/s 2623288.05 | norm 0.4608 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3891 | train 3.044043 | val 3.495013 | tok/s 2628823.00 | norm 0.4888 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3892 | train 3.027913 | val 3.495013 | tok/s 2627826.80 | norm 0.4835 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3893 | train 3.033211 | val 3.495013 | tok/s 2630405.30 | norm 0.5297 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3894 | train 3.010929 | val 3.495013 | tok/s 2624339.70 | norm 0.4759 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3895 | train 3.130740 | val 3.495013 | tok/s 2621747.54 | norm 0.4962 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3896 | train 3.205523 | val 3.495013 | tok/s 2629373.86 | norm 0.4845 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3897 | train 3.097927 | val 3.495013 | tok/s 2628391.58 | norm 0.4722 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3898 | train 3.052631 | val 3.495013 | tok/s 2629025.58 | norm 0.4392 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3899 | train 3.065960 | val 3.495013 | tok/s 2625887.52 | norm 0.4625 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3900 | train 3.080535 | val 3.495013 | tok/s 2626780.24 | norm 0.4753 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3901 | train 3.043843 | val 3.495013 | tok/s 2631216.88 | norm 0.4632 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3902 | train 2.984773 | val 3.495013 | tok/s 2627423.09 | norm 0.4711 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3903 | train 3.084219 | val 3.495013 | tok/s 2629294.12 | norm 0.4504 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3904 | train 3.209929 | val 3.495013 | tok/s 2628823.38 | norm 0.6678 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3905 | train 3.079710 | val 3.495013 | tok/s 2614813.59 | norm 0.5227 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3906 | train 3.028145 | val 3.495013 | tok/s 2631128.82 | norm 0.4840 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3907 | train 3.137976 | val 3.495013 | tok/s 2627019.90 | norm 0.5286 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3908 | train 3.239835 | val 3.495013 | tok/s 2631118.44 | norm 0.4992 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3909 | train 3.046484 | val 3.495013 | tok/s 2628716.43 | norm 0.4577 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3910 | train 3.117412 | val 3.495013 | tok/s 2629170.84 | norm 0.5028 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3911 | train 3.104780 | val 3.495013 | tok/s 2629416.20 | norm 0.4906 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3912 | train 3.168542 | val 3.495013 | tok/s 2486621.82 | norm 0.4486 | dt 210.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3913 | train 3.068886 | val 3.495013 | tok/s 2624468.26 | norm 0.4541 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3914 | train 3.066174 | val 3.495013 | tok/s 2625002.60 | norm 0.4502 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3915 | train 3.221430 | val 3.495013 | tok/s 2629996.88 | norm 0.4583 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3916 | train 3.165727 | val 3.495013 | tok/s 2625910.73 | norm 0.4961 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3917 | train 3.078155 | val 3.495013 | tok/s 2621499.52 | norm 0.4621 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3918 | train 2.978281 | val 3.495013 | tok/s 2631294.62 | norm 0.5109 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3919 | train 3.058165 | val 3.495013 | tok/s 2627243.61 | norm 0.5918 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3920 | train 2.969232 | val 3.495013 | tok/s 2629965.18 | norm 0.5011 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3921 | train 3.190274 | val 3.495013 | tok/s 2625497.96 | norm 0.4027 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3922 | train 3.169382 | val 3.495013 | tok/s 2624018.22 | norm 0.4543 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3923 | train 3.033155 | val 3.495013 | tok/s 2625175.45 | norm 0.4787 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3924 | train 3.181031 | val 3.495013 | tok/s 2628915.89 | norm 0.5073 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3925 | train 3.236774 | val 3.495013 | tok/s 2630731.89 | norm 0.5379 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3926 | train 3.135959 | val 3.495013 | tok/s 2625903.62 | norm 0.4859 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3927 | train 3.159496 | val 3.495013 | tok/s 2626328.36 | norm 0.4464 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3928 | train 3.175053 | val 3.495013 | tok/s 2630886.38 | norm 0.4884 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3929 | train 3.187957 | val 3.495013 | tok/s 2628568.69 | norm 0.5076 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3930 | train 3.020817 | val 3.495013 | tok/s 2628937.98 | norm 0.5163 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3931 | train 3.132620 | val 3.495013 | tok/s 2629812.60 | norm 0.5315 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3932 | train 3.063079 | val 3.495013 | tok/s 2630119.02 | norm 0.4977 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3933 | train 3.068851 | val 3.495013 | tok/s 2629805.40 | norm 0.4735 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3934 | train 3.164601 | val 3.495013 | tok/s 2630261.32 | norm 0.4926 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3935 | train 3.193616 | val 3.495013 | tok/s 2626189.19 | norm 0.4824 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3936 | train 3.172750 | val 3.495013 | tok/s 2627956.69 | norm 0.4734 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3937 | train 3.107313 | val 3.495013 | tok/s 2626343.13 | norm 0.4463 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3938 | train 3.124878 | val 3.495013 | tok/s 2631857.77 | norm 0.4376 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3939 | train 3.081826 | val 3.495013 | tok/s 2626397.11 | norm 0.4698 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3940 | train 3.178990 | val 3.495013 | tok/s 2629265.46 | norm 0.4860 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3941 | train 3.199011 | val 3.495013 | tok/s 2623033.07 | norm 0.4894 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3942 | train 3.187178 | val 3.495013 | tok/s 2622758.50 | norm 0.4549 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3943 | train 3.224597 | val 3.495013 | tok/s 2627409.05 | norm 0.4710 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3944 | train 3.085459 | val 3.495013 | tok/s 2624542.91 | norm 0.5290 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3945 | train 2.975846 | val 3.495013 | tok/s 2627725.66 | norm 0.4953 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3946 | train 3.060129 | val 3.495013 | tok/s 2625003.13 | norm 0.5011 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3947 | train 2.985421 | val 3.495013 | tok/s 2625150.79 | norm 0.4722 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3948 | train 3.014306 | val 3.495013 | tok/s 2629143.67 | norm 0.4992 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3949 | train 2.971958 | val 3.495013 | tok/s 2629502.57 | norm 0.5051 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3950 | train 3.081296 | val 3.495013 | tok/s 2628515.29 | norm 0.4871 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3951 | train 3.026325 | val 3.495013 | tok/s 2622797.69 | norm 0.5188 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3952 | train 3.115524 | val 3.495013 | tok/s 2529756.05 | norm 0.4422 | dt 207.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3953 | train 2.940466 | val 3.495013 | tok/s 2616817.55 | norm 0.4639 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3954 | train 3.027352 | val 3.495013 | tok/s 2627678.86 | norm 0.4757 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3955 | train 3.162082 | val 3.495013 | tok/s 2630542.27 | norm 0.4679 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3956 | train 3.151576 | val 3.495013 | tok/s 2629613.38 | norm 0.4326 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3957 | train 3.112439 | val 3.495013 | tok/s 2631688.64 | norm 0.4776 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3958 | train 3.213958 | val 3.495013 | tok/s 2475083.88 | norm 0.4818 | dt 211.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3959 | train 3.263452 | val 3.495013 | tok/s 2625920.40 | norm 0.4696 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3960 | train 3.087703 | val 3.495013 | tok/s 2629051.39 | norm 0.4655 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3961 | train 3.108911 | val 3.495013 | tok/s 2625995.70 | norm 0.5392 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3962 | train 2.972773 | val 3.495013 | tok/s 2627010.27 | norm 0.5407 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3963 | train 3.055352 | val 3.495013 | tok/s 2619520.86 | norm 0.4787 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3964 | train 3.056017 | val 3.495013 | tok/s 2621203.10 | norm 0.4298 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3965 | train 3.156047 | val 3.495013 | tok/s 2627789.28 | norm 0.4871 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3966 | train 3.079171 | val 3.495013 | tok/s 2625513.77 | norm 0.5282 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3967 | train 3.110776 | val 3.495013 | tok/s 2628088.20 | norm 0.4821 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3968 | train 3.267594 | val 3.495013 | tok/s 2625511.71 | norm 0.5062 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3969 | train 3.094534 | val 3.495013 | tok/s 2629523.52 | norm 0.4561 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3970 | train 3.180128 | val 3.495013 | tok/s 2629761.06 | norm 0.4866 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3971 | train 3.036080 | val 3.495013 | tok/s 2628279.19 | norm 0.5021 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3972 | train 3.057778 | val 3.495013 | tok/s 2626110.37 | norm 0.4345 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3973 | train 3.091052 | val 3.495013 | tok/s 2622301.48 | norm 0.4541 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3974 | train 3.113261 | val 3.495013 | tok/s 2626932.35 | norm 0.5062 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3975 | train 3.027717 | val 3.495013 | tok/s 2630172.13 | norm 0.4991 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3976 | train 2.986689 | val 3.495013 | tok/s 2625857.11 | norm 0.4746 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3977 | train 3.116606 | val 3.495013 | tok/s 2630672.37 | norm 0.5064 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3978 | train 3.147597 | val 3.495013 | tok/s 2624351.28 | norm 0.4835 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3979 | train 2.985532 | val 3.495013 | tok/s 2625191.06 | norm 0.4869 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3980 | train 3.067735 | val 3.495013 | tok/s 2630832.62 | norm 0.4748 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3981 | train 3.153355 | val 3.495013 | tok/s 2625557.44 | norm 0.5005 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3982 | train 3.108271 | val 3.495013 | tok/s 2629284.15 | norm 0.4741 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3983 | train 3.042675 | val 3.495013 | tok/s 2622072.68 | norm 0.4288 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3984 | train 3.171158 | val 3.495013 | tok/s 2624278.70 | norm 0.4497 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3985 | train 3.071547 | val 3.495013 | tok/s 2632091.11 | norm 0.4663 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3986 | train 3.126962 | val 3.495013 | tok/s 2629451.00 | norm 0.4801 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3987 | train 3.090093 | val 3.495013 | tok/s 2628867.51 | norm 0.4988 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3988 | train 3.081774 | val 3.495013 | tok/s 2625692.26 | norm 0.4631 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3989 | train 3.165141 | val 3.495013 | tok/s 2629270.10 | norm 0.4406 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3990 | train 3.083296 | val 3.495013 | tok/s 2629610.52 | norm 0.4750 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3991 | train 3.273327 | val 3.495013 | tok/s 2630136.31 | norm 0.4936 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3992 | train 3.126399 | val 3.495013 | tok/s 2547889.05 | norm 0.4684 | dt 205.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3993 | train 3.215081 | val 3.495013 | tok/s 2628806.34 | norm 0.4532 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3994 | train 3.151516 | val 3.495013 | tok/s 2628897.05 | norm 0.4497 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3995 | train 3.071728 | val 3.495013 | tok/s 2630188.31 | norm 0.4774 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3996 | train 3.219661 | val 3.495013 | tok/s 2624234.55 | norm 0.4427 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3997 | train 3.119633 | val 3.495013 | tok/s 2628681.16 | norm 0.4499 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3998 | train 3.122873 | val 3.495013 | tok/s 2629720.28 | norm 0.5338 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 3999 | train 3.108670 | val 3.495013 | tok/s 2613671.97 | norm 0.4683 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 4000 | val_loss 3.474666 step 4000 | train 3.156219 | val 3.474666 | tok/s 2167278.82 | norm 0.4233 | dt 241.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4001 | train 3.100334 | val 3.474666 | tok/s 2621632.68 | norm 0.4754 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4002 | train 3.099276 | val 3.474666 | tok/s 2626112.94 | norm 0.4784 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4003 | train 3.167835 | val 3.474666 | tok/s 2629049.11 | norm 0.5161 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4004 | train 3.161267 | val 3.474666 | tok/s 2627590.90 | norm 0.5116 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4005 | train 3.116271 | val 3.474666 | tok/s 2630586.11 | norm 0.5120 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4006 | train 3.248004 | val 3.474666 | tok/s 2621634.47 | norm 0.5442 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4007 | train 3.040016 | val 3.474666 | tok/s 2625987.54 | norm 0.5276 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4008 | train 3.122813 | val 3.474666 | tok/s 2609712.66 | norm 0.4515 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4009 | train 3.158482 | val 3.474666 | tok/s 2624697.92 | norm 0.4564 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4010 | train 3.128280 | val 3.474666 | tok/s 2626099.91 | norm 0.4710 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4011 | train 3.051763 | val 3.474666 | tok/s 2622246.64 | norm 0.4934 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4012 | train 3.167202 | val 3.474666 | tok/s 2628571.89 | norm 0.4665 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4013 | train 3.110041 | val 3.474666 | tok/s 2623090.40 | norm 0.4524 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4014 | train 3.048321 | val 3.474666 | tok/s 2629744.23 | norm 0.4378 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4015 | train 3.099379 | val 3.474666 | tok/s 2619174.36 | norm 0.4528 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4016 | train 3.179255 | val 3.474666 | tok/s 2623330.49 | norm 0.4567 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4017 | train 3.175374 | val 3.474666 | tok/s 2628553.29 | norm 0.4921 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4018 | train 3.148067 | val 3.474666 | tok/s 2626010.47 | norm 0.4862 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4019 | train 3.246611 | val 3.474666 | tok/s 2626570.44 | norm 0.5040 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4020 | train 3.091185 | val 3.474666 | tok/s 2620339.17 | norm 0.4891 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4021 | train 3.123441 | val 3.474666 | tok/s 2626649.10 | norm 0.4972 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4022 | train 3.215900 | val 3.474666 | tok/s 2626110.17 | norm 0.4665 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4023 | train 3.219159 | val 3.474666 | tok/s 2628904.33 | norm 0.4495 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4024 | train 3.155821 | val 3.474666 | tok/s 2626414.11 | norm 0.4199 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4025 | train 3.003608 | val 3.474666 | tok/s 2616697.50 | norm 0.4113 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4026 | train 3.062281 | val 3.474666 | tok/s 2623463.29 | norm 0.4353 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4027 | train 3.217359 | val 3.474666 | tok/s 2629084.26 | norm 0.4605 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4028 | train 3.166310 | val 3.474666 | tok/s 2623155.81 | norm 0.5252 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4029 | train 3.128034 | val 3.474666 | tok/s 2627983.86 | norm 0.4855 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4030 | train 3.199305 | val 3.474666 | tok/s 2625968.19 | norm 0.4543 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4031 | train 3.138627 | val 3.474666 | tok/s 2628082.62 | norm 0.7009 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4032 | train 3.161410 | val 3.474666 | tok/s 2628647.04 | norm 0.5068 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4033 | train 3.010659 | val 3.474666 | tok/s 2625189.32 | norm 0.4031 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4034 | train 3.102567 | val 3.474666 | tok/s 2628351.08 | norm 0.4826 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4035 | train 3.107549 | val 3.474666 | tok/s 2622247.59 | norm 0.4288 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4036 | train 3.005423 | val 3.474666 | tok/s 2626769.16 | norm 0.4666 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4037 | train 3.099553 | val 3.474666 | tok/s 2631017.47 | norm 0.4780 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4038 | train 3.102465 | val 3.474666 | tok/s 2626621.48 | norm 0.4329 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4039 | train 3.038206 | val 3.474666 | tok/s 2626321.33 | norm 0.4716 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4040 | train 3.181546 | val 3.474666 | tok/s 2624153.01 | norm 0.5283 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4041 | train 3.088952 | val 3.474666 | tok/s 2627879.37 | norm 0.5019 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4042 | train 3.136996 | val 3.474666 | tok/s 2630000.94 | norm 0.4938 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4043 | train 3.111704 | val 3.474666 | tok/s 2626536.47 | norm 0.4763 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4044 | train 3.045518 | val 3.474666 | tok/s 2629043.07 | norm 0.4832 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4045 | train 3.077571 | val 3.474666 | tok/s 2624366.89 | norm 0.4353 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4046 | train 3.154108 | val 3.474666 | tok/s 2627386.31 | norm 0.4597 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4047 | train 3.360513 | val 3.474666 | tok/s 2625605.17 | norm 0.4477 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4048 | train 3.221881 | val 3.474666 | tok/s 2618068.86 | norm 0.4469 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4049 | train 3.157773 | val 3.474666 | tok/s 2629522.88 | norm 0.4711 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4050 | train 3.028217 | val 3.474666 | tok/s 2625781.61 | norm 0.4612 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4051 | train 3.091490 | val 3.474666 | tok/s 2627986.67 | norm 0.4564 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4052 | train 3.194945 | val 3.474666 | tok/s 2627288.58 | norm 0.4747 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4053 | train 3.099801 | val 3.474666 | tok/s 2622225.91 | norm 0.4317 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4054 | train 3.060208 | val 3.474666 | tok/s 2624402.20 | norm 0.4991 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4055 | train 2.990692 | val 3.474666 | tok/s 2620959.32 | norm 0.5778 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4056 | train 3.014529 | val 3.474666 | tok/s 2626136.91 | norm 0.5081 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4057 | train 3.103683 | val 3.474666 | tok/s 2538065.12 | norm 0.4032 | dt 206.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4058 | train 3.143078 | val 3.474666 | tok/s 2625243.03 | norm 0.4344 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4059 | train 3.234305 | val 3.474666 | tok/s 2632891.11 | norm 0.4700 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4060 | train 3.159192 | val 3.474666 | tok/s 2623110.86 | norm 0.4605 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4061 | train 3.045525 | val 3.474666 | tok/s 2630057.70 | norm 0.4585 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4062 | train 3.129164 | val 3.474666 | tok/s 2626374.66 | norm 1.0114 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4063 | train 3.133121 | val 3.474666 | tok/s 2626064.37 | norm 0.3946 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4064 | train 3.212732 | val 3.474666 | tok/s 2626129.80 | norm 0.3813 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4065 | train 3.104899 | val 3.474666 | tok/s 2626559.60 | norm 0.4589 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4066 | train 3.205451 | val 3.474666 | tok/s 2622570.59 | norm 0.5061 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4067 | train 3.016461 | val 3.474666 | tok/s 2624774.82 | norm 0.4502 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4068 | train 3.049974 | val 3.474666 | tok/s 2626795.52 | norm 0.5041 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4069 | train 3.151670 | val 3.474666 | tok/s 2628311.52 | norm 0.4398 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4070 | train 3.081545 | val 3.474666 | tok/s 2626002.59 | norm 0.4361 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4071 | train 3.122487 | val 3.474666 | tok/s 2628888.28 | norm 0.4651 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4072 | train 3.104989 | val 3.474666 | tok/s 2624009.90 | norm 0.4796 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4073 | train 3.049070 | val 3.474666 | tok/s 2626505.62 | norm 0.4436 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4074 | train 3.137104 | val 3.474666 | tok/s 2629903.81 | norm 0.4426 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4075 | train 3.169090 | val 3.474666 | tok/s 2629664.18 | norm 0.4601 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4076 | train 3.087447 | val 3.474666 | tok/s 2628685.88 | norm 0.4982 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4077 | train 3.092102 | val 3.474666 | tok/s 2622292.31 | norm 0.5060 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4078 | train 3.202119 | val 3.474666 | tok/s 2602109.90 | norm 0.5119 | dt 201.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4079 | train 3.127294 | val 3.474666 | tok/s 2629689.37 | norm 0.5408 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4080 | train 3.083299 | val 3.474666 | tok/s 2628641.44 | norm 0.5187 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4081 | train 3.138436 | val 3.474666 | tok/s 2629293.15 | norm 0.4847 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4082 | train 3.146667 | val 3.474666 | tok/s 2628163.89 | norm 0.4888 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4083 | train 3.087675 | val 3.474666 | tok/s 2627190.31 | norm 0.4977 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4084 | train 3.083839 | val 3.474666 | tok/s 2629938.58 | norm 0.4688 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4085 | train 3.215932 | val 3.474666 | tok/s 2620149.46 | norm 0.4339 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4086 | train 3.137260 | val 3.474666 | tok/s 2629012.16 | norm 0.4747 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4087 | train 3.101343 | val 3.474666 | tok/s 2625721.91 | norm 0.5002 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4088 | train 3.092568 | val 3.474666 | tok/s 2630053.79 | norm 0.4536 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4089 | train 3.148871 | val 3.474666 | tok/s 2621738.98 | norm 0.4529 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4090 | train 3.074789 | val 3.474666 | tok/s 2621201.00 | norm 0.5030 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4091 | train 3.172683 | val 3.474666 | tok/s 2628734.45 | norm 0.5086 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4092 | train 3.107376 | val 3.474666 | tok/s 2629435.42 | norm 0.4990 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4093 | train 3.100774 | val 3.474666 | tok/s 2626637.34 | norm 0.4690 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4094 | train 3.057234 | val 3.474666 | tok/s 2623699.74 | norm 0.4386 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4095 | train 3.178931 | val 3.474666 | tok/s 2619051.80 | norm 0.4600 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4096 | train 3.090057 | val 3.474666 | tok/s 2630180.47 | norm 0.4805 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4097 | train 3.109084 | val 3.474666 | tok/s 2627380.73 | norm 0.4466 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4098 | train 3.107969 | val 3.474666 | tok/s 2626966.82 | norm 0.4500 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4099 | train 3.103972 | val 3.474666 | tok/s 2622455.98 | norm 0.4777 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4100 | train 3.155621 | val 3.474666 | tok/s 2625768.93 | norm 0.5031 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4101 | train 3.140571 | val 3.474666 | tok/s 2629039.62 | norm 0.4940 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4102 | train 3.065376 | val 3.474666 | tok/s 2624644.20 | norm 0.4648 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4103 | train 3.290568 | val 3.474666 | tok/s 2629449.88 | norm 0.4753 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4104 | train 3.150777 | val 3.474666 | tok/s 2627525.53 | norm 0.4490 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4105 | train 3.123640 | val 3.474666 | tok/s 2625030.48 | norm 0.4712 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4106 | train 3.426930 | val 3.474666 | tok/s 2627955.70 | norm 0.4817 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4107 | train 3.051175 | val 3.474666 | tok/s 2628253.33 | norm 0.4113 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4108 | train 3.275957 | val 3.474666 | tok/s 2629382.83 | norm 0.4887 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4109 | train 3.161105 | val 3.474666 | tok/s 2626502.93 | norm 0.4568 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4110 | train 3.074341 | val 3.474666 | tok/s 2628527.14 | norm 0.4384 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4111 | train 2.991180 | val 3.474666 | tok/s 2628829.47 | norm 0.4510 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4112 | train 3.040691 | val 3.474666 | tok/s 2625475.22 | norm 0.4769 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4113 | train 3.069623 | val 3.474666 | tok/s 2628198.37 | norm 0.4559 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4114 | train 3.150227 | val 3.474666 | tok/s 2623085.78 | norm 0.4949 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4115 | train 3.040489 | val 3.474666 | tok/s 2622088.88 | norm 0.5120 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4116 | train 3.125497 | val 3.474666 | tok/s 2628147.73 | norm 0.4749 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4117 | train 3.086417 | val 3.474666 | tok/s 2628871.36 | norm 0.4301 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4118 | train 3.092287 | val 3.474666 | tok/s 2629994.65 | norm 0.4424 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4119 | train 3.103560 | val 3.474666 | tok/s 2623384.90 | norm 0.4236 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4120 | train 3.243132 | val 3.474666 | tok/s 2624192.78 | norm 0.4310 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4121 | train 3.080991 | val 3.474666 | tok/s 2626755.83 | norm 0.4518 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4122 | train 3.198051 | val 3.474666 | tok/s 2628488.29 | norm 0.4800 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4123 | train 3.088018 | val 3.474666 | tok/s 2627361.07 | norm 0.4958 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4124 | train 3.092668 | val 3.474666 | tok/s 2626707.56 | norm 0.4306 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4125 | train 3.110956 | val 3.474666 | tok/s 2627704.51 | norm 0.4112 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4126 | train 3.187778 | val 3.474666 | tok/s 2630824.19 | norm 0.4291 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4127 | train 3.143017 | val 3.474666 | tok/s 2627475.81 | norm 0.4522 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4128 | train 3.120178 | val 3.474666 | tok/s 2628416.46 | norm 0.4699 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4129 | train 3.198045 | val 3.474666 | tok/s 2623460.34 | norm 0.4990 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4130 | train 3.066530 | val 3.474666 | tok/s 2625484.45 | norm 0.4703 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4131 | train 3.139372 | val 3.474666 | tok/s 2626460.20 | norm 0.4912 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4132 | train 3.099514 | val 3.474666 | tok/s 2622723.51 | norm 0.4817 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4133 | train 3.180908 | val 3.474666 | tok/s 2630753.53 | norm 0.4889 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4134 | train 3.071571 | val 3.474666 | tok/s 2627575.53 | norm 0.4555 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4135 | train 3.080591 | val 3.474666 | tok/s 2624089.08 | norm 0.4433 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4136 | train 3.413576 | val 3.474666 | tok/s 2624309.64 | norm 0.5506 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4137 | train 3.127054 | val 3.474666 | tok/s 2621035.97 | norm 0.4516 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4138 | train 3.116672 | val 3.474666 | tok/s 2630119.50 | norm 0.4245 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4139 | train 3.005111 | val 3.474666 | tok/s 2623899.72 | norm 0.4700 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4140 | train 3.145674 | val 3.474666 | tok/s 2627960.63 | norm 0.4861 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4141 | train 3.012129 | val 3.474666 | tok/s 2616689.15 | norm 0.4804 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4142 | train 3.262193 | val 3.474666 | tok/s 2623648.30 | norm 0.4748 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4143 | train 3.054194 | val 3.474666 | tok/s 2627642.79 | norm 0.4876 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4144 | train 3.048984 | val 3.474666 | tok/s 2625910.67 | norm 0.5093 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4145 | train 2.991866 | val 3.474666 | tok/s 2628271.56 | norm 0.4592 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4146 | train 3.016868 | val 3.474666 | tok/s 2623992.38 | norm 0.4362 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4147 | train 3.359714 | val 3.474666 | tok/s 2626448.31 | norm 0.4666 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4148 | train 3.038039 | val 3.474666 | tok/s 2629428.93 | norm 0.4555 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4149 | train 3.023633 | val 3.474666 | tok/s 2628418.69 | norm 0.4669 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4150 | train 3.117238 | val 3.474666 | tok/s 2629139.11 | norm 0.5309 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4151 | train 3.075629 | val 3.474666 | tok/s 2624328.48 | norm 0.5086 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4152 | train 3.124288 | val 3.474666 | tok/s 2625913.06 | norm 0.5005 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4153 | train 3.101174 | val 3.474666 | tok/s 2631755.09 | norm 0.5042 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4154 | train 3.023579 | val 3.474666 | tok/s 2626626.62 | norm 0.5029 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4155 | train 3.032027 | val 3.474666 | tok/s 2627934.53 | norm 0.4914 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4156 | train 3.307164 | val 3.474666 | tok/s 2626606.42 | norm 0.5029 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4157 | train 3.184632 | val 3.474666 | tok/s 2627896.57 | norm 0.4913 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4158 | train 3.128372 | val 3.474666 | tok/s 2630134.09 | norm 0.4894 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4159 | train 2.989163 | val 3.474666 | tok/s 2627840.61 | norm 0.4371 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4160 | train 3.057803 | val 3.474666 | tok/s 2630237.38 | norm 0.4413 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4161 | train 3.044938 | val 3.474666 | tok/s 2622050.83 | norm 0.4693 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4162 | train 2.983292 | val 3.474666 | tok/s 2623350.76 | norm 0.5060 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4163 | train 3.050154 | val 3.474666 | tok/s 2631664.71 | norm 0.4729 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4164 | train 3.147455 | val 3.474666 | tok/s 2624085.01 | norm 0.4785 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4165 | train 3.007305 | val 3.474666 | tok/s 2628728.86 | norm 0.4576 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4166 | train 3.063841 | val 3.474666 | tok/s 2627000.15 | norm 0.4067 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4167 | train 3.066353 | val 3.474666 | tok/s 2604864.63 | norm 0.4676 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4168 | train 3.206897 | val 3.474666 | tok/s 2611067.72 | norm 0.4649 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4169 | train 3.052260 | val 3.474666 | tok/s 2626167.42 | norm 0.5121 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4170 | train 3.114962 | val 3.474666 | tok/s 2629551.60 | norm 0.5107 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4171 | train 3.241594 | val 3.474666 | tok/s 2627947.44 | norm 0.4764 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4172 | train 3.170384 | val 3.474666 | tok/s 2628828.30 | norm 0.4438 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4173 | train 3.194450 | val 3.474666 | tok/s 2633391.62 | norm 0.4353 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4174 | train 3.432181 | val 3.474666 | tok/s 2629502.96 | norm 0.7294 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4175 | train 3.058300 | val 3.474666 | tok/s 2630573.29 | norm 0.4709 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4176 | train 2.962301 | val 3.474666 | tok/s 2628653.26 | norm 0.4354 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4177 | train 3.081182 | val 3.474666 | tok/s 2630758.59 | norm 0.4659 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4178 | train 3.053225 | val 3.474666 | tok/s 2632369.78 | norm 0.4519 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4179 | train 3.073386 | val 3.474666 | tok/s 2622054.90 | norm 0.5073 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4180 | train 3.136082 | val 3.474666 | tok/s 2629532.24 | norm 0.4643 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4181 | train 3.197846 | val 3.474666 | tok/s 2626382.09 | norm 0.4044 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4182 | train 3.103761 | val 3.474666 | tok/s 2626528.79 | norm 0.4353 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4183 | train 3.049086 | val 3.474666 | tok/s 2625333.53 | norm 0.4464 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4184 | train 3.098229 | val 3.474666 | tok/s 2624464.63 | norm 0.4510 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4185 | train 3.247792 | val 3.474666 | tok/s 2630009.35 | norm 0.4705 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4186 | train 3.085420 | val 3.474666 | tok/s 2626511.43 | norm 0.4466 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4187 | train 3.169396 | val 3.474666 | tok/s 2629064.08 | norm 0.4407 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4188 | train 3.104598 | val 3.474666 | tok/s 2626810.33 | norm 0.4422 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4189 | train 3.125951 | val 3.474666 | tok/s 2625969.71 | norm 0.4616 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4190 | train 3.096370 | val 3.474666 | tok/s 2631211.38 | norm 0.4894 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4191 | train 3.155951 | val 3.474666 | tok/s 2629496.61 | norm 0.5120 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4192 | train 3.029905 | val 3.474666 | tok/s 2630568.40 | norm 0.4732 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4193 | train 3.105799 | val 3.474666 | tok/s 2617518.29 | norm 0.5143 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4194 | train 2.984717 | val 3.474666 | tok/s 2624328.57 | norm 0.5049 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4195 | train 3.153608 | val 3.474666 | tok/s 2627018.82 | norm 0.4875 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4196 | train 3.143430 | val 3.474666 | tok/s 2627955.29 | norm 0.5445 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4197 | train 3.104874 | val 3.474666 | tok/s 2627642.04 | norm 0.4943 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4198 | train 3.109393 | val 3.474666 | tok/s 2619143.55 | norm 0.4315 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4199 | train 2.937630 | val 3.474666 | tok/s 2620314.69 | norm 0.4398 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4200 | train 3.133496 | val 3.474666 | tok/s 2630171.51 | norm 0.4650 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4201 | train 3.138080 | val 3.474666 | tok/s 2628231.40 | norm 0.5197 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4202 | train 3.092574 | val 3.474666 | tok/s 2623787.07 | norm 0.5159 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4203 | train 3.067829 | val 3.474666 | tok/s 2625530.93 | norm 0.4869 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4204 | train 2.982278 | val 3.474666 | tok/s 2458862.40 | norm 0.4688 | dt 213.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4205 | train 3.163347 | val 3.474666 | tok/s 2621796.17 | norm 0.4609 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4206 | train 3.170632 | val 3.474666 | tok/s 2623057.72 | norm 0.4848 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4207 | train 3.167229 | val 3.474666 | tok/s 2628337.41 | norm 0.4854 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4208 | train 3.055521 | val 3.474666 | tok/s 2585338.85 | norm 0.5098 | dt 202.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4209 | train 3.072772 | val 3.474666 | tok/s 2621314.82 | norm 0.4862 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4210 | train 3.084462 | val 3.474666 | tok/s 2587772.13 | norm 0.4660 | dt 202.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4211 | train 3.182693 | val 3.474666 | tok/s 2480869.12 | norm 0.4826 | dt 211.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4212 | train 3.153249 | val 3.474666 | tok/s 2629783.22 | norm 0.4873 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4213 | train 3.117920 | val 3.474666 | tok/s 2626725.18 | norm 0.4844 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4214 | train 3.029911 | val 3.474666 | tok/s 2630219.71 | norm 0.5261 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4215 | train 3.010102 | val 3.474666 | tok/s 2624393.23 | norm 0.4984 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4216 | train 3.084571 | val 3.474666 | tok/s 2622682.14 | norm 0.4622 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4217 | train 3.079857 | val 3.474666 | tok/s 2588074.79 | norm 0.4706 | dt 202.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4218 | train 2.986187 | val 3.474666 | tok/s 2626832.52 | norm 0.4781 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4219 | train 3.076520 | val 3.474666 | tok/s 2584906.32 | norm 0.4932 | dt 202.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4220 | train 3.134299 | val 3.474666 | tok/s 2623531.98 | norm 0.4885 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4221 | train 3.130153 | val 3.474666 | tok/s 2626356.45 | norm 0.4872 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4222 | train 3.007183 | val 3.474666 | tok/s 2627959.31 | norm 0.4969 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4223 | train 3.102662 | val 3.474666 | tok/s 2627816.01 | norm 0.5259 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4224 | train 3.014291 | val 3.474666 | tok/s 2622381.66 | norm 0.5132 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4225 | train 3.348821 | val 3.474666 | tok/s 2625953.62 | norm 0.5765 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4226 | train 2.980878 | val 3.474666 | tok/s 2626973.55 | norm 0.4405 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4227 | train 3.118771 | val 3.474666 | tok/s 2631471.38 | norm 0.4841 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4228 | train 3.093606 | val 3.474666 | tok/s 2625304.87 | norm 0.4624 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4229 | train 3.196871 | val 3.474666 | tok/s 2629288.43 | norm 0.4169 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4230 | train 3.160078 | val 3.474666 | tok/s 2624308.93 | norm 0.4837 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4231 | train 3.178348 | val 3.474666 | tok/s 2625858.37 | norm 0.5788 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4232 | train 3.028826 | val 3.474666 | tok/s 2629683.23 | norm 0.5071 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4233 | train 3.078071 | val 3.474666 | tok/s 2622739.10 | norm 0.4964 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4234 | train 3.016443 | val 3.474666 | tok/s 2631260.14 | norm 0.5137 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4235 | train 3.149564 | val 3.474666 | tok/s 2625991.66 | norm 0.4474 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4236 | train 3.308082 | val 3.474666 | tok/s 2627890.22 | norm 0.8131 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4237 | train 3.034378 | val 3.474666 | tok/s 2626936.05 | norm 0.4673 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4238 | train 3.144562 | val 3.474666 | tok/s 2622350.03 | norm 0.4638 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4239 | train 3.138522 | val 3.474666 | tok/s 2631452.77 | norm 0.4283 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4240 | train 3.147297 | val 3.474666 | tok/s 2624355.17 | norm 0.4245 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4241 | train 3.075094 | val 3.474666 | tok/s 2627469.17 | norm 0.4127 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4242 | train 3.152748 | val 3.474666 | tok/s 2623127.51 | norm 0.3841 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4243 | train 3.184951 | val 3.474666 | tok/s 2624631.24 | norm 0.4039 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4244 | train 3.141166 | val 3.474666 | tok/s 2628323.57 | norm 0.3906 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4245 | train 3.081782 | val 3.474666 | tok/s 2628382.28 | norm 0.4355 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4246 | train 3.103992 | val 3.474666 | tok/s 2631714.44 | norm 0.4524 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4247 | train 3.053326 | val 3.474666 | tok/s 2618758.51 | norm 0.4348 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4248 | train 3.065333 | val 3.474666 | tok/s 2626426.10 | norm 0.4381 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4249 | train 3.134371 | val 3.474666 | tok/s 2624926.91 | norm 0.4614 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 4250 | val_loss 3.469947 step 4250 | train 2.991404 | val 3.469947 | tok/s 2558671.59 | norm 0.5268 | dt 204.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4251 | train 3.163755 | val 3.469947 | tok/s 2625473.15 | norm 0.5350 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4252 | train 3.101614 | val 3.469947 | tok/s 2627403.40 | norm 0.4863 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4253 | train 3.106287 | val 3.469947 | tok/s 2621897.17 | norm 0.4816 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4254 | train 3.054774 | val 3.469947 | tok/s 2625002.18 | norm 0.4363 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4255 | train 3.224479 | val 3.469947 | tok/s 2614718.46 | norm 0.4119 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4256 | train 3.201961 | val 3.469947 | tok/s 2622857.35 | norm 0.4076 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4257 | train 3.108784 | val 3.469947 | tok/s 2621789.20 | norm 0.4592 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4258 | train 3.027126 | val 3.469947 | tok/s 2622273.40 | norm 0.5121 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4259 | train 3.078639 | val 3.469947 | tok/s 2621739.09 | norm 0.4989 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4260 | train 3.155746 | val 3.469947 | tok/s 2625310.02 | norm 0.4706 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4261 | train 3.284892 | val 3.469947 | tok/s 2624457.22 | norm 0.4794 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4262 | train 3.161060 | val 3.469947 | tok/s 2625743.90 | norm 0.4927 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4263 | train 3.131147 | val 3.469947 | tok/s 2622571.86 | norm 0.4972 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4264 | train 3.072032 | val 3.469947 | tok/s 2624167.65 | norm 0.4701 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4265 | train 3.079086 | val 3.469947 | tok/s 2629311.70 | norm 0.4259 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4266 | train 2.956415 | val 3.469947 | tok/s 2623627.16 | norm 0.4588 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4267 | train 2.982734 | val 3.469947 | tok/s 2623481.33 | norm 0.4916 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4268 | train 3.034415 | val 3.469947 | tok/s 2621782.50 | norm 0.4657 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4269 | train 3.038804 | val 3.469947 | tok/s 2626270.56 | norm 0.4770 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4270 | train 3.066756 | val 3.469947 | tok/s 2626343.38 | norm 0.4797 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4271 | train 3.104179 | val 3.469947 | tok/s 2628500.48 | norm 0.4290 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4272 | train 3.090953 | val 3.469947 | tok/s 2629525.87 | norm 0.4968 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4273 | train 3.122000 | val 3.469947 | tok/s 2627897.26 | norm 0.5005 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4274 | train 3.043814 | val 3.469947 | tok/s 2627108.13 | norm 0.4846 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4275 | train 3.096706 | val 3.469947 | tok/s 2628411.82 | norm 0.5264 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4276 | train 3.161856 | val 3.469947 | tok/s 2626266.38 | norm 0.4601 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4277 | train 3.102280 | val 3.469947 | tok/s 2629395.37 | norm 0.4678 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4278 | train 3.343330 | val 3.469947 | tok/s 2631299.57 | norm 0.5627 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4279 | train 3.143827 | val 3.469947 | tok/s 2627777.15 | norm 0.4354 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4280 | train 3.105550 | val 3.469947 | tok/s 2625592.85 | norm 0.4952 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4281 | train 3.068266 | val 3.469947 | tok/s 2622757.82 | norm 0.4655 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4282 | train 3.122767 | val 3.469947 | tok/s 2629539.89 | norm 0.4345 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4283 | train 3.103676 | val 3.469947 | tok/s 2627843.78 | norm 0.4548 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4284 | train 3.101705 | val 3.469947 | tok/s 2628788.42 | norm 0.4823 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4285 | train 3.153203 | val 3.469947 | tok/s 2621535.57 | norm 0.4870 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4286 | train 3.226909 | val 3.469947 | tok/s 2623418.45 | norm 0.5401 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4287 | train 3.200390 | val 3.469947 | tok/s 2629389.49 | norm 0.4504 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4288 | train 3.181317 | val 3.469947 | tok/s 2624739.17 | norm 0.4668 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4289 | train 3.132342 | val 3.469947 | tok/s 2627796.76 | norm 0.4919 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4290 | train 3.132136 | val 3.469947 | tok/s 2618091.06 | norm 0.4485 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4291 | train 3.116023 | val 3.469947 | tok/s 2623641.05 | norm 0.4338 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4292 | train 3.075958 | val 3.469947 | tok/s 2627039.88 | norm 0.4644 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4293 | train 2.955117 | val 3.469947 | tok/s 2624741.75 | norm 0.5516 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4294 | train 3.034712 | val 3.469947 | tok/s 2629179.87 | norm 0.5234 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4295 | train 3.099283 | val 3.469947 | tok/s 2624172.09 | norm 0.4831 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4296 | train 3.035331 | val 3.469947 | tok/s 2627092.69 | norm 0.4879 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4297 | train 3.059155 | val 3.469947 | tok/s 2629776.21 | norm 0.4649 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4298 | train 2.989062 | val 3.469947 | tok/s 2626336.54 | norm 0.4905 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4299 | train 3.163415 | val 3.469947 | tok/s 2626769.99 | norm 0.4769 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4300 | train 3.115707 | val 3.469947 | tok/s 2627576.69 | norm 0.4739 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4301 | train 2.991461 | val 3.469947 | tok/s 2630686.88 | norm 0.4721 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4302 | train 3.018874 | val 3.469947 | tok/s 2628133.83 | norm 0.4590 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4303 | train 3.069596 | val 3.469947 | tok/s 2626933.11 | norm 0.4297 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4304 | train 3.110800 | val 3.469947 | tok/s 2628351.26 | norm 0.4013 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4305 | train 3.079787 | val 3.469947 | tok/s 2629287.19 | norm 0.4590 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4306 | train 3.009379 | val 3.469947 | tok/s 2628374.97 | norm 0.4763 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4307 | train 3.155686 | val 3.469947 | tok/s 2629373.78 | norm 0.4514 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4308 | train 3.000546 | val 3.469947 | tok/s 2626037.25 | norm 0.3917 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4309 | train 3.071940 | val 3.469947 | tok/s 2625937.22 | norm 0.4503 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4310 | train 3.081511 | val 3.469947 | tok/s 2621773.73 | norm 0.5293 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4311 | train 3.146103 | val 3.469947 | tok/s 2628793.09 | norm 0.5070 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4312 | train 3.217569 | val 3.469947 | tok/s 2623995.48 | norm 0.4922 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4313 | train 3.169266 | val 3.469947 | tok/s 2623382.61 | norm 0.4787 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4314 | train 3.118476 | val 3.469947 | tok/s 2623219.63 | norm 0.4711 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4315 | train 3.207952 | val 3.469947 | tok/s 2620783.44 | norm 0.4952 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4316 | train 3.054677 | val 3.469947 | tok/s 2624728.98 | norm 0.5200 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4317 | train 3.097406 | val 3.469947 | tok/s 2626215.98 | norm 0.5055 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4318 | train 3.031395 | val 3.469947 | tok/s 2626522.15 | norm 0.4683 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4319 | train 2.969620 | val 3.469947 | tok/s 2630439.15 | norm 0.4510 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4320 | train 3.022406 | val 3.469947 | tok/s 2629945.38 | norm 0.4535 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4321 | train 3.114261 | val 3.469947 | tok/s 2624199.03 | norm 0.4786 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4322 | train 3.038811 | val 3.469947 | tok/s 2627313.65 | norm 0.4712 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4323 | train 3.095826 | val 3.469947 | tok/s 2618718.39 | norm 0.4399 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4324 | train 3.048596 | val 3.469947 | tok/s 2627658.77 | norm 0.4349 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4325 | train 2.991736 | val 3.469947 | tok/s 2619698.52 | norm 0.4809 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4326 | train 3.138762 | val 3.469947 | tok/s 2625057.37 | norm 0.4716 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4327 | train 3.056391 | val 3.469947 | tok/s 2625051.41 | norm 0.4826 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4328 | train 3.125579 | val 3.469947 | tok/s 2626351.09 | norm 0.4788 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4329 | train 3.214528 | val 3.469947 | tok/s 2630294.57 | norm 0.5049 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4330 | train 3.086431 | val 3.469947 | tok/s 2627820.85 | norm 0.4345 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4331 | train 3.053663 | val 3.469947 | tok/s 2628807.83 | norm 0.4575 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4332 | train 3.085304 | val 3.469947 | tok/s 2622390.04 | norm 0.4505 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4333 | train 3.073470 | val 3.469947 | tok/s 2625551.05 | norm 0.4367 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4334 | train 3.105134 | val 3.469947 | tok/s 2627677.34 | norm 0.4509 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4335 | train 3.188713 | val 3.469947 | tok/s 2621462.98 | norm 0.4493 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4336 | train 3.260626 | val 3.469947 | tok/s 2628803.79 | norm 0.4788 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4337 | train 2.962821 | val 3.469947 | tok/s 2618944.93 | norm 0.5118 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4338 | train 3.145995 | val 3.469947 | tok/s 2623444.49 | norm 0.4401 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4339 | train 3.148013 | val 3.469947 | tok/s 2630245.55 | norm 0.4287 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4340 | train 2.936148 | val 3.469947 | tok/s 2626908.29 | norm 0.4536 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4341 | train 3.152746 | val 3.469947 | tok/s 2628618.05 | norm 0.4183 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4342 | train 3.013868 | val 3.469947 | tok/s 2627073.96 | norm 0.3965 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4343 | train 3.061918 | val 3.469947 | tok/s 2622983.23 | norm 0.4110 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4344 | train 3.051267 | val 3.469947 | tok/s 2625629.09 | norm 0.3966 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4345 | train 3.348297 | val 3.469947 | tok/s 2627241.62 | norm 0.4632 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4346 | train 3.207289 | val 3.469947 | tok/s 2628782.19 | norm 0.5024 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4347 | train 3.111490 | val 3.469947 | tok/s 2623227.31 | norm 0.4288 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4348 | train 3.032914 | val 3.469947 | tok/s 2627408.47 | norm 0.4539 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4349 | train 3.042769 | val 3.469947 | tok/s 2627857.07 | norm 0.4890 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4350 | train 3.077732 | val 3.469947 | tok/s 2626656.96 | norm 0.4709 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4351 | train 3.130022 | val 3.469947 | tok/s 2628534.40 | norm 0.4954 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4352 | train 3.188551 | val 3.469947 | tok/s 2612080.20 | norm 0.5053 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4353 | train 3.046332 | val 3.469947 | tok/s 2623160.03 | norm 0.4514 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4354 | train 3.006489 | val 3.469947 | tok/s 2627345.89 | norm 0.4632 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4355 | train 3.073419 | val 3.469947 | tok/s 2624931.75 | norm 0.4706 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4356 | train 3.104488 | val 3.469947 | tok/s 2627146.33 | norm 0.4654 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4357 | train 3.092411 | val 3.469947 | tok/s 2622631.24 | norm 0.4436 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4358 | train 3.040129 | val 3.469947 | tok/s 2602827.18 | norm 0.4386 | dt 201.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4359 | train 3.128320 | val 3.469947 | tok/s 2630927.57 | norm 0.4381 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4360 | train 3.091694 | val 3.469947 | tok/s 2620780.59 | norm 0.4672 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4361 | train 3.257383 | val 3.469947 | tok/s 2630147.47 | norm 0.5329 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4362 | train 2.991048 | val 3.469947 | tok/s 2627817.59 | norm 0.4733 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4363 | train 3.080656 | val 3.469947 | tok/s 2630834.04 | norm 0.5013 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4364 | train 3.111375 | val 3.469947 | tok/s 2630023.41 | norm 0.5248 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4365 | train 3.113556 | val 3.469947 | tok/s 2624579.51 | norm 0.4731 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4366 | train 3.192776 | val 3.469947 | tok/s 2625017.90 | norm 0.4922 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4367 | train 3.124928 | val 3.469947 | tok/s 2628152.35 | norm 0.4908 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4368 | train 3.103025 | val 3.469947 | tok/s 2629470.55 | norm 0.4583 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4369 | train 3.041727 | val 3.469947 | tok/s 2626626.31 | norm 0.4396 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4370 | train 3.131355 | val 3.469947 | tok/s 2622682.42 | norm 0.4595 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4371 | train 3.068196 | val 3.469947 | tok/s 2630373.91 | norm 0.4828 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4372 | train 3.062324 | val 3.469947 | tok/s 2626227.72 | norm 0.4288 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4373 | train 3.065355 | val 3.469947 | tok/s 2629410.76 | norm 0.4412 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4374 | train 3.074648 | val 3.469947 | tok/s 2626589.15 | norm 0.4639 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4375 | train 3.210862 | val 3.469947 | tok/s 2623321.85 | norm 0.4935 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4376 | train 3.153282 | val 3.469947 | tok/s 2629324.40 | norm 0.4772 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4377 | train 2.936454 | val 3.469947 | tok/s 2626958.56 | norm 0.4968 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4378 | train 3.167997 | val 3.469947 | tok/s 2628578.27 | norm 0.4947 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4379 | train 3.126941 | val 3.469947 | tok/s 2618421.65 | norm 0.4583 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4380 | train 3.045330 | val 3.469947 | tok/s 2626687.54 | norm 0.4575 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4381 | train 3.024808 | val 3.469947 | tok/s 2631742.23 | norm 0.4749 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4382 | train 3.100011 | val 3.469947 | tok/s 2482237.75 | norm 0.4431 | dt 211.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4383 | train 3.187102 | val 3.469947 | tok/s 2548248.37 | norm 0.4442 | dt 205.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4384 | train 3.105973 | val 3.469947 | tok/s 2625757.65 | norm 0.4313 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4385 | train 3.065076 | val 3.469947 | tok/s 2627038.07 | norm 0.4763 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4386 | train 3.025235 | val 3.469947 | tok/s 2630432.92 | norm 0.4339 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4387 | train 3.096040 | val 3.469947 | tok/s 2625786.92 | norm 0.3911 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4388 | train 3.164403 | val 3.469947 | tok/s 2630135.54 | norm 0.5397 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4389 | train 3.280627 | val 3.469947 | tok/s 2622276.52 | norm 0.5130 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4390 | train 2.985327 | val 3.469947 | tok/s 2623144.10 | norm 0.4833 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4391 | train 3.101149 | val 3.469947 | tok/s 2628378.20 | norm 0.5376 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4392 | train 2.995436 | val 3.469947 | tok/s 2622298.74 | norm 0.5252 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4393 | train 3.110347 | val 3.469947 | tok/s 2630784.98 | norm 0.5072 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4394 | train 3.138924 | val 3.469947 | tok/s 2621621.32 | norm 0.4583 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4395 | train 3.104944 | val 3.469947 | tok/s 2626434.51 | norm 0.4420 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4396 | train 3.100566 | val 3.469947 | tok/s 2631599.66 | norm 0.4446 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4397 | train 3.119422 | val 3.469947 | tok/s 2626959.82 | norm 0.4637 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4398 | train 3.094592 | val 3.469947 | tok/s 2630928.56 | norm 0.4653 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4399 | train 3.028292 | val 3.469947 | tok/s 2626665.91 | norm 0.4522 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4400 | train 3.155982 | val 3.469947 | tok/s 2629820.22 | norm 0.4880 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4401 | train 3.089651 | val 3.469947 | tok/s 2631699.02 | norm 0.4374 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4402 | train 2.912509 | val 3.469947 | tok/s 2625648.52 | norm 0.4300 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4403 | train 3.021696 | val 3.469947 | tok/s 2625185.44 | norm 0.4819 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4404 | train 3.134950 | val 3.469947 | tok/s 2623025.63 | norm 0.4218 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4405 | train 3.058156 | val 3.469947 | tok/s 2629570.04 | norm 0.4418 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4406 | train 3.056206 | val 3.469947 | tok/s 2629387.72 | norm 0.4589 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4407 | train 3.052402 | val 3.469947 | tok/s 2623275.72 | norm 0.4167 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4408 | train 3.081777 | val 3.469947 | tok/s 2631640.95 | norm 0.4551 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4409 | train 3.018186 | val 3.469947 | tok/s 2625977.10 | norm 0.4866 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4410 | train 3.055611 | val 3.469947 | tok/s 2629732.41 | norm 0.4881 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4411 | train 2.983941 | val 3.469947 | tok/s 2628309.89 | norm 0.4679 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4412 | train 3.086759 | val 3.469947 | tok/s 2625938.10 | norm 0.4634 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4413 | train 3.150860 | val 3.469947 | tok/s 2629411.56 | norm 0.5452 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4414 | train 3.043255 | val 3.469947 | tok/s 2625522.41 | norm 0.4973 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4415 | train 2.966047 | val 3.469947 | tok/s 2627274.59 | norm 0.4025 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4416 | train 3.296993 | val 3.469947 | tok/s 2618268.58 | norm 0.4837 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4417 | train 3.081480 | val 3.469947 | tok/s 2626153.89 | norm 0.4545 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4418 | train 3.063203 | val 3.469947 | tok/s 2561538.74 | norm 0.4057 | dt 204.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4419 | train 3.189760 | val 3.469947 | tok/s 2601402.92 | norm 0.4435 | dt 201.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4420 | train 3.017637 | val 3.469947 | tok/s 2631541.67 | norm 0.4641 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4421 | train 3.059476 | val 3.469947 | tok/s 2626555.72 | norm 0.4756 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4422 | train 3.023419 | val 3.469947 | tok/s 2629303.76 | norm 0.5038 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4423 | train 3.092478 | val 3.469947 | tok/s 2632104.89 | norm 0.4857 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4424 | train 3.166280 | val 3.469947 | tok/s 2628189.15 | norm 0.4779 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4425 | train 3.226577 | val 3.469947 | tok/s 2491830.54 | norm 0.4446 | dt 210.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4426 | train 3.143061 | val 3.469947 | tok/s 2625438.70 | norm 0.4506 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4427 | train 3.061903 | val 3.469947 | tok/s 2623120.99 | norm 0.4283 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4428 | train 3.106317 | val 3.469947 | tok/s 2626558.27 | norm 0.4381 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4429 | train 3.131909 | val 3.469947 | tok/s 2628066.54 | norm 0.4578 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4430 | train 3.084618 | val 3.469947 | tok/s 2629860.19 | norm 0.4921 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4431 | train 3.131735 | val 3.469947 | tok/s 2625511.14 | norm 0.4779 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4432 | train 2.969937 | val 3.469947 | tok/s 2605335.09 | norm 0.4558 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4433 | train 3.094506 | val 3.469947 | tok/s 2627747.84 | norm 0.4330 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4434 | train 3.145050 | val 3.469947 | tok/s 2625438.95 | norm 0.4101 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4435 | train 3.115409 | val 3.469947 | tok/s 2627171.52 | norm 0.4430 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4436 | train 3.079922 | val 3.469947 | tok/s 2624331.69 | norm 0.4622 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4437 | train 3.108586 | val 3.469947 | tok/s 2616292.86 | norm 0.4542 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4438 | train 3.067702 | val 3.469947 | tok/s 2627381.23 | norm 0.4681 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4439 | train 3.132751 | val 3.469947 | tok/s 2628493.62 | norm 0.4486 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4440 | train 3.086940 | val 3.469947 | tok/s 2626532.18 | norm 0.4549 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4441 | train 3.065387 | val 3.469947 | tok/s 2628845.09 | norm 0.4614 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4442 | train 3.017055 | val 3.469947 | tok/s 2629037.80 | norm 0.4406 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4443 | train 3.054714 | val 3.469947 | tok/s 2595608.66 | norm 0.4610 | dt 201.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4444 | train 3.094128 | val 3.469947 | tok/s 2622184.78 | norm 0.4489 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4445 | train 2.997672 | val 3.469947 | tok/s 2629190.08 | norm 0.5003 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4446 | train 2.995322 | val 3.469947 | tok/s 2624972.83 | norm 0.5090 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4447 | train 3.079017 | val 3.469947 | tok/s 2628222.80 | norm 0.4650 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4448 | train 3.140914 | val 3.469947 | tok/s 2629890.53 | norm 0.4623 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4449 | train 3.122539 | val 3.469947 | tok/s 2627811.89 | norm 0.4194 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4450 | train 3.117695 | val 3.469947 | tok/s 2628151.01 | norm 0.4346 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4451 | train 3.071664 | val 3.469947 | tok/s 2584561.28 | norm 0.4645 | dt 202.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4452 | train 3.177457 | val 3.469947 | tok/s 2627170.65 | norm 0.5012 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4453 | train 3.205848 | val 3.469947 | tok/s 2470139.14 | norm 0.5016 | dt 212.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4454 | train 3.082885 | val 3.469947 | tok/s 2592285.38 | norm 0.4193 | dt 202.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4455 | train 3.104915 | val 3.469947 | tok/s 2630492.15 | norm 0.4242 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4456 | train 3.126054 | val 3.469947 | tok/s 2628024.53 | norm 0.4882 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4457 | train 3.151751 | val 3.469947 | tok/s 2630643.02 | norm 0.4749 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4458 | train 3.088020 | val 3.469947 | tok/s 2626462.39 | norm 0.4286 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4459 | train 3.071853 | val 3.469947 | tok/s 2626415.90 | norm 0.4742 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4460 | train 3.055527 | val 3.469947 | tok/s 2477525.57 | norm 0.4724 | dt 211.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4461 | train 3.068859 | val 3.469947 | tok/s 2626925.69 | norm 0.5136 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4462 | train 3.181906 | val 3.469947 | tok/s 2625832.43 | norm 0.5110 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4463 | train 3.116241 | val 3.469947 | tok/s 2622370.76 | norm 0.4603 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4464 | train 3.037880 | val 3.469947 | tok/s 2626726.71 | norm 0.4550 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4465 | train 3.056367 | val 3.469947 | tok/s 2628190.35 | norm 0.4532 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4466 | train 3.153730 | val 3.469947 | tok/s 2627840.55 | norm 0.4808 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4467 | train 3.133994 | val 3.469947 | tok/s 2618206.12 | norm 0.4755 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4468 | train 2.972683 | val 3.469947 | tok/s 2608885.03 | norm 0.4224 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4469 | train 3.103636 | val 3.469947 | tok/s 2627907.24 | norm 0.4447 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4470 | train 3.050062 | val 3.469947 | tok/s 2631812.76 | norm 0.4533 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4471 | train 3.081069 | val 3.469947 | tok/s 2627486.79 | norm 0.4687 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4472 | train 3.050960 | val 3.469947 | tok/s 2627480.87 | norm 0.4627 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4473 | train 3.152873 | val 3.469947 | tok/s 2625373.97 | norm 0.4707 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4474 | train 3.110784 | val 3.469947 | tok/s 2628732.60 | norm 0.5267 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4475 | train 3.039944 | val 3.469947 | tok/s 2627930.78 | norm 0.4100 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4476 | train 3.134666 | val 3.469947 | tok/s 2624333.35 | norm 0.4002 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4477 | train 3.143157 | val 3.469947 | tok/s 2626683.89 | norm 0.4452 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4478 | train 3.368274 | val 3.469947 | tok/s 2628152.63 | norm 0.4950 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4479 | train 3.284813 | val 3.469947 | tok/s 2629967.21 | norm 0.4909 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4480 | train 2.978667 | val 3.469947 | tok/s 2632485.57 | norm 0.4308 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4481 | train 3.065672 | val 3.469947 | tok/s 2627007.63 | norm 0.4119 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4482 | train 3.024376 | val 3.469947 | tok/s 2633453.95 | norm 0.4616 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4483 | train 3.251286 | val 3.469947 | tok/s 2624294.71 | norm 0.5143 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4484 | train 3.185360 | val 3.469947 | tok/s 2629577.69 | norm 0.4646 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4485 | train 3.128538 | val 3.469947 | tok/s 2630390.94 | norm 0.4632 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4486 | train 3.039714 | val 3.469947 | tok/s 2626744.80 | norm 0.4665 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4487 | train 3.112534 | val 3.469947 | tok/s 2629395.34 | norm 0.4618 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4488 | train 3.063535 | val 3.469947 | tok/s 2624537.91 | norm 0.4178 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4489 | train 3.011858 | val 3.469947 | tok/s 2628634.40 | norm 0.4059 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4490 | train 3.109618 | val 3.469947 | tok/s 2621245.03 | norm 0.4460 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4491 | train 3.190548 | val 3.469947 | tok/s 2620780.40 | norm 0.4398 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4492 | train 3.076134 | val 3.469947 | tok/s 2629647.16 | norm 0.4257 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4493 | train 3.171935 | val 3.469947 | tok/s 2624408.03 | norm 0.4643 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4494 | train 3.141756 | val 3.469947 | tok/s 2625804.70 | norm 0.4082 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4495 | train 3.151942 | val 3.469947 | tok/s 2625953.98 | norm 0.4639 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4496 | train 3.015954 | val 3.469947 | tok/s 2626007.39 | norm 0.4388 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4497 | train 3.146610 | val 3.469947 | tok/s 2632138.21 | norm 0.3751 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4498 | train 3.005013 | val 3.469947 | tok/s 2630193.77 | norm 0.4445 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4499 | train 3.159822 | val 3.469947 | tok/s 2627147.12 | norm 0.4844 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 4500 | val_loss 3.466599 step 4500 | train 3.113365 | val 3.466599 | tok/s 2504298.11 | norm 0.4774 | dt 209.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4501 | train 3.125475 | val 3.466599 | tok/s 2619852.78 | norm 0.4803 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4502 | train 3.023481 | val 3.466599 | tok/s 2627413.83 | norm 0.3991 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4503 | train 3.133061 | val 3.466599 | tok/s 2632221.79 | norm 0.3605 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4504 | train 3.136459 | val 3.466599 | tok/s 2630264.21 | norm 0.4167 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4505 | train 3.213861 | val 3.466599 | tok/s 2503789.20 | norm 0.4977 | dt 209.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4506 | train 3.116456 | val 3.466599 | tok/s 2618793.85 | norm 0.4966 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4507 | train 3.080353 | val 3.466599 | tok/s 2626906.58 | norm 0.4634 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4508 | train 3.088466 | val 3.466599 | tok/s 2635662.40 | norm 0.5210 | dt 198.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4509 | train 3.080078 | val 3.466599 | tok/s 2630527.55 | norm 0.7108 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4510 | train 3.070358 | val 3.466599 | tok/s 2630775.70 | norm 0.4011 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4511 | train 3.152650 | val 3.466599 | tok/s 2627639.81 | norm 0.3941 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4512 | train 3.128158 | val 3.466599 | tok/s 2626265.76 | norm 0.4932 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4513 | train 3.092686 | val 3.466599 | tok/s 2629635.42 | norm 0.4873 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4514 | train 2.968514 | val 3.466599 | tok/s 2631759.72 | norm 0.4544 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4515 | train 3.082916 | val 3.466599 | tok/s 2632963.07 | norm 0.5154 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4516 | train 3.101633 | val 3.466599 | tok/s 2626514.04 | norm 0.4603 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4517 | train 3.081240 | val 3.466599 | tok/s 2629497.04 | norm 0.4640 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4518 | train 3.056366 | val 3.466599 | tok/s 2630955.67 | norm 0.4353 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4519 | train 3.044222 | val 3.466599 | tok/s 2631051.10 | norm 0.4217 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4520 | train 3.055727 | val 3.466599 | tok/s 2630955.44 | norm 0.4437 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4521 | train 3.164087 | val 3.466599 | tok/s 2623076.22 | norm 0.4550 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4522 | train 3.078093 | val 3.466599 | tok/s 2624441.32 | norm 0.4726 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4523 | train 3.060314 | val 3.466599 | tok/s 2629840.50 | norm 0.4963 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4524 | train 3.065416 | val 3.466599 | tok/s 2626624.17 | norm 0.4853 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4525 | train 3.012648 | val 3.466599 | tok/s 2631867.21 | norm 0.4959 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4526 | train 3.013923 | val 3.466599 | tok/s 2624896.21 | norm 0.5474 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4527 | train 3.031022 | val 3.466599 | tok/s 2625817.80 | norm 0.5432 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4528 | train 3.014974 | val 3.466599 | tok/s 2634474.14 | norm 0.4829 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4529 | train 2.982551 | val 3.466599 | tok/s 2630670.26 | norm 0.5043 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4530 | train 3.011775 | val 3.466599 | tok/s 2628415.84 | norm 0.4979 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4531 | train 3.072814 | val 3.466599 | tok/s 2625911.14 | norm 0.4728 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4532 | train 3.145946 | val 3.466599 | tok/s 2626508.33 | norm 0.5137 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4533 | train 3.197336 | val 3.466599 | tok/s 2630853.07 | norm 0.4984 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4534 | train 3.190837 | val 3.466599 | tok/s 2627694.82 | norm 0.4722 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4535 | train 3.021588 | val 3.466599 | tok/s 2627095.44 | norm 0.5796 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4536 | train 3.061948 | val 3.466599 | tok/s 2621491.79 | norm 0.4359 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4537 | train 3.178623 | val 3.466599 | tok/s 2623559.19 | norm 0.4329 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4538 | train 3.104060 | val 3.466599 | tok/s 2629664.82 | norm 0.4542 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4539 | train 3.117957 | val 3.466599 | tok/s 2627389.03 | norm 0.4604 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4540 | train 3.218470 | val 3.466599 | tok/s 2631421.19 | norm 0.4421 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4541 | train 3.023108 | val 3.466599 | tok/s 2629072.41 | norm 0.4701 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4542 | train 3.033247 | val 3.466599 | tok/s 2631062.29 | norm 0.4525 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4543 | train 3.148296 | val 3.466599 | tok/s 2626003.96 | norm 0.4586 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4544 | train 3.073253 | val 3.466599 | tok/s 2624052.82 | norm 0.4797 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4545 | train 3.000531 | val 3.466599 | tok/s 2629058.83 | norm 0.4805 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4546 | train 3.075361 | val 3.466599 | tok/s 2626033.25 | norm 0.4779 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4547 | train 3.027969 | val 3.466599 | tok/s 2629228.88 | norm 0.4914 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4548 | train 3.137973 | val 3.466599 | tok/s 2625934.47 | norm 0.4648 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4549 | train 3.083350 | val 3.466599 | tok/s 2457377.42 | norm 0.4287 | dt 213.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4550 | train 3.054726 | val 3.466599 | tok/s 2598341.18 | norm 0.5043 | dt 201.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4551 | train 3.172234 | val 3.466599 | tok/s 2603124.70 | norm 0.5753 | dt 201.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4552 | train 3.179624 | val 3.466599 | tok/s 2631136.81 | norm 0.4660 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4553 | train 3.108456 | val 3.466599 | tok/s 2624467.06 | norm 0.4080 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4554 | train 3.008682 | val 3.466599 | tok/s 2627208.54 | norm 0.4501 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4555 | train 3.007310 | val 3.466599 | tok/s 2503327.37 | norm 0.4832 | dt 209.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4556 | train 2.989418 | val 3.466599 | tok/s 2481972.39 | norm 0.4594 | dt 211.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4557 | train 3.114396 | val 3.466599 | tok/s 2598058.68 | norm 0.4817 | dt 201.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4558 | train 3.096871 | val 3.466599 | tok/s 2480739.84 | norm 0.5158 | dt 211.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4559 | train 3.098460 | val 3.466599 | tok/s 2594334.56 | norm 0.4744 | dt 202.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4560 | train 3.170857 | val 3.466599 | tok/s 2593186.64 | norm 0.4378 | dt 202.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4561 | train 3.120709 | val 3.466599 | tok/s 2627170.61 | norm 0.4223 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4562 | train 3.059778 | val 3.466599 | tok/s 2623811.41 | norm 0.4495 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4563 | train 3.200421 | val 3.466599 | tok/s 2625280.05 | norm 0.4556 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4564 | train 3.116077 | val 3.466599 | tok/s 2615931.83 | norm 0.4218 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4565 | train 2.939922 | val 3.466599 | tok/s 2596757.95 | norm 0.4177 | dt 201.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4566 | train 3.031991 | val 3.466599 | tok/s 2451427.75 | norm 0.4545 | dt 213.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4567 | train 3.156438 | val 3.466599 | tok/s 2620164.03 | norm 0.4604 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4568 | train 3.015726 | val 3.466599 | tok/s 2623808.88 | norm 0.4914 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4569 | train 3.168585 | val 3.466599 | tok/s 2628418.52 | norm 0.4574 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4570 | train 3.189352 | val 3.466599 | tok/s 2624835.13 | norm 0.4420 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4571 | train 3.008389 | val 3.466599 | tok/s 2614990.26 | norm 0.4639 | dt 200.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4572 | train 3.040485 | val 3.466599 | tok/s 2626721.77 | norm 0.4978 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4573 | train 3.048023 | val 3.466599 | tok/s 2625146.10 | norm 0.4867 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4574 | train 3.131981 | val 3.466599 | tok/s 2627520.16 | norm 0.4971 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4575 | train 3.054378 | val 3.466599 | tok/s 2617973.99 | norm 0.4971 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4576 | train 3.076137 | val 3.466599 | tok/s 2626967.45 | norm 0.4580 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4577 | train 3.118893 | val 3.466599 | tok/s 2625350.14 | norm 0.4348 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4578 | train 3.205050 | val 3.466599 | tok/s 2441810.27 | norm 0.4407 | dt 214.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4579 | train 3.240038 | val 3.466599 | tok/s 2615822.19 | norm 0.3806 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4580 | train 3.155742 | val 3.466599 | tok/s 2460432.02 | norm 0.4347 | dt 213.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4581 | train 3.031255 | val 3.466599 | tok/s 2625972.13 | norm 0.4612 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4582 | train 3.070168 | val 3.466599 | tok/s 2630753.25 | norm 0.4539 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4583 | train 3.068155 | val 3.466599 | tok/s 2625389.83 | norm 0.4949 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4584 | train 3.057486 | val 3.466599 | tok/s 2633567.67 | norm 0.4515 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4585 | train 3.032682 | val 3.466599 | tok/s 2631286.78 | norm 0.4342 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4586 | train 3.048611 | val 3.466599 | tok/s 2625056.63 | norm 0.4292 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4587 | train 3.003342 | val 3.466599 | tok/s 2628272.43 | norm 0.4505 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4588 | train 3.180885 | val 3.466599 | tok/s 2629322.84 | norm 0.4792 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4589 | train 3.105348 | val 3.466599 | tok/s 2632501.34 | norm 0.4701 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4590 | train 3.121845 | val 3.466599 | tok/s 2627714.84 | norm 0.4791 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4591 | train 3.016826 | val 3.466599 | tok/s 2629965.66 | norm 0.5059 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4592 | train 3.177046 | val 3.466599 | tok/s 2631679.36 | norm 0.5877 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4593 | train 3.013048 | val 3.466599 | tok/s 2623723.82 | norm 0.4915 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4594 | train 3.007257 | val 3.466599 | tok/s 2627039.73 | norm 0.6046 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4595 | train 3.117969 | val 3.466599 | tok/s 2623780.83 | norm 0.5246 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4596 | train 3.045998 | val 3.466599 | tok/s 2619652.81 | norm 0.4360 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4597 | train 3.085846 | val 3.466599 | tok/s 2629393.77 | norm 0.4360 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4598 | train 3.157361 | val 3.466599 | tok/s 2629208.75 | norm 0.4310 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4599 | train 2.982197 | val 3.466599 | tok/s 2626516.61 | norm 0.4215 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4600 | train 3.059904 | val 3.466599 | tok/s 2626289.31 | norm 0.4296 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4601 | train 3.232208 | val 3.466599 | tok/s 2624885.67 | norm 0.4226 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4602 | train 3.019863 | val 3.466599 | tok/s 2625592.66 | norm 0.4417 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4603 | train 3.100460 | val 3.466599 | tok/s 2621020.31 | norm 0.4623 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4604 | train 3.133867 | val 3.466599 | tok/s 2627768.97 | norm 0.4636 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4605 | train 3.037412 | val 3.466599 | tok/s 2499016.36 | norm 0.4329 | dt 209.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4606 | train 3.093880 | val 3.466599 | tok/s 2596727.30 | norm 0.4594 | dt 201.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4607 | train 2.996787 | val 3.466599 | tok/s 2457565.25 | norm 0.4676 | dt 213.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4608 | train 2.977490 | val 3.466599 | tok/s 2605445.47 | norm 0.4645 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4609 | train 2.972995 | val 3.466599 | tok/s 2616516.17 | norm 0.4602 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4610 | train 3.037009 | val 3.466599 | tok/s 2627214.06 | norm 0.4429 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4611 | train 3.017811 | val 3.466599 | tok/s 2632054.32 | norm 0.4440 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4612 | train 3.195488 | val 3.466599 | tok/s 2622391.48 | norm 0.4514 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4613 | train 3.132623 | val 3.466599 | tok/s 2629024.33 | norm 0.4496 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4614 | train 3.050547 | val 3.466599 | tok/s 2631805.84 | norm 0.4445 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4615 | train 3.081376 | val 3.466599 | tok/s 2630673.84 | norm 0.4963 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4616 | train 3.089362 | val 3.466599 | tok/s 2633176.74 | norm 0.4741 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4617 | train 3.122138 | val 3.466599 | tok/s 2624101.15 | norm 0.4481 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4618 | train 3.232255 | val 3.466599 | tok/s 2623858.73 | norm 0.4465 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4619 | train 3.065639 | val 3.466599 | tok/s 2633008.29 | norm 0.4877 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4620 | train 3.052475 | val 3.466599 | tok/s 2627505.21 | norm 0.4797 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4621 | train 3.092339 | val 3.466599 | tok/s 2633219.22 | norm 0.4640 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4622 | train 3.062019 | val 3.466599 | tok/s 2627971.26 | norm 0.4504 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4623 | train 3.000426 | val 3.466599 | tok/s 2628033.41 | norm 0.4480 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4624 | train 3.109559 | val 3.466599 | tok/s 2630616.52 | norm 0.4401 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4625 | train 3.197030 | val 3.466599 | tok/s 2479171.32 | norm 0.4798 | dt 211.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4626 | train 3.031089 | val 3.466599 | tok/s 2630520.57 | norm 0.4510 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4627 | train 3.023572 | val 3.466599 | tok/s 2629275.41 | norm 0.4281 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4628 | train 3.005382 | val 3.466599 | tok/s 2628689.93 | norm 0.4493 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4629 | train 3.119678 | val 3.466599 | tok/s 2632778.46 | norm 0.4610 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4630 | train 3.098726 | val 3.466599 | tok/s 2614819.67 | norm 0.4607 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4631 | train 3.094337 | val 3.466599 | tok/s 2623118.34 | norm 0.4569 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4632 | train 3.044705 | val 3.466599 | tok/s 2630575.08 | norm 0.4204 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4633 | train 3.040581 | val 3.466599 | tok/s 2628262.92 | norm 0.4261 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4634 | train 3.161832 | val 3.466599 | tok/s 2630163.81 | norm 0.4383 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4635 | train 3.037642 | val 3.466599 | tok/s 2625156.89 | norm 0.4288 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4636 | train 3.128874 | val 3.466599 | tok/s 2628034.42 | norm 0.4373 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4637 | train 3.098089 | val 3.466599 | tok/s 2625941.69 | norm 0.4818 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4638 | train 3.003218 | val 3.466599 | tok/s 2628589.18 | norm 0.4558 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4639 | train 3.140001 | val 3.466599 | tok/s 2631103.41 | norm 0.4583 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4640 | train 3.064199 | val 3.466599 | tok/s 2612977.99 | norm 0.4531 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4641 | train 3.094585 | val 3.466599 | tok/s 2629470.71 | norm 0.4556 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4642 | train 3.075161 | val 3.466599 | tok/s 2625631.99 | norm 0.4643 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4643 | train 3.060206 | val 3.466599 | tok/s 2627045.54 | norm 0.4758 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4644 | train 2.834131 | val 3.466599 | tok/s 2631694.04 | norm 0.4842 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4645 | train 3.069136 | val 3.466599 | tok/s 2628625.28 | norm 0.4325 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4646 | train 3.155718 | val 3.466599 | tok/s 2631977.74 | norm 0.4108 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4647 | train 3.084898 | val 3.466599 | tok/s 2628132.91 | norm 0.4574 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4648 | train 3.148397 | val 3.466599 | tok/s 2631106.95 | norm 0.4936 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4649 | train 2.985812 | val 3.466599 | tok/s 2620783.34 | norm 0.4771 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4650 | train 3.191940 | val 3.466599 | tok/s 2626378.35 | norm 0.5614 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4651 | train 3.100128 | val 3.466599 | tok/s 2629400.48 | norm 0.4554 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4652 | train 3.061328 | val 3.466599 | tok/s 2624168.32 | norm 0.4380 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4653 | train 3.167445 | val 3.466599 | tok/s 2630334.29 | norm 0.5363 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4654 | train 3.060639 | val 3.466599 | tok/s 2629778.15 | norm 0.4683 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4655 | train 3.126064 | val 3.466599 | tok/s 2626108.05 | norm 0.4200 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4656 | train 3.068045 | val 3.466599 | tok/s 2626199.24 | norm 0.4533 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4657 | train 3.050628 | val 3.466599 | tok/s 2627119.93 | norm 0.4629 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4658 | train 2.980532 | val 3.466599 | tok/s 2630985.11 | norm 0.4777 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4659 | train 3.138918 | val 3.466599 | tok/s 2628883.83 | norm 0.4867 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4660 | train 3.006015 | val 3.466599 | tok/s 2628794.03 | norm 0.4493 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4661 | train 3.114881 | val 3.466599 | tok/s 2625540.16 | norm 0.4636 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4662 | train 3.069740 | val 3.466599 | tok/s 2630532.54 | norm 0.4891 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4663 | train 3.135304 | val 3.466599 | tok/s 2629458.52 | norm 0.4734 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4664 | train 3.117298 | val 3.466599 | tok/s 2624543.35 | norm 0.4378 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4665 | train 3.135617 | val 3.466599 | tok/s 2626442.66 | norm 0.4534 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4666 | train 3.024887 | val 3.466599 | tok/s 2629060.53 | norm 0.4935 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4667 | train 2.959788 | val 3.466599 | tok/s 2627561.84 | norm 0.4217 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4668 | train 3.123745 | val 3.466599 | tok/s 2628352.28 | norm 0.4095 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4669 | train 3.085976 | val 3.466599 | tok/s 2627822.44 | norm 0.4210 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4670 | train 3.140105 | val 3.466599 | tok/s 2624725.12 | norm 0.4685 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4671 | train 3.134516 | val 3.466599 | tok/s 2631765.63 | norm 0.4590 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4672 | train 3.008436 | val 3.466599 | tok/s 2629626.05 | norm 0.4582 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4673 | train 3.020738 | val 3.466599 | tok/s 2630420.04 | norm 0.4798 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4674 | train 2.963135 | val 3.466599 | tok/s 2632230.82 | norm 0.4223 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4675 | train 3.259598 | val 3.466599 | tok/s 2629156.93 | norm 0.4538 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4676 | train 3.123996 | val 3.466599 | tok/s 2633460.60 | norm 0.4166 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4677 | train 3.147054 | val 3.466599 | tok/s 2624039.77 | norm 0.4333 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4678 | train 3.077457 | val 3.466599 | tok/s 2629204.80 | norm 0.4598 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4679 | train 3.064711 | val 3.466599 | tok/s 2624239.17 | norm 0.4583 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4680 | train 3.030420 | val 3.466599 | tok/s 2489978.51 | norm 0.4557 | dt 210.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4681 | train 3.219986 | val 3.466599 | tok/s 2632781.89 | norm 0.4681 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4682 | train 3.141316 | val 3.466599 | tok/s 2629773.87 | norm 0.5347 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4683 | train 2.969039 | val 3.466599 | tok/s 2631104.12 | norm 0.4914 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4684 | train 3.035328 | val 3.466599 | tok/s 2622739.28 | norm 0.4061 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4685 | train 3.154695 | val 3.466599 | tok/s 2629188.84 | norm 0.4440 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4686 | train 3.132489 | val 3.466599 | tok/s 2628772.87 | norm 0.5092 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4687 | train 3.034173 | val 3.466599 | tok/s 2626110.33 | norm 0.4526 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4688 | train 3.124747 | val 3.466599 | tok/s 2629841.65 | norm 0.4824 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4689 | train 3.111996 | val 3.466599 | tok/s 2624759.90 | norm 0.4852 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4690 | train 3.159899 | val 3.466599 | tok/s 2628800.32 | norm 0.4326 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4691 | train 3.139729 | val 3.466599 | tok/s 2631608.43 | norm 0.4941 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4692 | train 3.154322 | val 3.466599 | tok/s 2624761.43 | norm 0.4839 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4693 | train 3.064177 | val 3.466599 | tok/s 2629946.67 | norm 0.4521 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4694 | train 3.109001 | val 3.466599 | tok/s 2622302.58 | norm 0.4596 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4695 | train 2.988158 | val 3.466599 | tok/s 2628784.32 | norm 0.4886 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4696 | train 2.976760 | val 3.466599 | tok/s 2628966.56 | norm 0.4778 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4697 | train 3.136357 | val 3.466599 | tok/s 2627169.50 | norm 0.4876 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4698 | train 2.968585 | val 3.466599 | tok/s 2627208.52 | norm 0.4302 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4699 | train 3.084132 | val 3.466599 | tok/s 2627563.37 | norm 0.4153 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4700 | train 2.999546 | val 3.466599 | tok/s 2628696.49 | norm 0.4112 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4701 | train 3.181607 | val 3.466599 | tok/s 2612763.33 | norm 0.4183 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4702 | train 3.250011 | val 3.466599 | tok/s 2628362.88 | norm 0.5006 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4703 | train 2.995887 | val 3.466599 | tok/s 2630998.01 | norm 0.4496 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4704 | train 3.101606 | val 3.466599 | tok/s 2629240.06 | norm 0.4079 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4705 | train 3.033500 | val 3.466599 | tok/s 2628183.72 | norm 0.4715 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4706 | train 3.197908 | val 3.466599 | tok/s 2623319.49 | norm 0.5156 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4707 | train 3.060862 | val 3.466599 | tok/s 2626501.37 | norm 0.4547 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4708 | train 3.116900 | val 3.466599 | tok/s 2627013.31 | norm 0.4184 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4709 | train 3.056065 | val 3.466599 | tok/s 2630365.46 | norm 0.4312 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4710 | train 3.187576 | val 3.466599 | tok/s 2631688.36 | norm 0.4397 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4711 | train 3.077976 | val 3.466599 | tok/s 2626857.86 | norm 0.5080 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4712 | train 3.243931 | val 3.466599 | tok/s 2629879.80 | norm 0.5254 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4713 | train 3.001557 | val 3.466599 | tok/s 2631504.35 | norm 0.4969 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4714 | train 3.034469 | val 3.466599 | tok/s 2628128.47 | norm 0.4275 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4715 | train 3.067151 | val 3.466599 | tok/s 2629799.99 | norm 0.4170 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4716 | train 3.190579 | val 3.466599 | tok/s 2627260.95 | norm 0.4509 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4717 | train 3.082942 | val 3.466599 | tok/s 2624998.71 | norm 0.4093 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4718 | train 3.056355 | val 3.466599 | tok/s 2629022.94 | norm 0.4022 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4719 | train 3.066008 | val 3.466599 | tok/s 2633423.16 | norm 0.4447 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4720 | train 3.252052 | val 3.466599 | tok/s 2633245.11 | norm 0.4536 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4721 | train 3.103555 | val 3.466599 | tok/s 2628029.31 | norm 0.4421 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4722 | train 3.119145 | val 3.466599 | tok/s 2626948.82 | norm 0.4490 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4723 | train 3.024910 | val 3.466599 | tok/s 2629702.28 | norm 0.5137 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4724 | train 3.078127 | val 3.466599 | tok/s 2626480.14 | norm 0.4808 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4725 | train 3.221026 | val 3.466599 | tok/s 2630475.78 | norm 0.4065 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4726 | train 3.202217 | val 3.466599 | tok/s 2625365.84 | norm 0.4197 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4727 | train 3.105670 | val 3.466599 | tok/s 2625043.00 | norm 0.4570 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4728 | train 3.109113 | val 3.466599 | tok/s 2629696.77 | norm 0.4794 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4729 | train 3.036650 | val 3.466599 | tok/s 2629360.59 | norm 0.4528 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4730 | train 3.000312 | val 3.466599 | tok/s 2630714.64 | norm 0.4600 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4731 | train 3.020865 | val 3.466599 | tok/s 2630574.54 | norm 0.4592 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4732 | train 3.068318 | val 3.466599 | tok/s 2626689.22 | norm 0.4273 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4733 | train 3.112551 | val 3.466599 | tok/s 2629393.18 | norm 0.4717 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4734 | train 3.177212 | val 3.466599 | tok/s 2629417.92 | norm 0.5099 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4735 | train 3.060976 | val 3.466599 | tok/s 2630713.42 | norm 0.4470 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4736 | train 3.061111 | val 3.466599 | tok/s 2627159.33 | norm 0.4740 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4737 | train 2.979697 | val 3.466599 | tok/s 2624712.15 | norm 0.5366 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4738 | train 3.000402 | val 3.466599 | tok/s 2631466.29 | norm 0.4676 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4739 | train 3.256061 | val 3.466599 | tok/s 2627820.14 | norm 0.4626 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4740 | train 3.131641 | val 3.466599 | tok/s 2632174.01 | norm 0.4627 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4741 | train 3.073407 | val 3.466599 | tok/s 2629833.51 | norm 0.4157 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4742 | train 3.077105 | val 3.466599 | tok/s 2600254.28 | norm 0.4173 | dt 201.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4743 | train 3.026599 | val 3.466599 | tok/s 2628565.50 | norm 0.4553 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4744 | train 3.024114 | val 3.466599 | tok/s 2632810.53 | norm 0.4979 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4745 | train 3.060148 | val 3.466599 | tok/s 2630601.17 | norm 0.5157 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4746 | train 3.016830 | val 3.466599 | tok/s 2560543.49 | norm 0.4690 | dt 204.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4747 | train 3.028409 | val 3.466599 | tok/s 2632659.71 | norm 0.4562 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4748 | train 3.178285 | val 3.466599 | tok/s 2632152.43 | norm 0.4639 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4749 | train 3.164541 | val 3.466599 | tok/s 2627463.64 | norm 0.4331 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 4750 | val_loss 3.452990 step 4750 | train 3.092401 | val 3.452990 | tok/s 2612476.16 | norm 0.6097 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4751 | train 3.183424 | val 3.452990 | tok/s 2631586.29 | norm 0.4829 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4752 | train 3.085906 | val 3.452990 | tok/s 2623616.07 | norm 0.4508 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4753 | train 3.111971 | val 3.452990 | tok/s 2509672.43 | norm 0.4835 | dt 208.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4754 | train 3.043179 | val 3.452990 | tok/s 2631989.28 | norm 0.4266 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4755 | train 3.015542 | val 3.452990 | tok/s 2630811.75 | norm 0.4061 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4756 | train 3.029982 | val 3.452990 | tok/s 2630883.36 | norm 0.4069 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4757 | train 2.943087 | val 3.452990 | tok/s 2631902.31 | norm 0.4092 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4758 | train 3.117637 | val 3.452990 | tok/s 2631629.66 | norm 0.4209 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4759 | train 3.100709 | val 3.452990 | tok/s 2635391.26 | norm 0.4101 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4760 | train 3.070694 | val 3.452990 | tok/s 2615702.74 | norm 0.3965 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4761 | train 3.017114 | val 3.452990 | tok/s 2631431.22 | norm 0.4313 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4762 | train 3.044439 | val 3.452990 | tok/s 2636668.74 | norm 0.4437 | dt 198.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4763 | train 3.095699 | val 3.452990 | tok/s 2630217.06 | norm 0.4804 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4764 | train 3.170367 | val 3.452990 | tok/s 2629827.02 | norm 0.4673 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4765 | train 3.036676 | val 3.452990 | tok/s 2622981.63 | norm 0.4689 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4766 | train 3.150483 | val 3.452990 | tok/s 2630854.29 | norm 0.4773 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4767 | train 3.137767 | val 3.452990 | tok/s 2625405.69 | norm 0.4622 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4768 | train 3.098043 | val 3.452990 | tok/s 2628732.45 | norm 0.4805 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4769 | train 3.323680 | val 3.452990 | tok/s 2629410.19 | norm 0.4726 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4770 | train 3.183076 | val 3.452990 | tok/s 2629767.96 | norm 0.4787 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4771 | train 3.036687 | val 3.452990 | tok/s 2627755.71 | norm 0.4309 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4772 | train 3.094033 | val 3.452990 | tok/s 2628296.90 | norm 0.4330 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4773 | train 3.285089 | val 3.452990 | tok/s 2631285.43 | norm 0.4680 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4774 | train 3.029079 | val 3.452990 | tok/s 2622504.28 | norm 0.4354 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4775 | train 3.016996 | val 3.452990 | tok/s 2626803.22 | norm 0.4226 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4776 | train 3.037275 | val 3.452990 | tok/s 2631880.18 | norm 0.4900 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4777 | train 3.194106 | val 3.452990 | tok/s 2626063.38 | norm 0.4428 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4778 | train 3.084411 | val 3.452990 | tok/s 2632894.58 | norm 0.4726 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4779 | train 3.139134 | val 3.452990 | tok/s 2625339.39 | norm 0.4905 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4780 | train 3.092031 | val 3.452990 | tok/s 2626373.56 | norm 0.4822 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4781 | train 3.054382 | val 3.452990 | tok/s 2627152.23 | norm 0.4787 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4782 | train 3.010782 | val 3.452990 | tok/s 2628570.43 | norm 0.5116 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4783 | train 3.042123 | val 3.452990 | tok/s 2634347.10 | norm 0.4985 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4784 | train 2.962401 | val 3.452990 | tok/s 2622637.44 | norm 0.4295 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4785 | train 3.062191 | val 3.452990 | tok/s 2628960.75 | norm 0.4605 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4786 | train 2.987523 | val 3.452990 | tok/s 2632108.54 | norm 0.4844 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4787 | train 3.075975 | val 3.452990 | tok/s 2628203.08 | norm 0.4799 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4788 | train 2.825278 | val 3.452990 | tok/s 2629365.68 | norm 0.5445 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4789 | train 3.038285 | val 3.452990 | tok/s 2620115.40 | norm 0.4959 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4790 | train 3.028208 | val 3.452990 | tok/s 2625501.77 | norm 0.4089 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4791 | train 3.108647 | val 3.452990 | tok/s 2628815.95 | norm 0.5074 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4792 | train 3.093172 | val 3.452990 | tok/s 2628957.52 | norm 0.4723 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4793 | train 3.022816 | val 3.452990 | tok/s 2631269.50 | norm 0.4208 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4794 | train 3.192610 | val 3.452990 | tok/s 2626376.62 | norm 0.4356 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4795 | train 3.048525 | val 3.452990 | tok/s 2628143.70 | norm 0.4281 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4796 | train 3.121900 | val 3.452990 | tok/s 2631682.56 | norm 0.4763 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4797 | train 3.171197 | val 3.452990 | tok/s 2627765.58 | norm 0.4788 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4798 | train 3.082147 | val 3.452990 | tok/s 2631059.79 | norm 0.4508 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4799 | train 3.091874 | val 3.452990 | tok/s 2621575.64 | norm 0.4896 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4800 | train 3.058557 | val 3.452990 | tok/s 2627964.96 | norm 0.5093 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4801 | train 3.051013 | val 3.452990 | tok/s 2632164.08 | norm 0.4722 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4802 | train 3.128946 | val 3.452990 | tok/s 2629565.57 | norm 0.4794 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4803 | train 3.073617 | val 3.452990 | tok/s 2633345.59 | norm 0.6141 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4804 | train 3.083198 | val 3.452990 | tok/s 2629826.42 | norm 0.4264 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4805 | train 3.006275 | val 3.452990 | tok/s 2624307.15 | norm 0.4237 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4806 | train 3.199044 | val 3.452990 | tok/s 2634313.03 | norm 0.4667 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4807 | train 3.095992 | val 3.452990 | tok/s 2479801.72 | norm 0.4215 | dt 211.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4808 | train 3.077053 | val 3.452990 | tok/s 2537477.36 | norm 0.4451 | dt 206.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4809 | train 2.989018 | val 3.452990 | tok/s 2629809.28 | norm 0.4824 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4810 | train 2.960337 | val 3.452990 | tok/s 2629155.64 | norm 0.4550 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4811 | train 3.104065 | val 3.452990 | tok/s 2629685.67 | norm 0.4465 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4812 | train 3.311924 | val 3.452990 | tok/s 2625652.74 | norm 0.4735 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4813 | train 3.137833 | val 3.452990 | tok/s 2630090.14 | norm 0.4246 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4814 | train 3.083567 | val 3.452990 | tok/s 2625001.47 | norm 0.4051 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4815 | train 3.070480 | val 3.452990 | tok/s 2629452.76 | norm 0.4660 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4816 | train 2.969843 | val 3.452990 | tok/s 2632241.87 | norm 0.4589 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4817 | train 3.077503 | val 3.452990 | tok/s 2611476.21 | norm 0.4114 | dt 200.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4818 | train 3.184345 | val 3.452990 | tok/s 2629594.95 | norm 0.4491 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4819 | train 3.104557 | val 3.452990 | tok/s 2626322.49 | norm 0.4833 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4820 | train 3.211671 | val 3.452990 | tok/s 2630040.81 | norm 0.4892 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4821 | train 3.038973 | val 3.452990 | tok/s 2623752.81 | norm 0.4785 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4822 | train 2.992277 | val 3.452990 | tok/s 2620103.71 | norm 0.4657 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4823 | train 3.013999 | val 3.452990 | tok/s 2625052.24 | norm 0.4795 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4824 | train 3.134135 | val 3.452990 | tok/s 2632053.65 | norm 0.4849 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4825 | train 3.288137 | val 3.452990 | tok/s 2633218.78 | norm 0.4677 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4826 | train 2.991451 | val 3.452990 | tok/s 2627058.98 | norm 0.4058 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4827 | train 3.122369 | val 3.452990 | tok/s 2627855.33 | norm 0.4297 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4828 | train 3.110265 | val 3.452990 | tok/s 2628288.97 | norm 0.4551 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4829 | train 3.097107 | val 3.452990 | tok/s 2626975.37 | norm 0.4632 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4830 | train 2.999837 | val 3.452990 | tok/s 2629365.24 | norm 0.4267 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4831 | train 3.237073 | val 3.452990 | tok/s 2631730.59 | norm 0.4464 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4832 | train 3.078709 | val 3.452990 | tok/s 2629285.95 | norm 0.4883 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4833 | train 3.122501 | val 3.452990 | tok/s 2629913.86 | norm 0.4807 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4834 | train 3.011599 | val 3.452990 | tok/s 2625598.13 | norm 0.4433 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4835 | train 3.059663 | val 3.452990 | tok/s 2627571.12 | norm 0.4560 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4836 | train 3.043867 | val 3.452990 | tok/s 2618808.28 | norm 0.4626 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4837 | train 3.207331 | val 3.452990 | tok/s 2625081.21 | norm 0.4857 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4838 | train 2.957426 | val 3.452990 | tok/s 2627384.81 | norm 0.4387 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4839 | train 3.226137 | val 3.452990 | tok/s 2628202.52 | norm 0.4779 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4840 | train 3.129264 | val 3.452990 | tok/s 2629837.29 | norm 0.5167 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4841 | train 3.141253 | val 3.452990 | tok/s 2624204.19 | norm 0.5157 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4842 | train 3.039373 | val 3.452990 | tok/s 2627182.51 | norm 0.4803 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4843 | train 3.140297 | val 3.452990 | tok/s 2630965.33 | norm 0.4062 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4844 | train 3.055169 | val 3.452990 | tok/s 2630003.28 | norm 0.4079 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4845 | train 3.169662 | val 3.452990 | tok/s 2631712.75 | norm 0.4558 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4846 | train 3.170965 | val 3.452990 | tok/s 2624958.45 | norm 0.4548 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4847 | train 3.098409 | val 3.452990 | tok/s 2625191.15 | norm 0.4388 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4848 | train 3.134738 | val 3.452990 | tok/s 2628068.54 | norm 0.4606 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4849 | train 3.115460 | val 3.452990 | tok/s 2621693.75 | norm 0.4650 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4850 | train 3.102582 | val 3.452990 | tok/s 2625277.78 | norm 0.4549 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4851 | train 3.092506 | val 3.452990 | tok/s 2628308.80 | norm 0.3948 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4852 | train 3.224643 | val 3.452990 | tok/s 2538781.00 | norm 0.4583 | dt 206.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4853 | train 3.064781 | val 3.452990 | tok/s 2630283.27 | norm 0.3939 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4854 | train 3.234906 | val 3.452990 | tok/s 2631581.46 | norm 0.4932 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4855 | train 3.356699 | val 3.452990 | tok/s 2630197.91 | norm 0.6220 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4856 | train 3.233416 | val 3.452990 | tok/s 2626049.40 | norm 0.4947 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4857 | train 3.190578 | val 3.452990 | tok/s 2629241.92 | norm 0.4532 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4858 | train 3.132125 | val 3.452990 | tok/s 2629705.31 | norm 0.5110 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4859 | train 3.131488 | val 3.452990 | tok/s 2627926.13 | norm 0.4941 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4860 | train 3.031608 | val 3.452990 | tok/s 2626028.59 | norm 0.4885 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4861 | train 3.048629 | val 3.452990 | tok/s 2626036.48 | norm 0.4405 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4862 | train 3.075024 | val 3.452990 | tok/s 2627272.41 | norm 0.4801 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4863 | train 3.006103 | val 3.452990 | tok/s 2631205.84 | norm 0.4787 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4864 | train 2.999284 | val 3.452990 | tok/s 2628613.20 | norm 0.4902 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4865 | train 3.040915 | val 3.452990 | tok/s 2629290.59 | norm 0.5079 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4866 | train 3.188468 | val 3.452990 | tok/s 2629722.73 | norm 0.4655 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4867 | train 3.143283 | val 3.452990 | tok/s 2621478.39 | norm 0.4343 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4868 | train 2.921851 | val 3.452990 | tok/s 2631824.93 | norm 0.4142 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4869 | train 3.145423 | val 3.452990 | tok/s 2627181.43 | norm 0.4415 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4870 | train 2.999231 | val 3.452990 | tok/s 2630606.82 | norm 0.4225 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4871 | train 3.063709 | val 3.452990 | tok/s 2625287.38 | norm 0.4196 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4872 | train 3.162271 | val 3.452990 | tok/s 2630623.36 | norm 0.4256 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4873 | train 3.010573 | val 3.452990 | tok/s 2630393.31 | norm 0.4101 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4874 | train 3.068178 | val 3.452990 | tok/s 2628372.45 | norm 0.4421 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4875 | train 3.112835 | val 3.452990 | tok/s 2630718.57 | norm 0.4557 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4876 | train 3.054021 | val 3.452990 | tok/s 2631921.69 | norm 0.4523 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4877 | train 3.108159 | val 3.452990 | tok/s 2631372.33 | norm 0.4720 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4878 | train 3.187804 | val 3.452990 | tok/s 2624448.80 | norm 0.4738 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4879 | train 3.048078 | val 3.452990 | tok/s 2622855.32 | norm 0.4352 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4880 | train 3.097634 | val 3.452990 | tok/s 2620406.42 | norm 0.4651 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4881 | train 3.040345 | val 3.452990 | tok/s 2624364.95 | norm 0.4485 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4882 | train 3.195996 | val 3.452990 | tok/s 2627110.83 | norm 0.4578 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4883 | train 3.076518 | val 3.452990 | tok/s 2626245.81 | norm 0.4749 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4884 | train 3.209143 | val 3.452990 | tok/s 2622270.22 | norm 0.4926 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4885 | train 2.967742 | val 3.452990 | tok/s 2626915.92 | norm 0.4766 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4886 | train 3.001257 | val 3.452990 | tok/s 2629592.66 | norm 0.4696 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4887 | train 3.227345 | val 3.452990 | tok/s 2629182.86 | norm 0.4714 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4888 | train 3.069689 | val 3.452990 | tok/s 2626085.69 | norm 0.4456 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4889 | train 3.017858 | val 3.452990 | tok/s 2628002.66 | norm 0.4632 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4890 | train 3.053521 | val 3.452990 | tok/s 2630613.90 | norm 0.4472 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4891 | train 3.008216 | val 3.452990 | tok/s 2625489.80 | norm 0.4203 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4892 | train 3.116623 | val 3.452990 | tok/s 2627963.43 | norm 0.4402 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4893 | train 3.105485 | val 3.452990 | tok/s 2621742.55 | norm 0.4256 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4894 | train 2.983123 | val 3.452990 | tok/s 2626847.60 | norm 0.4162 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4895 | train 3.199066 | val 3.452990 | tok/s 2631729.92 | norm 0.4623 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4896 | train 3.028941 | val 3.452990 | tok/s 2631872.86 | norm 0.4780 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4897 | train 3.030590 | val 3.452990 | tok/s 2630563.97 | norm 0.4629 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4898 | train 2.996482 | val 3.452990 | tok/s 2623777.84 | norm 0.4476 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4899 | train 2.950770 | val 3.452990 | tok/s 2624391.53 | norm 0.4216 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4900 | train 3.013188 | val 3.452990 | tok/s 2628991.08 | norm 0.4193 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4901 | train 3.025051 | val 3.452990 | tok/s 2626272.68 | norm 0.4755 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4902 | train 2.956754 | val 3.452990 | tok/s 2629103.38 | norm 0.4721 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4903 | train 3.124734 | val 3.452990 | tok/s 2625948.06 | norm 0.4685 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4904 | train 2.982775 | val 3.452990 | tok/s 2628921.33 | norm 0.4699 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4905 | train 3.072531 | val 3.452990 | tok/s 2628485.01 | norm 0.4890 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4906 | train 3.158338 | val 3.452990 | tok/s 2626157.87 | norm 0.4885 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4907 | train 3.177229 | val 3.452990 | tok/s 2630970.49 | norm 0.4353 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4908 | train 3.283533 | val 3.452990 | tok/s 2626920.64 | norm 0.4014 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4909 | train 2.988215 | val 3.452990 | tok/s 2626485.56 | norm 0.4184 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4910 | train 3.081312 | val 3.452990 | tok/s 2632837.14 | norm 0.4748 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4911 | train 3.158898 | val 3.452990 | tok/s 2627308.43 | norm 0.4592 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4912 | train 3.199024 | val 3.452990 | tok/s 2631441.20 | norm 0.4326 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4913 | train 3.060572 | val 3.452990 | tok/s 2623140.33 | norm 0.4077 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4914 | train 3.185835 | val 3.452990 | tok/s 2628415.10 | norm 0.4419 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4915 | train 3.057918 | val 3.452990 | tok/s 2629035.05 | norm 0.4135 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4916 | train 3.083251 | val 3.452990 | tok/s 2628913.99 | norm 0.4465 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4917 | train 3.131760 | val 3.452990 | tok/s 2630324.99 | norm 0.4795 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4918 | train 3.116247 | val 3.452990 | tok/s 2633148.64 | norm 0.4920 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4919 | train 2.999968 | val 3.452990 | tok/s 2462322.26 | norm 0.4716 | dt 212.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4920 | train 3.267307 | val 3.452990 | tok/s 2628019.23 | norm 0.4631 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4921 | train 3.189327 | val 3.452990 | tok/s 2627283.12 | norm 0.4584 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4922 | train 3.098134 | val 3.452990 | tok/s 2624725.30 | norm 0.4202 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4923 | train 3.026762 | val 3.452990 | tok/s 2629464.77 | norm 0.5152 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4924 | train 3.005477 | val 3.452990 | tok/s 2632196.71 | norm 0.4966 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4925 | train 3.098286 | val 3.452990 | tok/s 2628566.03 | norm 0.4652 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4926 | train 3.095731 | val 3.452990 | tok/s 2626023.46 | norm 0.4816 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4927 | train 3.062205 | val 3.452990 | tok/s 2632662.51 | norm 0.5071 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4928 | train 2.977194 | val 3.452990 | tok/s 2625243.35 | norm 0.4785 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4929 | train 2.970408 | val 3.452990 | tok/s 2630039.91 | norm 0.4435 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4930 | train 3.084547 | val 3.452990 | tok/s 2627224.07 | norm 0.4495 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4931 | train 3.022323 | val 3.452990 | tok/s 2621697.75 | norm 0.4507 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4932 | train 3.000404 | val 3.452990 | tok/s 2627642.19 | norm 0.4370 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4933 | train 3.174101 | val 3.452990 | tok/s 2626492.70 | norm 0.4718 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4934 | train 3.108925 | val 3.452990 | tok/s 2613165.03 | norm 0.4851 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4935 | train 3.118728 | val 3.452990 | tok/s 2630408.56 | norm 0.4384 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4936 | train 3.280577 | val 3.452990 | tok/s 2629004.01 | norm 0.5964 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4937 | train 3.000861 | val 3.452990 | tok/s 2633549.08 | norm 0.4455 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4938 | train 2.961752 | val 3.452990 | tok/s 2628037.82 | norm 0.4353 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4939 | train 3.166424 | val 3.452990 | tok/s 2631964.32 | norm 0.4851 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4940 | train 3.079327 | val 3.452990 | tok/s 2627940.96 | norm 0.4526 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4941 | train 3.111937 | val 3.452990 | tok/s 2625471.87 | norm 0.4461 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4942 | train 3.254189 | val 3.452990 | tok/s 2628446.29 | norm 0.4574 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4943 | train 3.105134 | val 3.452990 | tok/s 2625587.48 | norm 0.4158 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4944 | train 3.131743 | val 3.452990 | tok/s 2628386.68 | norm 0.4537 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4945 | train 3.117321 | val 3.452990 | tok/s 2620175.91 | norm 0.4659 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4946 | train 3.090423 | val 3.452990 | tok/s 2623413.90 | norm 0.4383 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4947 | train 3.167797 | val 3.452990 | tok/s 2628359.49 | norm 0.4403 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4948 | train 2.932906 | val 3.452990 | tok/s 2630162.79 | norm 0.4415 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4949 | train 3.093253 | val 3.452990 | tok/s 2628117.24 | norm 0.4143 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4950 | train 3.023116 | val 3.452990 | tok/s 2628441.92 | norm 0.4117 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4951 | train 3.092795 | val 3.452990 | tok/s 2631809.28 | norm 0.4454 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4952 | train 3.067234 | val 3.452990 | tok/s 2633616.88 | norm 0.4899 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4953 | train 3.084227 | val 3.452990 | tok/s 2629482.31 | norm 0.4834 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4954 | train 3.147362 | val 3.452990 | tok/s 2630037.34 | norm 0.4541 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4955 | train 3.008614 | val 3.452990 | tok/s 2622092.91 | norm 0.4594 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4956 | train 2.987806 | val 3.452990 | tok/s 2625867.86 | norm 0.4747 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4957 | train 3.016623 | val 3.452990 | tok/s 2631942.33 | norm 0.4777 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4958 | train 3.172997 | val 3.452990 | tok/s 2628802.43 | norm 0.5087 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4959 | train 2.967489 | val 3.452990 | tok/s 2627401.68 | norm 0.5254 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4960 | train 3.089748 | val 3.452990 | tok/s 2634326.16 | norm 0.4699 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4961 | train 3.119411 | val 3.452990 | tok/s 2629622.29 | norm 0.4546 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4962 | train 3.153754 | val 3.452990 | tok/s 2627729.29 | norm 0.3983 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4963 | train 3.073406 | val 3.452990 | tok/s 2623849.97 | norm 0.4240 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4964 | train 3.098512 | val 3.452990 | tok/s 2628391.36 | norm 0.4425 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4965 | train 3.165604 | val 3.452990 | tok/s 2624521.28 | norm 0.4521 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4966 | train 3.045144 | val 3.452990 | tok/s 2627847.95 | norm 0.4240 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4967 | train 3.104858 | val 3.452990 | tok/s 2626079.60 | norm 0.4247 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4968 | train 2.991573 | val 3.452990 | tok/s 2629262.62 | norm 0.4402 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4969 | train 3.234995 | val 3.452990 | tok/s 2632391.18 | norm 0.4731 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4970 | train 3.058258 | val 3.452990 | tok/s 2627720.37 | norm 0.4248 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4971 | train 3.182856 | val 3.452990 | tok/s 2627924.27 | norm 0.4368 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4972 | train 3.123075 | val 3.452990 | tok/s 2629014.97 | norm 0.4539 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4973 | train 3.162556 | val 3.452990 | tok/s 2631557.91 | norm 0.4289 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4974 | train 3.077637 | val 3.452990 | tok/s 2628729.05 | norm 0.4776 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4975 | train 3.043504 | val 3.452990 | tok/s 2623625.36 | norm 0.4630 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4976 | train 3.010370 | val 3.452990 | tok/s 2622808.54 | norm 0.4329 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4977 | train 3.038199 | val 3.452990 | tok/s 2629444.95 | norm 0.4581 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4978 | train 3.231976 | val 3.452990 | tok/s 2631519.67 | norm 0.4646 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4979 | train 3.068718 | val 3.452990 | tok/s 2631437.81 | norm 0.4093 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4980 | train 3.095545 | val 3.452990 | tok/s 2619789.41 | norm 0.4121 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4981 | train 3.074472 | val 3.452990 | tok/s 2628141.74 | norm 0.4380 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4982 | train 3.051217 | val 3.452990 | tok/s 2632214.72 | norm 0.4271 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4983 | train 3.019206 | val 3.452990 | tok/s 2626578.61 | norm 0.4445 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4984 | train 3.019972 | val 3.452990 | tok/s 2629791.90 | norm 0.4708 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4985 | train 3.141662 | val 3.452990 | tok/s 2631603.42 | norm 0.4347 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4986 | train 3.084030 | val 3.452990 | tok/s 2630426.60 | norm 0.4198 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4987 | train 3.073423 | val 3.452990 | tok/s 2483068.51 | norm 0.4177 | dt 211.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4988 | train 3.159857 | val 3.452990 | tok/s 2626408.93 | norm 0.4519 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4989 | train 3.187907 | val 3.452990 | tok/s 2567016.98 | norm 0.4691 | dt 204.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4990 | train 3.047945 | val 3.452990 | tok/s 2627675.65 | norm 0.4693 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4991 | train 3.263263 | val 3.452990 | tok/s 2625531.77 | norm 0.4605 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4992 | train 3.011715 | val 3.452990 | tok/s 2624358.00 | norm 0.4726 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4993 | train 3.077350 | val 3.452990 | tok/s 2624960.56 | norm 0.4653 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4994 | train 3.060068 | val 3.452990 | tok/s 2630046.39 | norm 0.4776 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4995 | train 3.097256 | val 3.452990 | tok/s 2626623.73 | norm 0.4656 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4996 | train 3.077894 | val 3.452990 | tok/s 2630095.86 | norm 0.4414 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4997 | train 3.076046 | val 3.452990 | tok/s 2627926.50 | norm 0.4387 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4998 | train 2.937428 | val 3.452990 | tok/s 2627080.45 | norm 0.4701 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 4999 | train 3.097426 | val 3.452990 | tok/s 2627955.17 | norm 0.4571 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 5000 | val_loss 3.442658 step 5000 | train 3.208389 | val 3.442658 | tok/s 2611201.69 | norm 0.4456 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5001 | train 3.015983 | val 3.442658 | tok/s 2631393.97 | norm 0.4414 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5002 | train 3.147620 | val 3.442658 | tok/s 2631513.51 | norm 0.4720 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5003 | train 3.118642 | val 3.442658 | tok/s 2627574.18 | norm 0.4602 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5004 | train 3.113834 | val 3.442658 | tok/s 2625537.59 | norm 0.3970 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5005 | train 3.140625 | val 3.442658 | tok/s 2628327.19 | norm 0.4283 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5006 | train 3.111424 | val 3.442658 | tok/s 2627559.44 | norm 0.4505 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5007 | train 3.087728 | val 3.442658 | tok/s 2631506.05 | norm 0.4318 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5008 | train 3.128545 | val 3.442658 | tok/s 2625447.88 | norm 0.4673 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5009 | train 3.043895 | val 3.442658 | tok/s 2628559.36 | norm 0.4729 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5010 | train 3.129422 | val 3.442658 | tok/s 2631708.84 | norm 0.4952 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5011 | train 3.166400 | val 3.442658 | tok/s 2629700.39 | norm 0.4720 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5012 | train 3.157314 | val 3.442658 | tok/s 2628151.75 | norm 0.4949 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5013 | train 2.964838 | val 3.442658 | tok/s 2628875.48 | norm 0.4535 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5014 | train 3.088958 | val 3.442658 | tok/s 2626416.13 | norm 0.4052 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5015 | train 3.038635 | val 3.442658 | tok/s 2625914.99 | norm 0.4261 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5016 | train 3.096902 | val 3.442658 | tok/s 2629148.41 | norm 0.4476 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5017 | train 3.162827 | val 3.442658 | tok/s 2632947.47 | norm 0.4371 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5018 | train 3.062712 | val 3.442658 | tok/s 2628935.50 | norm 0.4293 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5019 | train 3.137944 | val 3.442658 | tok/s 2627497.45 | norm 0.4492 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5020 | train 3.052662 | val 3.442658 | tok/s 2632635.76 | norm 0.5037 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5021 | train 3.124213 | val 3.442658 | tok/s 2621996.73 | norm 0.4829 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5022 | train 3.042248 | val 3.442658 | tok/s 2630383.55 | norm 0.4500 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5023 | train 3.140214 | val 3.442658 | tok/s 2625152.57 | norm 0.4747 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5024 | train 3.055799 | val 3.442658 | tok/s 2626363.45 | norm 0.4505 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5025 | train 3.029603 | val 3.442658 | tok/s 2632046.91 | norm 0.4450 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5026 | train 3.053673 | val 3.442658 | tok/s 2623849.59 | norm 0.5138 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5027 | train 3.037541 | val 3.442658 | tok/s 2629867.09 | norm 0.4843 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5028 | train 2.941740 | val 3.442658 | tok/s 2625567.83 | norm 0.4367 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5029 | train 2.964375 | val 3.442658 | tok/s 2627718.95 | norm 0.4400 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5030 | train 3.275699 | val 3.442658 | tok/s 2626120.01 | norm 0.4611 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5031 | train 3.088491 | val 3.442658 | tok/s 2630195.57 | norm 0.4449 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5032 | train 2.947106 | val 3.442658 | tok/s 2626161.63 | norm 0.4129 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5033 | train 3.132824 | val 3.442658 | tok/s 2626474.04 | norm 0.4311 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5034 | train 3.092376 | val 3.442658 | tok/s 2628086.79 | norm 0.4343 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5035 | train 3.118577 | val 3.442658 | tok/s 2588933.66 | norm 0.4253 | dt 202.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5036 | train 3.027440 | val 3.442658 | tok/s 2572618.81 | norm 0.4301 | dt 203.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5037 | train 3.062472 | val 3.442658 | tok/s 2632872.06 | norm 0.4676 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5038 | train 3.063858 | val 3.442658 | tok/s 2631904.58 | norm 0.4673 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5039 | train 3.307781 | val 3.442658 | tok/s 2627328.67 | norm 1.1032 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5040 | train 3.075320 | val 3.442658 | tok/s 2627578.93 | norm 0.4712 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5041 | train 2.958083 | val 3.442658 | tok/s 2629607.63 | norm 0.4209 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5042 | train 3.140748 | val 3.442658 | tok/s 2628372.66 | norm 0.4759 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5043 | train 2.974251 | val 3.442658 | tok/s 2627786.02 | norm 0.4962 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5044 | train 3.044295 | val 3.442658 | tok/s 2625751.08 | norm 0.3822 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5045 | train 3.082109 | val 3.442658 | tok/s 2627020.74 | norm 0.4262 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5046 | train 3.005670 | val 3.442658 | tok/s 2622620.81 | norm 0.4350 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5047 | train 3.029742 | val 3.442658 | tok/s 2629333.97 | norm 0.4247 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5048 | train 3.095136 | val 3.442658 | tok/s 2628684.17 | norm 0.4345 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5049 | train 3.115283 | val 3.442658 | tok/s 2625914.18 | norm 0.4407 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5050 | train 3.073902 | val 3.442658 | tok/s 2628342.43 | norm 0.4326 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5051 | train 3.071605 | val 3.442658 | tok/s 2630511.48 | norm 0.4431 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5052 | train 2.987120 | val 3.442658 | tok/s 2625776.33 | norm 0.4674 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5053 | train 3.092070 | val 3.442658 | tok/s 2629627.28 | norm 0.4564 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5054 | train 3.012307 | val 3.442658 | tok/s 2628308.81 | norm 0.4379 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5055 | train 3.052776 | val 3.442658 | tok/s 2625057.85 | norm 0.4592 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5056 | train 2.941999 | val 3.442658 | tok/s 2607495.08 | norm 0.4452 | dt 201.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5057 | train 3.095571 | val 3.442658 | tok/s 2632118.36 | norm 0.4710 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5058 | train 3.041314 | val 3.442658 | tok/s 2631825.14 | norm 0.4013 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5059 | train 3.071815 | val 3.442658 | tok/s 2630693.38 | norm 0.4477 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5060 | train 3.069353 | val 3.442658 | tok/s 2631268.64 | norm 0.4599 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5061 | train 3.161882 | val 3.442658 | tok/s 2628348.39 | norm 0.3730 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5062 | train 2.972615 | val 3.442658 | tok/s 2632217.75 | norm 0.3971 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5063 | train 3.043920 | val 3.442658 | tok/s 2629339.22 | norm 0.4657 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5064 | train 3.108077 | val 3.442658 | tok/s 2631180.67 | norm 0.4487 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5065 | train 3.110298 | val 3.442658 | tok/s 2628447.22 | norm 0.5093 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5066 | train 3.062375 | val 3.442658 | tok/s 2627986.58 | norm 0.5262 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5067 | train 3.088451 | val 3.442658 | tok/s 2630841.97 | norm 0.4952 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5068 | train 3.080948 | val 3.442658 | tok/s 2626940.29 | norm 0.5012 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5069 | train 3.146190 | val 3.442658 | tok/s 2627517.34 | norm 0.5030 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5070 | train 3.079668 | val 3.442658 | tok/s 2624654.05 | norm 0.4469 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5071 | train 3.134352 | val 3.442658 | tok/s 2626962.39 | norm 0.4835 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5072 | train 3.131493 | val 3.442658 | tok/s 2629774.06 | norm 0.4710 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5073 | train 3.059496 | val 3.442658 | tok/s 2628671.95 | norm 0.4702 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5074 | train 3.100940 | val 3.442658 | tok/s 2629445.08 | norm 0.4735 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5075 | train 3.120214 | val 3.442658 | tok/s 2626206.71 | norm 0.5192 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5076 | train 3.094347 | val 3.442658 | tok/s 2629731.28 | norm 0.4876 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5077 | train 2.988046 | val 3.442658 | tok/s 2629540.25 | norm 0.4265 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5078 | train 3.207659 | val 3.442658 | tok/s 2626463.19 | norm 0.5392 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5079 | train 2.983357 | val 3.442658 | tok/s 2628199.37 | norm 0.4784 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5080 | train 2.962005 | val 3.442658 | tok/s 2627223.62 | norm 0.5287 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5081 | train 3.012837 | val 3.442658 | tok/s 2626725.18 | norm 0.4728 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5082 | train 3.056840 | val 3.442658 | tok/s 2618144.66 | norm 0.4001 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5083 | train 3.017535 | val 3.442658 | tok/s 2621223.33 | norm 0.4241 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5084 | train 3.079243 | val 3.442658 | tok/s 2610304.40 | norm 0.4771 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5085 | train 3.070402 | val 3.442658 | tok/s 2617828.17 | norm 0.4676 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5086 | train 3.065109 | val 3.442658 | tok/s 2624422.53 | norm 0.4661 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5087 | train 3.194566 | val 3.442658 | tok/s 2627257.47 | norm 0.4623 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5088 | train 3.123168 | val 3.442658 | tok/s 2631360.36 | norm 0.4285 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5089 | train 3.091264 | val 3.442658 | tok/s 2629510.43 | norm 0.4636 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5090 | train 3.097352 | val 3.442658 | tok/s 2570894.25 | norm 0.4388 | dt 203.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5091 | train 3.225646 | val 3.442658 | tok/s 2625339.74 | norm 0.4785 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5092 | train 3.086277 | val 3.442658 | tok/s 2631070.78 | norm 0.5357 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5093 | train 3.014033 | val 3.442658 | tok/s 2625556.53 | norm 0.3992 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5094 | train 3.085436 | val 3.442658 | tok/s 2627574.22 | norm 0.4532 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5095 | train 3.095285 | val 3.442658 | tok/s 2628310.05 | norm 0.4657 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5096 | train 3.027121 | val 3.442658 | tok/s 2500889.16 | norm 0.4191 | dt 209.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5097 | train 3.071350 | val 3.442658 | tok/s 2623276.31 | norm 0.3946 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5098 | train 3.121132 | val 3.442658 | tok/s 2630300.48 | norm 0.4242 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5099 | train 3.042346 | val 3.442658 | tok/s 2621797.25 | norm 0.4286 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5100 | train 3.210948 | val 3.442658 | tok/s 2625569.44 | norm 0.4401 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5101 | train 3.098166 | val 3.442658 | tok/s 2627886.64 | norm 0.4419 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5102 | train 3.144849 | val 3.442658 | tok/s 2629025.69 | norm 0.4437 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5103 | train 3.027075 | val 3.442658 | tok/s 2622789.01 | norm 0.4716 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5104 | train 3.114317 | val 3.442658 | tok/s 2628183.62 | norm 0.4822 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5105 | train 3.129697 | val 3.442658 | tok/s 2547167.03 | norm 0.4370 | dt 205.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5106 | train 3.132078 | val 3.442658 | tok/s 2627974.84 | norm 0.4233 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5107 | train 3.166170 | val 3.442658 | tok/s 2623992.61 | norm 0.4340 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5108 | train 3.176790 | val 3.442658 | tok/s 2625536.91 | norm 0.4726 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5109 | train 2.978750 | val 3.442658 | tok/s 2627662.48 | norm 0.4898 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5110 | train 3.163877 | val 3.442658 | tok/s 2625749.33 | norm 0.4698 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5111 | train 3.028697 | val 3.442658 | tok/s 2630660.21 | norm 0.4383 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5112 | train 2.976326 | val 3.442658 | tok/s 2629618.63 | norm 0.4779 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5113 | train 3.112096 | val 3.442658 | tok/s 2620061.31 | norm 0.5098 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5114 | train 3.104140 | val 3.442658 | tok/s 2632317.98 | norm 0.4517 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5115 | train 3.109627 | val 3.442658 | tok/s 2626779.36 | norm 0.4571 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5116 | train 3.095054 | val 3.442658 | tok/s 2626770.41 | norm 0.4418 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5117 | train 3.063475 | val 3.442658 | tok/s 2620006.90 | norm 0.4228 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5118 | train 2.985235 | val 3.442658 | tok/s 2624565.71 | norm 0.4512 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5119 | train 3.078104 | val 3.442658 | tok/s 2630686.27 | norm 0.4337 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5120 | train 3.026448 | val 3.442658 | tok/s 2615860.71 | norm 0.4069 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5121 | train 3.019402 | val 3.442658 | tok/s 2630825.24 | norm 0.4675 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5122 | train 3.210816 | val 3.442658 | tok/s 2624383.35 | norm 0.4948 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5123 | train 3.036232 | val 3.442658 | tok/s 2630655.22 | norm 0.4609 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5124 | train 3.086460 | val 3.442658 | tok/s 2630715.02 | norm 0.4308 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5125 | train 3.028404 | val 3.442658 | tok/s 2626442.26 | norm 0.4688 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5126 | train 3.015318 | val 3.442658 | tok/s 2632507.71 | norm 0.5057 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5127 | train 3.104899 | val 3.442658 | tok/s 2623026.33 | norm 0.4400 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5128 | train 3.008749 | val 3.442658 | tok/s 2627208.14 | norm 0.4734 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5129 | train 2.984720 | val 3.442658 | tok/s 2629469.20 | norm 0.4953 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5130 | train 3.073533 | val 3.442658 | tok/s 2624951.77 | norm 0.4950 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5131 | train 3.161990 | val 3.442658 | tok/s 2621436.63 | norm 0.4585 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5132 | train 3.091939 | val 3.442658 | tok/s 2624924.08 | norm 0.4850 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5133 | train 3.243357 | val 3.442658 | tok/s 2625305.82 | norm 0.4684 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5134 | train 3.009488 | val 3.442658 | tok/s 2630467.26 | norm 0.4472 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5135 | train 3.213412 | val 3.442658 | tok/s 2629349.79 | norm 0.4378 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5136 | train 2.998178 | val 3.442658 | tok/s 2630234.48 | norm 0.3759 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5137 | train 3.062116 | val 3.442658 | tok/s 2619541.47 | norm 0.3847 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5138 | train 2.986890 | val 3.442658 | tok/s 2628132.05 | norm 0.4065 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5139 | train 2.985011 | val 3.442658 | tok/s 2631584.80 | norm 0.4116 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5140 | train 3.104958 | val 3.442658 | tok/s 2625449.23 | norm 0.4059 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5141 | train 3.199715 | val 3.442658 | tok/s 2630221.51 | norm 0.4318 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5142 | train 3.059670 | val 3.442658 | tok/s 2627767.15 | norm 0.4122 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5143 | train 3.054389 | val 3.442658 | tok/s 2628349.10 | norm 0.4184 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5144 | train 3.088756 | val 3.442658 | tok/s 2530442.11 | norm 0.5160 | dt 207.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5145 | train 3.241486 | val 3.442658 | tok/s 2627315.53 | norm 0.4980 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5146 | train 3.065963 | val 3.442658 | tok/s 2628155.33 | norm 0.4636 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5147 | train 3.027044 | val 3.442658 | tok/s 2626554.24 | norm 0.4432 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5148 | train 3.118198 | val 3.442658 | tok/s 2632296.33 | norm 0.4466 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5149 | train 3.082494 | val 3.442658 | tok/s 2627766.34 | norm 0.4579 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5150 | train 3.054219 | val 3.442658 | tok/s 2623170.44 | norm 0.4683 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5151 | train 3.000371 | val 3.442658 | tok/s 2462465.45 | norm 0.4321 | dt 212.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5152 | train 2.988948 | val 3.442658 | tok/s 2621349.98 | norm 0.4243 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5153 | train 2.986882 | val 3.442658 | tok/s 2625087.43 | norm 0.4790 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5154 | train 2.985614 | val 3.442658 | tok/s 2626136.41 | norm 0.4907 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5155 | train 3.152977 | val 3.442658 | tok/s 2627431.91 | norm 0.4401 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5156 | train 3.131526 | val 3.442658 | tok/s 2633179.61 | norm 0.3831 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5157 | train 3.073645 | val 3.442658 | tok/s 2625758.58 | norm 0.4513 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5158 | train 3.002889 | val 3.442658 | tok/s 2628587.33 | norm 0.4455 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5159 | train 3.003615 | val 3.442658 | tok/s 2615942.88 | norm 0.4527 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5160 | train 3.027001 | val 3.442658 | tok/s 2625048.77 | norm 0.4306 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5161 | train 3.045749 | val 3.442658 | tok/s 2632840.44 | norm 0.4483 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5162 | train 3.022422 | val 3.442658 | tok/s 2628317.90 | norm 0.4622 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5163 | train 3.043841 | val 3.442658 | tok/s 2630254.73 | norm 0.4931 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5164 | train 2.970447 | val 3.442658 | tok/s 2626413.99 | norm 0.4740 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5165 | train 3.010426 | val 3.442658 | tok/s 2631041.12 | norm 0.4448 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5166 | train 3.101209 | val 3.442658 | tok/s 2626168.98 | norm 0.4084 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5167 | train 3.096322 | val 3.442658 | tok/s 2631118.17 | norm 0.4284 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5168 | train 3.226014 | val 3.442658 | tok/s 2632368.54 | norm 0.4801 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5169 | train 3.164971 | val 3.442658 | tok/s 2614643.03 | norm 0.5284 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5170 | train 2.961808 | val 3.442658 | tok/s 2625763.13 | norm 0.4600 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5171 | train 2.984696 | val 3.442658 | tok/s 2634637.95 | norm 0.4496 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5172 | train 3.065145 | val 3.442658 | tok/s 2627204.16 | norm 0.4935 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5173 | train 3.071795 | val 3.442658 | tok/s 2630263.51 | norm 0.4690 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5174 | train 3.091130 | val 3.442658 | tok/s 2628409.95 | norm 0.4531 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5175 | train 2.971118 | val 3.442658 | tok/s 2629582.42 | norm 0.4545 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5176 | train 3.068954 | val 3.442658 | tok/s 2630204.64 | norm 0.4591 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5177 | train 2.988955 | val 3.442658 | tok/s 2627886.01 | norm 0.4570 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5178 | train 3.004744 | val 3.442658 | tok/s 2630172.46 | norm 0.4776 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5179 | train 3.034076 | val 3.442658 | tok/s 2621521.92 | norm 0.4794 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5180 | train 2.980677 | val 3.442658 | tok/s 2625648.81 | norm 0.4324 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5181 | train 3.020310 | val 3.442658 | tok/s 2632576.81 | norm 0.4769 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5182 | train 3.170061 | val 3.442658 | tok/s 2625339.81 | norm 0.5741 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5183 | train 3.088743 | val 3.442658 | tok/s 2625211.33 | norm 0.5124 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5184 | train 3.197931 | val 3.442658 | tok/s 2625871.32 | norm 0.5397 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5185 | train 3.148564 | val 3.442658 | tok/s 2630633.96 | norm 0.4975 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5186 | train 2.981237 | val 3.442658 | tok/s 2633893.71 | norm 0.3903 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5187 | train 3.144633 | val 3.442658 | tok/s 2631829.29 | norm 0.4198 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5188 | train 2.973842 | val 3.442658 | tok/s 2629518.35 | norm 0.3678 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5189 | train 3.079253 | val 3.442658 | tok/s 2628080.05 | norm 0.3712 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5190 | train 3.052741 | val 3.442658 | tok/s 2627097.93 | norm 0.3955 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5191 | train 2.969173 | val 3.442658 | tok/s 2591967.76 | norm 0.4130 | dt 202.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5192 | train 3.147316 | val 3.442658 | tok/s 2627036.06 | norm 0.4298 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5193 | train 3.153452 | val 3.442658 | tok/s 2633119.72 | norm 0.4204 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5194 | train 3.017223 | val 3.442658 | tok/s 2626039.86 | norm 0.4104 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5195 | train 3.083317 | val 3.442658 | tok/s 2628331.48 | norm 0.4227 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5196 | train 3.130563 | val 3.442658 | tok/s 2634052.77 | norm 0.4559 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5197 | train 3.286071 | val 3.442658 | tok/s 2629155.24 | norm 0.5295 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5198 | train 3.227400 | val 3.442658 | tok/s 2627113.46 | norm 0.4610 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5199 | train 2.931201 | val 3.442658 | tok/s 2623738.72 | norm 0.4128 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5200 | train 3.017196 | val 3.442658 | tok/s 2628120.56 | norm 0.4587 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5201 | train 3.017017 | val 3.442658 | tok/s 2633376.41 | norm 0.4560 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5202 | train 3.063604 | val 3.442658 | tok/s 2624869.71 | norm 0.4523 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5203 | train 3.040899 | val 3.442658 | tok/s 2626253.33 | norm 0.4229 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5204 | train 3.053332 | val 3.442658 | tok/s 2626297.98 | norm 0.3993 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5205 | train 3.166376 | val 3.442658 | tok/s 2628181.09 | norm 0.4472 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5206 | train 3.135108 | val 3.442658 | tok/s 2634101.31 | norm 0.4544 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5207 | train 3.226859 | val 3.442658 | tok/s 2629249.82 | norm 0.4565 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5208 | train 2.939447 | val 3.442658 | tok/s 2633881.42 | norm 0.4606 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5209 | train 2.984512 | val 3.442658 | tok/s 2629045.16 | norm 0.4364 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5210 | train 3.074735 | val 3.442658 | tok/s 2458895.21 | norm 0.4326 | dt 213.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5211 | train 3.065961 | val 3.442658 | tok/s 2585271.32 | norm 0.4529 | dt 202.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5212 | train 3.101820 | val 3.442658 | tok/s 2623511.75 | norm 0.4300 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5213 | train 3.065961 | val 3.442658 | tok/s 2628099.27 | norm 0.4410 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5214 | train 3.157829 | val 3.442658 | tok/s 2624884.18 | norm 0.5015 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5215 | train 3.171454 | val 3.442658 | tok/s 2627835.44 | norm 0.4752 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5216 | train 3.006937 | val 3.442658 | tok/s 2619695.45 | norm 0.4770 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5217 | train 3.091234 | val 3.442658 | tok/s 2622894.45 | norm 0.4396 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5218 | train 3.086907 | val 3.442658 | tok/s 2627847.74 | norm 0.4455 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5219 | train 3.154000 | val 3.442658 | tok/s 2614093.90 | norm 0.5280 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5220 | train 3.255314 | val 3.442658 | tok/s 2629022.39 | norm 0.5054 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5221 | train 3.070806 | val 3.442658 | tok/s 2632476.86 | norm 0.4616 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5222 | train 3.110920 | val 3.442658 | tok/s 2632624.20 | norm 0.5354 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5223 | train 2.969712 | val 3.442658 | tok/s 2631393.05 | norm 0.4570 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5224 | train 3.100132 | val 3.442658 | tok/s 2627899.34 | norm 0.4102 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5225 | train 3.141519 | val 3.442658 | tok/s 2624195.96 | norm 0.4510 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5226 | train 3.082440 | val 3.442658 | tok/s 2625415.85 | norm 0.4424 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5227 | train 3.090245 | val 3.442658 | tok/s 2628620.89 | norm 0.4453 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5228 | train 3.020180 | val 3.442658 | tok/s 2627290.81 | norm 0.4827 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5229 | train 3.153187 | val 3.442658 | tok/s 2627177.19 | norm 0.4791 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5230 | train 3.048006 | val 3.442658 | tok/s 2625535.23 | norm 0.4191 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5231 | train 3.020288 | val 3.442658 | tok/s 2628759.68 | norm 0.4473 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5232 | train 3.023193 | val 3.442658 | tok/s 2626660.93 | norm 0.4385 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5233 | train 3.022731 | val 3.442658 | tok/s 2628410.15 | norm 0.4664 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5234 | train 3.047038 | val 3.442658 | tok/s 2622126.24 | norm 0.4889 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5235 | train 3.120222 | val 3.442658 | tok/s 2629483.68 | norm 0.4619 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5236 | train 2.971937 | val 3.442658 | tok/s 2622329.87 | norm 0.4412 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5237 | train 2.941899 | val 3.442658 | tok/s 2630081.71 | norm 0.4997 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5238 | train 2.972706 | val 3.442658 | tok/s 2630791.77 | norm 0.5220 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5239 | train 3.122591 | val 3.442658 | tok/s 2628579.21 | norm 0.4821 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5240 | train 3.063345 | val 3.442658 | tok/s 2630589.93 | norm 0.4426 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5241 | train 3.057494 | val 3.442658 | tok/s 2625644.59 | norm 0.4514 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5242 | train 3.043111 | val 3.442658 | tok/s 2625281.89 | norm 0.4752 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5243 | train 3.193640 | val 3.442658 | tok/s 2628926.87 | norm 0.4940 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5244 | train 3.124467 | val 3.442658 | tok/s 2627826.70 | norm 0.4577 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5245 | train 2.969887 | val 3.442658 | tok/s 2632204.29 | norm 0.4528 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5246 | train 3.030851 | val 3.442658 | tok/s 2622352.55 | norm 0.4609 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5247 | train 3.034726 | val 3.442658 | tok/s 2539939.91 | norm 0.4185 | dt 206.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5248 | train 2.996021 | val 3.442658 | tok/s 2531581.70 | norm 0.4272 | dt 207.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5249 | train 3.083336 | val 3.442658 | tok/s 2608437.62 | norm 0.4365 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 5250 | val_loss 3.432976 step 5250 | train 3.017730 | val 3.432976 | tok/s 2610205.39 | norm 0.4426 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5251 | train 3.089612 | val 3.432976 | tok/s 2627697.39 | norm 0.4761 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5252 | train 3.146297 | val 3.432976 | tok/s 2632266.23 | norm 0.4094 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5253 | train 3.052402 | val 3.432976 | tok/s 2632072.65 | norm 0.3976 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5254 | train 3.187490 | val 3.432976 | tok/s 2533364.75 | norm 0.4372 | dt 206.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5255 | train 3.202271 | val 3.432976 | tok/s 2617180.72 | norm 0.4245 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5256 | train 3.024655 | val 3.432976 | tok/s 2628957.93 | norm 0.4052 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5257 | train 3.038466 | val 3.432976 | tok/s 2628983.96 | norm 0.4357 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5258 | train 3.099180 | val 3.432976 | tok/s 2629467.69 | norm 0.4508 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5259 | train 3.083493 | val 3.432976 | tok/s 2611913.93 | norm 0.4340 | dt 200.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5260 | train 3.000731 | val 3.432976 | tok/s 2630453.90 | norm 0.4411 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5261 | train 3.111650 | val 3.432976 | tok/s 2630256.69 | norm 0.4660 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5262 | train 3.116324 | val 3.432976 | tok/s 2624792.41 | norm 0.5196 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5263 | train 3.064395 | val 3.432976 | tok/s 2630206.03 | norm 0.4955 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5264 | train 3.118090 | val 3.432976 | tok/s 2631141.10 | norm 0.4550 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5265 | train 2.934198 | val 3.432976 | tok/s 2628456.82 | norm 0.4376 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5266 | train 3.112183 | val 3.432976 | tok/s 2630096.70 | norm 0.4455 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5267 | train 3.169478 | val 3.432976 | tok/s 2627565.97 | norm 0.4467 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5268 | train 3.060608 | val 3.432976 | tok/s 2629180.31 | norm 0.4510 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5269 | train 2.932064 | val 3.432976 | tok/s 2628570.33 | norm 0.4229 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5270 | train 3.051152 | val 3.432976 | tok/s 2627583.38 | norm 0.4520 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5271 | train 3.149036 | val 3.432976 | tok/s 2629449.30 | norm 0.4890 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5272 | train 3.002824 | val 3.432976 | tok/s 2629023.06 | norm 0.5065 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5273 | train 3.175189 | val 3.432976 | tok/s 2626981.99 | norm 0.5201 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5274 | train 3.017143 | val 3.432976 | tok/s 2628477.88 | norm 0.3724 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5275 | train 3.122092 | val 3.432976 | tok/s 2626687.94 | norm 0.4349 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5276 | train 3.222562 | val 3.432976 | tok/s 2630899.49 | norm 0.4882 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5277 | train 3.020648 | val 3.432976 | tok/s 2627622.15 | norm 0.4421 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5278 | train 3.059541 | val 3.432976 | tok/s 2627860.77 | norm 0.4381 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5279 | train 3.074634 | val 3.432976 | tok/s 2624227.03 | norm 0.4594 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5280 | train 3.034215 | val 3.432976 | tok/s 2627840.61 | norm 0.4568 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5281 | train 3.049430 | val 3.432976 | tok/s 2629490.33 | norm 0.4401 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5282 | train 3.126889 | val 3.432976 | tok/s 2625881.77 | norm 0.4467 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5283 | train 2.964897 | val 3.432976 | tok/s 2633041.23 | norm 0.4644 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5284 | train 3.024734 | val 3.432976 | tok/s 2629939.76 | norm 0.4806 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5285 | train 3.060495 | val 3.432976 | tok/s 2627947.60 | norm 0.4541 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5286 | train 3.117540 | val 3.432976 | tok/s 2629074.73 | norm 0.4549 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5287 | train 3.080209 | val 3.432976 | tok/s 2628437.29 | norm 0.4777 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5288 | train 3.143563 | val 3.432976 | tok/s 2629010.29 | norm 0.4767 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5289 | train 3.056012 | val 3.432976 | tok/s 2624304.80 | norm 0.4743 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5290 | train 3.151831 | val 3.432976 | tok/s 2625741.81 | norm 0.4469 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5291 | train 3.013442 | val 3.432976 | tok/s 2627407.47 | norm 0.4191 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5292 | train 3.143226 | val 3.432976 | tok/s 2627490.01 | norm 0.4434 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5293 | train 3.079819 | val 3.432976 | tok/s 2629379.49 | norm 0.4787 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5294 | train 3.123757 | val 3.432976 | tok/s 2627647.12 | norm 0.4785 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5295 | train 3.207029 | val 3.432976 | tok/s 2627968.73 | norm 0.4669 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5296 | train 3.201667 | val 3.432976 | tok/s 2631712.42 | norm 0.5092 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5297 | train 3.084948 | val 3.432976 | tok/s 2629171.68 | norm 0.5123 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5298 | train 3.150261 | val 3.432976 | tok/s 2626989.92 | norm 0.4366 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5299 | train 3.030355 | val 3.432976 | tok/s 2625007.52 | norm 0.4611 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5300 | train 3.131985 | val 3.432976 | tok/s 2626706.44 | norm 0.5102 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5301 | train 3.012245 | val 3.432976 | tok/s 2605540.89 | norm 0.4752 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5302 | train 3.069957 | val 3.432976 | tok/s 2621104.70 | norm 0.4339 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5303 | train 2.983856 | val 3.432976 | tok/s 2628744.31 | norm 0.4594 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5304 | train 3.069856 | val 3.432976 | tok/s 2629372.58 | norm 0.4568 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5305 | train 3.048666 | val 3.432976 | tok/s 2623455.29 | norm 0.4544 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5306 | train 3.103468 | val 3.432976 | tok/s 2627040.97 | norm 0.4384 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5307 | train 3.094421 | val 3.432976 | tok/s 2632003.00 | norm 0.4431 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5308 | train 2.961879 | val 3.432976 | tok/s 2628874.31 | norm 0.3833 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5309 | train 3.134220 | val 3.432976 | tok/s 2628429.98 | norm 0.3915 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5310 | train 3.176286 | val 3.432976 | tok/s 2625778.46 | norm 0.4233 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5311 | train 3.079640 | val 3.432976 | tok/s 2629698.04 | norm 0.4361 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5312 | train 3.047583 | val 3.432976 | tok/s 2630094.42 | norm 0.4189 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5313 | train 3.137201 | val 3.432976 | tok/s 2624040.57 | norm 0.4268 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5314 | train 3.261184 | val 3.432976 | tok/s 2628477.35 | norm 0.4726 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5315 | train 3.005105 | val 3.432976 | tok/s 2627366.37 | norm 0.4984 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5316 | train 3.028567 | val 3.432976 | tok/s 2629655.91 | norm 0.4503 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5317 | train 3.008539 | val 3.432976 | tok/s 2616623.53 | norm 0.4248 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5318 | train 3.173091 | val 3.432976 | tok/s 2628481.66 | norm 0.4233 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5319 | train 3.038716 | val 3.432976 | tok/s 2627329.88 | norm 0.4450 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5320 | train 3.120807 | val 3.432976 | tok/s 2625252.48 | norm 0.4308 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5321 | train 3.139360 | val 3.432976 | tok/s 2629390.38 | norm 0.4200 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5322 | train 3.017878 | val 3.432976 | tok/s 2620926.44 | norm 0.4415 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5323 | train 2.981039 | val 3.432976 | tok/s 2627501.83 | norm 0.4252 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5324 | train 3.150946 | val 3.432976 | tok/s 2622671.11 | norm 0.4297 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5325 | train 3.014770 | val 3.432976 | tok/s 2627406.29 | norm 0.4300 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5326 | train 3.007819 | val 3.432976 | tok/s 2609488.76 | norm 0.4293 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5327 | train 3.030253 | val 3.432976 | tok/s 2622811.07 | norm 0.4822 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5328 | train 3.126880 | val 3.432976 | tok/s 2629373.52 | norm 0.4879 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5329 | train 3.297799 | val 3.432976 | tok/s 2627500.26 | norm 0.4764 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5330 | train 3.046914 | val 3.432976 | tok/s 2623385.08 | norm 0.4759 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5331 | train 3.018984 | val 3.432976 | tok/s 2617872.60 | norm 0.4664 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5332 | train 3.023219 | val 3.432976 | tok/s 2620124.07 | norm 0.4027 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5333 | train 3.126823 | val 3.432976 | tok/s 2631550.52 | norm 0.4026 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5334 | train 3.083200 | val 3.432976 | tok/s 2625936.72 | norm 0.4025 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5335 | train 3.065346 | val 3.432976 | tok/s 2626102.97 | norm 0.4043 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5336 | train 3.052483 | val 3.432976 | tok/s 2629034.91 | norm 0.4689 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5337 | train 3.000508 | val 3.432976 | tok/s 2623218.35 | norm 0.4779 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5338 | train 2.994888 | val 3.432976 | tok/s 2626637.95 | norm 0.4834 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5339 | train 2.974176 | val 3.432976 | tok/s 2621336.89 | norm 0.4644 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5340 | train 3.049016 | val 3.432976 | tok/s 2623993.32 | norm 0.4382 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5341 | train 3.017875 | val 3.432976 | tok/s 2619127.90 | norm 0.4299 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5342 | train 3.064935 | val 3.432976 | tok/s 2615662.02 | norm 0.4451 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5343 | train 3.101354 | val 3.432976 | tok/s 2626230.44 | norm 0.4340 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5344 | train 3.115118 | val 3.432976 | tok/s 2594904.51 | norm 0.4382 | dt 202.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5345 | train 2.977180 | val 3.432976 | tok/s 2620861.72 | norm 0.4229 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5346 | train 3.011077 | val 3.432976 | tok/s 2621001.05 | norm 0.4447 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5347 | train 3.127449 | val 3.432976 | tok/s 2627627.01 | norm 0.4302 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5348 | train 2.991959 | val 3.432976 | tok/s 2604077.26 | norm 0.4062 | dt 201.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5349 | train 3.022961 | val 3.432976 | tok/s 2624678.49 | norm 0.4778 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5350 | train 3.059173 | val 3.432976 | tok/s 2621768.89 | norm 0.4547 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5351 | train 3.127370 | val 3.432976 | tok/s 2624617.47 | norm 0.4222 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5352 | train 2.974030 | val 3.432976 | tok/s 2630413.42 | norm 0.4776 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5353 | train 3.235837 | val 3.432976 | tok/s 2629515.25 | norm 0.4977 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5354 | train 2.980797 | val 3.432976 | tok/s 2628368.60 | norm 0.3642 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5355 | train 3.172473 | val 3.432976 | tok/s 2622529.62 | norm 0.4284 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5356 | train 3.156213 | val 3.432976 | tok/s 2622441.27 | norm 0.4728 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5357 | train 3.113979 | val 3.432976 | tok/s 2628380.84 | norm 0.6092 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5358 | train 2.934858 | val 3.432976 | tok/s 2623406.33 | norm 0.4284 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5359 | train 3.020574 | val 3.432976 | tok/s 2618756.36 | norm 0.4272 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5360 | train 3.059282 | val 3.432976 | tok/s 2623040.92 | norm 0.4619 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5361 | train 3.039712 | val 3.432976 | tok/s 2621730.45 | norm 0.4173 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5362 | train 3.115143 | val 3.432976 | tok/s 2625462.47 | norm 0.4590 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5363 | train 3.087208 | val 3.432976 | tok/s 2624846.42 | norm 0.4336 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5364 | train 3.292378 | val 3.432976 | tok/s 2625541.62 | norm 0.4524 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5365 | train 3.120904 | val 3.432976 | tok/s 2622932.00 | norm 0.4212 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5366 | train 3.157765 | val 3.432976 | tok/s 2620376.78 | norm 0.4423 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5367 | train 2.991837 | val 3.432976 | tok/s 2625825.60 | norm 0.4460 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5368 | train 3.104337 | val 3.432976 | tok/s 2622707.34 | norm 0.4256 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5369 | train 3.146102 | val 3.432976 | tok/s 2623789.03 | norm 0.4448 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5370 | train 3.007124 | val 3.432976 | tok/s 2626913.31 | norm 0.4212 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5371 | train 2.959816 | val 3.432976 | tok/s 2624413.00 | norm 0.4440 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5372 | train 3.060534 | val 3.432976 | tok/s 2627269.46 | norm 0.4458 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5373 | train 3.122349 | val 3.432976 | tok/s 2624930.91 | norm 0.4305 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5374 | train 3.212145 | val 3.432976 | tok/s 2627328.68 | norm 0.4544 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5375 | train 3.395859 | val 3.432976 | tok/s 2495938.51 | norm 0.5129 | dt 210.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5376 | train 3.156188 | val 3.432976 | tok/s 2625912.11 | norm 0.4689 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5377 | train 3.253800 | val 3.432976 | tok/s 2622756.81 | norm 0.4449 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5378 | train 3.024790 | val 3.432976 | tok/s 2619307.71 | norm 0.4432 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5379 | train 3.131770 | val 3.432976 | tok/s 2629133.44 | norm 0.4345 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5380 | train 3.087551 | val 3.432976 | tok/s 2627872.51 | norm 0.4421 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5381 | train 3.009921 | val 3.432976 | tok/s 2626777.05 | norm 0.4055 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5382 | train 3.125134 | val 3.432976 | tok/s 2626488.91 | norm 0.4353 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5383 | train 2.950871 | val 3.432976 | tok/s 2621108.68 | norm 0.4780 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5384 | train 2.976772 | val 3.432976 | tok/s 2628420.76 | norm 0.4567 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5385 | train 3.037284 | val 3.432976 | tok/s 2625986.40 | norm 0.4846 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5386 | train 3.101958 | val 3.432976 | tok/s 2627299.46 | norm 0.4813 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5387 | train 3.030243 | val 3.432976 | tok/s 2622193.67 | norm 0.4646 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5388 | train 3.065589 | val 3.432976 | tok/s 2622788.35 | norm 0.4346 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5389 | train 3.036581 | val 3.432976 | tok/s 2629918.52 | norm 0.4036 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5390 | train 3.132335 | val 3.432976 | tok/s 2625064.60 | norm 0.3872 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5391 | train 3.099355 | val 3.432976 | tok/s 2628227.24 | norm 0.4250 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5392 | train 3.002297 | val 3.432976 | tok/s 2621066.46 | norm 0.4350 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5393 | train 3.036814 | val 3.432976 | tok/s 2622791.63 | norm 0.4357 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5394 | train 3.083897 | val 3.432976 | tok/s 2625644.33 | norm 0.4406 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5395 | train 3.056426 | val 3.432976 | tok/s 2624825.58 | norm 0.4739 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5396 | train 3.383423 | val 3.432976 | tok/s 2631813.17 | norm 0.5105 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5397 | train 2.913936 | val 3.432976 | tok/s 2623684.21 | norm 0.4794 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5398 | train 2.986850 | val 3.432976 | tok/s 2622930.74 | norm 0.4414 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5399 | train 3.083834 | val 3.432976 | tok/s 2627762.96 | norm 0.4201 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5400 | train 3.104482 | val 3.432976 | tok/s 2626879.44 | norm 0.4429 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5401 | train 3.046344 | val 3.432976 | tok/s 2629398.33 | norm 0.4810 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5402 | train 3.183033 | val 3.432976 | tok/s 2624108.98 | norm 0.4278 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5403 | train 2.922465 | val 3.432976 | tok/s 2622598.72 | norm 0.4206 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5404 | train 3.059896 | val 3.432976 | tok/s 2623537.16 | norm 0.4157 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5405 | train 2.981504 | val 3.432976 | tok/s 2629419.83 | norm 0.4694 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5406 | train 3.124609 | val 3.432976 | tok/s 2627142.79 | norm 0.5075 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5407 | train 3.135367 | val 3.432976 | tok/s 2555496.81 | norm 0.5196 | dt 205.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5408 | train 3.084237 | val 3.432976 | tok/s 2522056.50 | norm 0.4472 | dt 207.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5409 | train 3.081256 | val 3.432976 | tok/s 2608883.31 | norm 0.4730 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5410 | train 3.165906 | val 3.432976 | tok/s 2617693.90 | norm 0.4555 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5411 | train 3.119850 | val 3.432976 | tok/s 2626289.14 | norm 0.4567 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5412 | train 3.159454 | val 3.432976 | tok/s 2620318.92 | norm 0.4713 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5413 | train 3.053455 | val 3.432976 | tok/s 2616335.44 | norm 0.4540 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5414 | train 3.157397 | val 3.432976 | tok/s 2621961.36 | norm 0.4267 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5415 | train 3.088889 | val 3.432976 | tok/s 2565114.23 | norm 0.4358 | dt 204.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5416 | train 3.047640 | val 3.432976 | tok/s 2629907.79 | norm 0.4088 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5417 | train 3.026027 | val 3.432976 | tok/s 2625391.21 | norm 0.4089 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5418 | train 3.058432 | val 3.432976 | tok/s 2629185.19 | norm 0.4482 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5419 | train 2.951265 | val 3.432976 | tok/s 2616547.93 | norm 0.4266 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5420 | train 2.990045 | val 3.432976 | tok/s 2621886.90 | norm 0.4169 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5421 | train 3.050942 | val 3.432976 | tok/s 2626661.57 | norm 0.4370 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5422 | train 2.996159 | val 3.432976 | tok/s 2627245.08 | norm 0.4555 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5423 | train 3.082049 | val 3.432976 | tok/s 2630452.02 | norm 0.4793 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5424 | train 3.078041 | val 3.432976 | tok/s 2629693.45 | norm 0.5260 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5425 | train 3.121944 | val 3.432976 | tok/s 2624031.69 | norm 0.4455 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5426 | train 3.107446 | val 3.432976 | tok/s 2631380.98 | norm 0.3655 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5427 | train 3.074570 | val 3.432976 | tok/s 2628221.38 | norm 0.4124 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5428 | train 2.996498 | val 3.432976 | tok/s 2630999.62 | norm 0.4197 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5429 | train 3.190779 | val 3.432976 | tok/s 2627572.44 | norm 0.4410 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5430 | train 3.117149 | val 3.432976 | tok/s 2622735.91 | norm 0.4494 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5431 | train 3.083074 | val 3.432976 | tok/s 2627307.39 | norm 0.4268 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5432 | train 3.053809 | val 3.432976 | tok/s 2628303.01 | norm 0.4396 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5433 | train 3.128273 | val 3.432976 | tok/s 2630062.55 | norm 0.4554 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5434 | train 3.079035 | val 3.432976 | tok/s 2632996.55 | norm 0.4584 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5435 | train 3.392423 | val 3.432976 | tok/s 2624915.40 | norm 0.5079 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5436 | train 3.023483 | val 3.432976 | tok/s 2630817.23 | norm 0.4142 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5437 | train 3.008461 | val 3.432976 | tok/s 2627255.97 | norm 0.4064 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5438 | train 2.973924 | val 3.432976 | tok/s 2629833.26 | norm 0.4475 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5439 | train 3.011469 | val 3.432976 | tok/s 2626341.37 | norm 0.4269 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5440 | train 3.177306 | val 3.432976 | tok/s 2623422.18 | norm 0.4484 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5441 | train 3.375276 | val 3.432976 | tok/s 2622402.61 | norm 0.6336 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5442 | train 3.086729 | val 3.432976 | tok/s 2619724.42 | norm 0.4184 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5443 | train 3.102801 | val 3.432976 | tok/s 2627357.22 | norm 0.3477 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5444 | train 3.062668 | val 3.432976 | tok/s 2627160.91 | norm 0.4075 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5445 | train 3.314816 | val 3.432976 | tok/s 2627992.84 | norm 0.4265 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5446 | train 3.051353 | val 3.432976 | tok/s 2623855.61 | norm 0.4067 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5447 | train 3.085137 | val 3.432976 | tok/s 2621920.43 | norm 0.3890 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5448 | train 3.183758 | val 3.432976 | tok/s 2630642.08 | norm 0.4208 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5449 | train 3.011028 | val 3.432976 | tok/s 2626490.89 | norm 0.4266 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5450 | train 3.120853 | val 3.432976 | tok/s 2625978.89 | norm 0.4213 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5451 | train 3.115308 | val 3.432976 | tok/s 2628539.01 | norm 0.4193 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5452 | train 3.040509 | val 3.432976 | tok/s 2618266.96 | norm 0.4262 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5453 | train 3.205929 | val 3.432976 | tok/s 2627553.00 | norm 0.4387 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5454 | train 3.192593 | val 3.432976 | tok/s 2623885.53 | norm 0.4738 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5455 | train 3.118793 | val 3.432976 | tok/s 2628810.88 | norm 0.4436 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5456 | train 3.010198 | val 3.432976 | tok/s 2621209.61 | norm 0.4287 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5457 | train 3.038596 | val 3.432976 | tok/s 2621019.33 | norm 0.4439 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5458 | train 3.186824 | val 3.432976 | tok/s 2632201.35 | norm 0.4545 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5459 | train 3.075832 | val 3.432976 | tok/s 2625095.33 | norm 0.4399 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5460 | train 3.098077 | val 3.432976 | tok/s 2626610.07 | norm 0.4886 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5461 | train 3.184997 | val 3.432976 | tok/s 2623532.60 | norm 0.4649 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5462 | train 3.096279 | val 3.432976 | tok/s 2618643.92 | norm 0.4555 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5463 | train 2.965742 | val 3.432976 | tok/s 2629520.79 | norm 0.4763 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5464 | train 3.049646 | val 3.432976 | tok/s 2628276.82 | norm 0.4456 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5465 | train 3.119481 | val 3.432976 | tok/s 2628109.51 | norm 0.4305 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5466 | train 3.084312 | val 3.432976 | tok/s 2627164.16 | norm 0.4154 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5467 | train 3.134405 | val 3.432976 | tok/s 2624664.91 | norm 0.4104 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5468 | train 3.058856 | val 3.432976 | tok/s 2629646.46 | norm 0.4189 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5469 | train 2.968712 | val 3.432976 | tok/s 2626433.59 | norm 0.4328 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5470 | train 3.009122 | val 3.432976 | tok/s 2629588.96 | norm 0.4459 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5471 | train 2.968249 | val 3.432976 | tok/s 2625216.28 | norm 0.4743 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5472 | train 3.011951 | val 3.432976 | tok/s 2623909.20 | norm 0.4957 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5473 | train 3.297710 | val 3.432976 | tok/s 2628898.76 | norm 0.4430 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5474 | train 3.082791 | val 3.432976 | tok/s 2629290.28 | norm 0.4194 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5475 | train 3.039482 | val 3.432976 | tok/s 2630020.89 | norm 0.3978 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5476 | train 3.095493 | val 3.432976 | tok/s 2614724.59 | norm 0.4390 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5477 | train 3.022766 | val 3.432976 | tok/s 2579436.70 | norm 0.4523 | dt 203.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5478 | train 3.161309 | val 3.432976 | tok/s 2633575.75 | norm 0.4338 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5479 | train 3.058199 | val 3.432976 | tok/s 2627479.38 | norm 0.4463 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5480 | train 2.923783 | val 3.432976 | tok/s 2629533.80 | norm 0.4288 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5481 | train 3.082156 | val 3.432976 | tok/s 2628176.92 | norm 0.4794 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5482 | train 3.003829 | val 3.432976 | tok/s 2628116.31 | norm 0.5306 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5483 | train 3.083800 | val 3.432976 | tok/s 2630713.70 | norm 0.4711 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5484 | train 3.083344 | val 3.432976 | tok/s 2619639.14 | norm 0.4369 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5485 | train 3.097309 | val 3.432976 | tok/s 2631532.90 | norm 0.4315 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5486 | train 2.972810 | val 3.432976 | tok/s 2626548.41 | norm 0.4996 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5487 | train 3.093090 | val 3.432976 | tok/s 2629589.31 | norm 0.4518 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5488 | train 2.752576 | val 3.432976 | tok/s 2626708.64 | norm 0.4546 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5489 | train 3.085097 | val 3.432976 | tok/s 2621266.75 | norm 0.3600 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5490 | train 2.943938 | val 3.432976 | tok/s 2631050.12 | norm 0.4238 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5491 | train 3.046670 | val 3.432976 | tok/s 2628542.32 | norm 0.4494 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5492 | train 3.151784 | val 3.432976 | tok/s 2629114.77 | norm 0.4522 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5493 | train 3.033767 | val 3.432976 | tok/s 2626428.82 | norm 0.4661 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5494 | train 3.167194 | val 3.432976 | tok/s 2619482.16 | norm 0.4026 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5495 | train 3.035658 | val 3.432976 | tok/s 2630150.86 | norm 0.3818 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5496 | train 2.986034 | val 3.432976 | tok/s 2627172.70 | norm 0.4233 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5497 | train 3.047748 | val 3.432976 | tok/s 2624623.69 | norm 0.4301 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5498 | train 3.049695 | val 3.432976 | tok/s 2614223.18 | norm 0.4282 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5499 | train 2.957221 | val 3.432976 | tok/s 2618140.61 | norm 0.4390 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 5500 | val_loss 3.431798 step 5500 | train 3.146930 | val 3.431798 | tok/s 2595124.11 | norm 0.4302 | dt 202.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5501 | train 3.079839 | val 3.431798 | tok/s 2622882.45 | norm 0.5045 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5502 | train 3.081798 | val 3.431798 | tok/s 2622603.10 | norm 0.4826 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5503 | train 3.077196 | val 3.431798 | tok/s 2605091.11 | norm 0.4306 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5504 | train 2.903467 | val 3.431798 | tok/s 2609261.51 | norm 0.4288 | dt 200.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5505 | train 2.977382 | val 3.431798 | tok/s 2620881.06 | norm 0.4374 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5506 | train 3.210010 | val 3.431798 | tok/s 2627048.41 | norm 0.4738 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5507 | train 3.027155 | val 3.431798 | tok/s 2620625.57 | norm 0.4296 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5508 | train 3.183001 | val 3.431798 | tok/s 2628822.17 | norm 0.4320 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5509 | train 2.944154 | val 3.431798 | tok/s 2623507.65 | norm 0.4478 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5510 | train 3.091192 | val 3.431798 | tok/s 2620693.72 | norm 0.4708 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5511 | train 3.034630 | val 3.431798 | tok/s 2630823.73 | norm 0.4438 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5512 | train 3.068595 | val 3.431798 | tok/s 2629674.06 | norm 0.4497 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5513 | train 3.130042 | val 3.431798 | tok/s 2630972.20 | norm 0.5018 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5514 | train 2.919658 | val 3.431798 | tok/s 2626158.27 | norm 0.4314 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5515 | train 3.198568 | val 3.431798 | tok/s 2623187.91 | norm 0.4688 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5516 | train 3.095434 | val 3.431798 | tok/s 2629157.54 | norm 0.4654 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5517 | train 3.160293 | val 3.431798 | tok/s 2625845.95 | norm 0.4420 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5518 | train 3.072106 | val 3.431798 | tok/s 2628560.56 | norm 0.4511 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5519 | train 3.263534 | val 3.431798 | tok/s 2594556.17 | norm 0.4564 | dt 202.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5520 | train 3.068516 | val 3.431798 | tok/s 2581600.85 | norm 0.4277 | dt 203.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5521 | train 2.915462 | val 3.431798 | tok/s 2632852.65 | norm 0.4230 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5522 | train 3.037783 | val 3.431798 | tok/s 2626399.10 | norm 0.4460 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5523 | train 2.975702 | val 3.431798 | tok/s 2631282.13 | norm 0.4844 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5524 | train 3.121356 | val 3.431798 | tok/s 2625958.93 | norm 0.4481 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5525 | train 3.112962 | val 3.431798 | tok/s 2625631.11 | norm 0.3799 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5526 | train 3.119543 | val 3.431798 | tok/s 2627936.73 | norm 0.4293 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5527 | train 3.065700 | val 3.431798 | tok/s 2626614.64 | norm 0.4681 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5528 | train 3.025922 | val 3.431798 | tok/s 2625687.19 | norm 0.4978 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5529 | train 3.023894 | val 3.431798 | tok/s 2625575.83 | norm 0.4845 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5530 | train 3.043270 | val 3.431798 | tok/s 2627953.29 | norm 0.4428 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5531 | train 3.062410 | val 3.431798 | tok/s 2629387.80 | norm 0.4663 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5532 | train 3.015480 | val 3.431798 | tok/s 2624143.71 | norm 0.4787 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5533 | train 3.012642 | val 3.431798 | tok/s 2631490.37 | norm 0.4343 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5534 | train 3.011354 | val 3.431798 | tok/s 2624539.06 | norm 0.4434 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5535 | train 3.082329 | val 3.431798 | tok/s 2624718.19 | norm 0.4418 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5536 | train 3.182883 | val 3.431798 | tok/s 2624375.34 | norm 0.4229 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5537 | train 3.067844 | val 3.431798 | tok/s 2622575.59 | norm 0.3963 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5538 | train 3.108357 | val 3.431798 | tok/s 2548473.00 | norm 0.3765 | dt 205.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5539 | train 3.133430 | val 3.431798 | tok/s 2593425.07 | norm 0.3990 | dt 202.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5540 | train 3.082677 | val 3.431798 | tok/s 2627307.90 | norm 0.4406 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5541 | train 3.050617 | val 3.431798 | tok/s 2625075.61 | norm 0.4444 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5542 | train 3.024182 | val 3.431798 | tok/s 2626401.05 | norm 0.4441 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5543 | train 3.075776 | val 3.431798 | tok/s 2628550.02 | norm 0.4869 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5544 | train 3.098473 | val 3.431798 | tok/s 2628418.44 | norm 0.4934 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5545 | train 3.062830 | val 3.431798 | tok/s 2632236.53 | norm 0.4362 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5546 | train 3.056517 | val 3.431798 | tok/s 2621732.83 | norm 0.4457 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5547 | train 3.011586 | val 3.431798 | tok/s 2627479.74 | norm 0.4753 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5548 | train 3.133988 | val 3.431798 | tok/s 2630630.37 | norm 0.4779 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5549 | train 2.950423 | val 3.431798 | tok/s 2627492.14 | norm 0.4798 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5550 | train 2.989148 | val 3.431798 | tok/s 2631909.78 | norm 0.4764 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5551 | train 3.009618 | val 3.431798 | tok/s 2627878.13 | norm 0.4749 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5552 | train 3.013477 | val 3.431798 | tok/s 2628537.79 | norm 0.4777 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5553 | train 3.106421 | val 3.431798 | tok/s 2632140.26 | norm 0.4809 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5554 | train 3.226928 | val 3.431798 | tok/s 2628562.73 | norm 0.4856 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5555 | train 3.050093 | val 3.431798 | tok/s 2631423.37 | norm 0.4340 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5556 | train 2.927656 | val 3.431798 | tok/s 2619630.74 | norm 0.4260 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5557 | train 3.062043 | val 3.431798 | tok/s 2625165.28 | norm 0.4308 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5558 | train 3.131330 | val 3.431798 | tok/s 2626757.33 | norm 0.4526 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5559 | train 3.417597 | val 3.431798 | tok/s 2627667.16 | norm 0.5630 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5560 | train 3.045794 | val 3.431798 | tok/s 2626811.46 | norm 0.4681 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5561 | train 3.048316 | val 3.431798 | tok/s 2627193.69 | norm 0.4031 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5562 | train 3.044460 | val 3.431798 | tok/s 2628996.47 | norm 0.4330 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5563 | train 3.038136 | val 3.431798 | tok/s 2630239.77 | norm 0.4448 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5564 | train 2.889388 | val 3.431798 | tok/s 2629111.77 | norm 0.4323 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5565 | train 3.031140 | val 3.431798 | tok/s 2630759.80 | norm 0.4964 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5566 | train 3.068125 | val 3.431798 | tok/s 2618766.02 | norm 0.4914 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5567 | train 3.040910 | val 3.431798 | tok/s 2624773.69 | norm 0.3893 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5568 | train 3.149291 | val 3.431798 | tok/s 2628641.66 | norm 0.4490 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5569 | train 3.036292 | val 3.431798 | tok/s 2628195.65 | norm 0.4719 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5570 | train 2.995940 | val 3.431798 | tok/s 2627729.44 | norm 0.4389 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5571 | train 3.096400 | val 3.431798 | tok/s 2631402.20 | norm 0.4133 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5572 | train 3.078517 | val 3.431798 | tok/s 2629390.40 | norm 0.4172 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5573 | train 3.104781 | val 3.431798 | tok/s 2629959.48 | norm 0.4380 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5574 | train 2.978906 | val 3.431798 | tok/s 2628325.91 | norm 0.4549 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5575 | train 3.127088 | val 3.431798 | tok/s 2632466.12 | norm 0.4585 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5576 | train 3.041839 | val 3.431798 | tok/s 2629993.45 | norm 0.4368 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5577 | train 3.095715 | val 3.431798 | tok/s 2627030.08 | norm 0.3939 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5578 | train 3.104318 | val 3.431798 | tok/s 2626950.33 | norm 0.4279 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5579 | train 3.097455 | val 3.431798 | tok/s 2633127.46 | norm 0.4227 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5580 | train 3.080164 | val 3.431798 | tok/s 2628377.67 | norm 0.4315 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5581 | train 3.124588 | val 3.431798 | tok/s 2558748.35 | norm 0.4498 | dt 204.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5582 | train 3.157958 | val 3.431798 | tok/s 2628575.70 | norm 0.4913 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5583 | train 3.006445 | val 3.431798 | tok/s 2634361.83 | norm 0.4752 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5584 | train 3.040253 | val 3.431798 | tok/s 2625291.96 | norm 0.4255 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5585 | train 3.066039 | val 3.431798 | tok/s 2616292.46 | norm 0.4218 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5586 | train 2.967455 | val 3.431798 | tok/s 2627048.43 | norm 0.4208 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5587 | train 3.091920 | val 3.431798 | tok/s 2630023.81 | norm 0.4284 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5588 | train 3.121013 | val 3.431798 | tok/s 2629366.46 | norm 0.5393 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5589 | train 3.045373 | val 3.431798 | tok/s 2626541.28 | norm 0.4124 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5590 | train 3.125815 | val 3.431798 | tok/s 2629058.56 | norm 0.4112 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5591 | train 3.150455 | val 3.431798 | tok/s 2627190.42 | norm 0.4526 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5592 | train 2.947877 | val 3.431798 | tok/s 2627208.40 | norm 0.4365 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5593 | train 3.010897 | val 3.431798 | tok/s 2625357.30 | norm 0.4161 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5594 | train 3.081429 | val 3.431798 | tok/s 2627548.67 | norm 0.4467 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5595 | train 3.024012 | val 3.431798 | tok/s 2552743.12 | norm 0.4433 | dt 205.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5596 | train 3.122530 | val 3.431798 | tok/s 2595685.77 | norm 0.4551 | dt 201.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5597 | train 2.986962 | val 3.431798 | tok/s 2628987.19 | norm 0.4565 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5598 | train 3.009836 | val 3.431798 | tok/s 2627383.03 | norm 0.4393 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5599 | train 3.041538 | val 3.431798 | tok/s 2626749.97 | norm 0.4485 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5600 | train 3.110570 | val 3.431798 | tok/s 2632049.40 | norm 0.4070 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5601 | train 3.043121 | val 3.431798 | tok/s 2630802.47 | norm 0.4248 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5602 | train 3.206447 | val 3.431798 | tok/s 2629169.50 | norm 0.4805 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5603 | train 2.997942 | val 3.431798 | tok/s 2626941.06 | norm 0.4721 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5604 | train 3.021321 | val 3.431798 | tok/s 2626753.44 | norm 0.4102 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5605 | train 3.077788 | val 3.431798 | tok/s 2632834.20 | norm 0.4383 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5606 | train 3.181650 | val 3.431798 | tok/s 2629385.36 | norm 0.4314 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5607 | train 3.120850 | val 3.431798 | tok/s 2629581.90 | norm 0.5019 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5608 | train 3.173259 | val 3.431798 | tok/s 2625288.86 | norm 0.5160 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5609 | train 3.081784 | val 3.431798 | tok/s 2627383.35 | norm 0.4859 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5610 | train 2.962470 | val 3.431798 | tok/s 2632093.52 | norm 0.4340 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5611 | train 2.999983 | val 3.431798 | tok/s 2629583.93 | norm 0.4234 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5612 | train 3.184754 | val 3.431798 | tok/s 2629683.56 | norm 0.4258 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5613 | train 3.247409 | val 3.431798 | tok/s 2620724.53 | norm 0.4529 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5614 | train 3.063217 | val 3.431798 | tok/s 2624474.05 | norm 0.4243 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5615 | train 3.050249 | val 3.431798 | tok/s 2631508.61 | norm 0.4161 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5616 | train 3.179407 | val 3.431798 | tok/s 2625369.73 | norm 0.4426 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5617 | train 3.224771 | val 3.431798 | tok/s 2627942.11 | norm 0.4137 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5618 | train 2.970005 | val 3.431798 | tok/s 2622532.63 | norm 0.4193 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5619 | train 3.083541 | val 3.431798 | tok/s 2625226.44 | norm 0.4082 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5620 | train 3.000169 | val 3.431798 | tok/s 2629435.98 | norm 0.4207 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5621 | train 3.113996 | val 3.431798 | tok/s 2627517.09 | norm 0.4752 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5622 | train 3.181381 | val 3.431798 | tok/s 2627391.66 | norm 0.4524 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5623 | train 3.186487 | val 3.431798 | tok/s 2622830.94 | norm 0.4017 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5624 | train 3.196284 | val 3.431798 | tok/s 2601176.61 | norm 0.4126 | dt 201.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5625 | train 3.083202 | val 3.431798 | tok/s 2629309.08 | norm 0.4289 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5626 | train 3.324498 | val 3.431798 | tok/s 2627039.28 | norm 0.4635 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5627 | train 3.060720 | val 3.431798 | tok/s 2631933.44 | norm 0.3885 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5628 | train 2.978989 | val 3.431798 | tok/s 2625576.46 | norm 0.4146 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5629 | train 3.127983 | val 3.431798 | tok/s 2628743.11 | norm 0.4550 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5630 | train 3.111604 | val 3.431798 | tok/s 2619063.89 | norm 0.4004 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5631 | train 3.009127 | val 3.431798 | tok/s 2589728.96 | norm 0.4156 | dt 202.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5632 | train 3.077347 | val 3.431798 | tok/s 2628840.99 | norm 0.4213 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5633 | train 3.072726 | val 3.431798 | tok/s 2623687.87 | norm 0.4401 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5634 | train 3.037581 | val 3.431798 | tok/s 2631537.35 | norm 0.4435 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5635 | train 3.058291 | val 3.431798 | tok/s 2631130.22 | norm 0.4152 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5636 | train 3.102686 | val 3.431798 | tok/s 2629098.58 | norm 0.4341 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5637 | train 2.951912 | val 3.431798 | tok/s 2526255.42 | norm 0.4688 | dt 207.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5638 | train 3.229238 | val 3.431798 | tok/s 2473254.44 | norm 0.4825 | dt 211.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5639 | train 2.936442 | val 3.431798 | tok/s 2627574.82 | norm 0.3850 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5640 | train 3.128100 | val 3.431798 | tok/s 2631855.96 | norm 0.4073 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5641 | train 3.012923 | val 3.431798 | tok/s 2630161.81 | norm 0.4291 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5642 | train 3.191489 | val 3.431798 | tok/s 2627566.51 | norm 0.3780 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5643 | train 3.208909 | val 3.431798 | tok/s 2628147.35 | norm 0.3944 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5644 | train 2.983449 | val 3.431798 | tok/s 2630400.66 | norm 0.4322 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5645 | train 3.111948 | val 3.431798 | tok/s 2598595.92 | norm 0.3973 | dt 201.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5646 | train 3.108633 | val 3.431798 | tok/s 2626886.39 | norm 0.4815 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5647 | train 3.140887 | val 3.431798 | tok/s 2627589.35 | norm 0.5059 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5648 | train 3.227622 | val 3.431798 | tok/s 2629011.12 | norm 0.4735 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5649 | train 3.063393 | val 3.431798 | tok/s 2629550.26 | norm 0.5086 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5650 | train 3.056626 | val 3.431798 | tok/s 2617621.71 | norm 0.4688 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5651 | train 3.145994 | val 3.431798 | tok/s 2622742.61 | norm 0.4715 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5652 | train 3.081873 | val 3.431798 | tok/s 2623927.86 | norm 0.4479 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5653 | train 3.174150 | val 3.431798 | tok/s 2626296.81 | norm 0.4999 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5654 | train 3.130883 | val 3.431798 | tok/s 2626883.64 | norm 0.4582 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5655 | train 3.055718 | val 3.431798 | tok/s 2628191.52 | norm 0.4410 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5656 | train 2.970910 | val 3.431798 | tok/s 2630173.23 | norm 0.4437 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5657 | train 3.060690 | val 3.431798 | tok/s 2629793.18 | norm 0.4013 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5658 | train 3.061013 | val 3.431798 | tok/s 2629251.74 | norm 0.4425 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5659 | train 3.006384 | val 3.431798 | tok/s 2630694.74 | norm 0.4741 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5660 | train 2.998463 | val 3.431798 | tok/s 2625409.59 | norm 0.4506 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5661 | train 3.084770 | val 3.431798 | tok/s 2627835.42 | norm 0.3989 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5662 | train 3.085655 | val 3.431798 | tok/s 2632425.80 | norm 0.4182 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5663 | train 3.059622 | val 3.431798 | tok/s 2625972.09 | norm 0.4119 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5664 | train 3.008760 | val 3.431798 | tok/s 2628903.75 | norm 0.4170 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5665 | train 2.980495 | val 3.431798 | tok/s 2626913.09 | norm 0.4247 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5666 | train 3.012937 | val 3.431798 | tok/s 2626413.22 | norm 0.4021 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5667 | train 3.134251 | val 3.431798 | tok/s 2630661.14 | norm 0.4309 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5668 | train 3.126212 | val 3.431798 | tok/s 2629902.91 | norm 0.4364 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5669 | train 3.141026 | val 3.431798 | tok/s 2630320.55 | norm 0.4431 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5670 | train 3.053272 | val 3.431798 | tok/s 2627817.87 | norm 0.4438 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5671 | train 3.027595 | val 3.431798 | tok/s 2628285.59 | norm 0.4458 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5672 | train 3.013959 | val 3.431798 | tok/s 2634558.12 | norm 0.4127 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5673 | train 2.994947 | val 3.431798 | tok/s 2627788.93 | norm 0.4976 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5674 | train 3.138603 | val 3.431798 | tok/s 2629144.33 | norm 0.4648 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5675 | train 3.057836 | val 3.431798 | tok/s 2624665.58 | norm 0.4021 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5676 | train 3.184531 | val 3.431798 | tok/s 2625271.03 | norm 0.4188 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5677 | train 3.031526 | val 3.431798 | tok/s 2627320.75 | norm 0.3999 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5678 | train 3.080803 | val 3.431798 | tok/s 2625346.48 | norm 0.4490 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5679 | train 3.050167 | val 3.431798 | tok/s 2627302.51 | norm 0.4659 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5680 | train 3.093374 | val 3.431798 | tok/s 2620715.19 | norm 0.4611 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5681 | train 3.017329 | val 3.431798 | tok/s 2619950.94 | norm 0.4251 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5682 | train 3.043352 | val 3.431798 | tok/s 2625161.05 | norm 0.4358 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5683 | train 3.036916 | val 3.431798 | tok/s 2625292.41 | norm 0.4537 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5684 | train 3.061492 | val 3.431798 | tok/s 2623829.47 | norm 0.4686 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5685 | train 3.060476 | val 3.431798 | tok/s 2622587.29 | norm 0.4588 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5686 | train 3.111073 | val 3.431798 | tok/s 2618509.91 | norm 0.6700 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5687 | train 3.319524 | val 3.431798 | tok/s 2623535.54 | norm 0.4894 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5688 | train 3.096484 | val 3.431798 | tok/s 2624873.58 | norm 0.5627 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5689 | train 3.060079 | val 3.431798 | tok/s 2626208.84 | norm 0.3873 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5690 | train 3.061926 | val 3.431798 | tok/s 2626901.79 | norm 0.3851 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5691 | train 3.094404 | val 3.431798 | tok/s 2627004.01 | norm 0.4432 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5692 | train 2.992091 | val 3.431798 | tok/s 2619471.04 | norm 0.4395 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5693 | train 2.988127 | val 3.431798 | tok/s 2621808.99 | norm 0.4312 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5694 | train 2.966773 | val 3.431798 | tok/s 2624121.31 | norm 0.4635 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5695 | train 3.011855 | val 3.431798 | tok/s 2622956.24 | norm 0.4669 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5696 | train 3.118151 | val 3.431798 | tok/s 2623053.47 | norm 0.4278 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5697 | train 3.115113 | val 3.431798 | tok/s 2621457.26 | norm 0.4230 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5698 | train 2.977310 | val 3.431798 | tok/s 2463458.23 | norm 0.4378 | dt 212.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5699 | train 3.051557 | val 3.431798 | tok/s 2624988.16 | norm 0.4321 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5700 | train 3.073898 | val 3.431798 | tok/s 2550039.17 | norm 0.4416 | dt 205.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5701 | train 3.097453 | val 3.431798 | tok/s 2626783.43 | norm 0.4459 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5702 | train 3.042276 | val 3.431798 | tok/s 2624538.14 | norm 0.4047 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5703 | train 2.981200 | val 3.431798 | tok/s 2627319.81 | norm 0.4941 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5704 | train 3.108186 | val 3.431798 | tok/s 2628441.39 | norm 0.4159 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5705 | train 2.986775 | val 3.431798 | tok/s 2623693.88 | norm 0.3749 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5706 | train 3.036428 | val 3.431798 | tok/s 2625111.90 | norm 0.4362 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5707 | train 2.944379 | val 3.431798 | tok/s 2620870.85 | norm 0.4652 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5708 | train 3.147820 | val 3.431798 | tok/s 2621162.93 | norm 0.4734 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5709 | train 2.989045 | val 3.431798 | tok/s 2627240.34 | norm 0.4679 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5710 | train 3.081029 | val 3.431798 | tok/s 2625987.75 | norm 0.4623 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5711 | train 3.106758 | val 3.431798 | tok/s 2628090.20 | norm 0.4615 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5712 | train 3.169192 | val 3.431798 | tok/s 2620342.40 | norm 0.4894 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5713 | train 3.221336 | val 3.431798 | tok/s 2624330.82 | norm 0.4679 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5714 | train 3.049526 | val 3.431798 | tok/s 2626814.69 | norm 0.4023 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5715 | train 3.058516 | val 3.431798 | tok/s 2624435.06 | norm 0.4034 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5716 | train 3.021718 | val 3.431798 | tok/s 2624384.89 | norm 0.4087 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5717 | train 3.133282 | val 3.431798 | tok/s 2624609.94 | norm 0.4135 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5718 | train 2.996546 | val 3.431798 | tok/s 2623931.99 | norm 0.4154 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5719 | train 2.993576 | val 3.431798 | tok/s 2626705.05 | norm 0.4897 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5720 | train 2.998857 | val 3.431798 | tok/s 2624500.69 | norm 0.5007 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5721 | train 3.095937 | val 3.431798 | tok/s 2630189.03 | norm 0.4567 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5722 | train 2.987129 | val 3.431798 | tok/s 2624924.09 | norm 0.4155 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5723 | train 3.110130 | val 3.431798 | tok/s 2626423.51 | norm 0.4098 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5724 | train 3.110955 | val 3.431798 | tok/s 2628235.43 | norm 0.3997 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5725 | train 3.100134 | val 3.431798 | tok/s 2625402.95 | norm 0.3972 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5726 | train 3.068111 | val 3.431798 | tok/s 2624488.49 | norm 0.4064 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5727 | train 3.060696 | val 3.431798 | tok/s 2622371.27 | norm 0.4058 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5728 | train 3.303182 | val 3.431798 | tok/s 2609732.73 | norm 0.4409 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5729 | train 3.066801 | val 3.431798 | tok/s 2631492.35 | norm 0.4438 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5730 | train 3.037784 | val 3.431798 | tok/s 2625345.85 | norm 0.4180 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5731 | train 3.064529 | val 3.431798 | tok/s 2628221.90 | norm 0.4130 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5732 | train 3.217829 | val 3.431798 | tok/s 2623455.47 | norm 0.4615 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5733 | train 3.122593 | val 3.431798 | tok/s 2622994.48 | norm 0.4754 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5734 | train 3.134157 | val 3.431798 | tok/s 2628494.93 | norm 0.4184 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5735 | train 3.011670 | val 3.431798 | tok/s 2626545.05 | norm 0.4464 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5736 | train 3.089974 | val 3.431798 | tok/s 2627205.40 | norm 0.4545 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5737 | train 3.184626 | val 3.431798 | tok/s 2625479.92 | norm 0.4942 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5738 | train 3.095063 | val 3.431798 | tok/s 2623524.02 | norm 0.5062 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5739 | train 3.054783 | val 3.431798 | tok/s 2628276.79 | norm 0.4366 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5740 | train 2.998467 | val 3.431798 | tok/s 2625618.22 | norm 0.4341 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5741 | train 3.047212 | val 3.431798 | tok/s 2550695.37 | norm 0.4328 | dt 205.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5742 | train 3.028438 | val 3.431798 | tok/s 2532235.64 | norm 0.4335 | dt 207.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5743 | train 3.037385 | val 3.431798 | tok/s 2588154.76 | norm 0.4462 | dt 202.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5744 | train 3.096156 | val 3.431798 | tok/s 2627688.80 | norm 0.4328 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5745 | train 3.032759 | val 3.431798 | tok/s 2624016.72 | norm 0.4225 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5746 | train 3.143481 | val 3.431798 | tok/s 2630892.64 | norm 0.3945 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5747 | train 3.011506 | val 3.431798 | tok/s 2621317.23 | norm 0.3849 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5748 | train 3.038776 | val 3.431798 | tok/s 2627162.92 | norm 0.3801 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5749 | train 2.993667 | val 3.431798 | tok/s 2613794.30 | norm 0.4098 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 5750 | val_loss 3.428548 step 5750 | train 2.943407 | val 3.428548 | tok/s 2597674.34 | norm 0.4645 | dt 201.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5751 | train 3.061126 | val 3.428548 | tok/s 2625825.76 | norm 0.4628 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5752 | train 3.136077 | val 3.428548 | tok/s 2627436.07 | norm 0.4700 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5753 | train 2.999642 | val 3.428548 | tok/s 2624565.12 | norm 0.4584 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5754 | train 3.034350 | val 3.428548 | tok/s 2626055.79 | norm 0.4305 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5755 | train 3.071706 | val 3.428548 | tok/s 2619009.00 | norm 0.4379 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5756 | train 3.009374 | val 3.428548 | tok/s 2619393.97 | norm 0.4417 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5757 | train 3.235895 | val 3.428548 | tok/s 2625818.63 | norm 0.4977 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5758 | train 3.027405 | val 3.428548 | tok/s 2626049.82 | norm 0.4504 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5759 | train 3.242173 | val 3.428548 | tok/s 2627525.68 | norm 0.4433 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5760 | train 3.020934 | val 3.428548 | tok/s 2626542.05 | norm 0.4639 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5761 | train 3.190998 | val 3.428548 | tok/s 2623933.47 | norm 0.4402 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5762 | train 3.018485 | val 3.428548 | tok/s 2395304.90 | norm 0.4126 | dt 218.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5763 | train 3.061794 | val 3.428548 | tok/s 2556543.19 | norm 0.4176 | dt 205.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5764 | train 3.026059 | val 3.428548 | tok/s 2627056.45 | norm 0.4024 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5765 | train 3.049859 | val 3.428548 | tok/s 2622759.43 | norm 0.4166 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5766 | train 3.056203 | val 3.428548 | tok/s 2625942.26 | norm 0.4271 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5767 | train 3.066898 | val 3.428548 | tok/s 2624165.08 | norm 0.4485 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5768 | train 3.194364 | val 3.428548 | tok/s 2625542.02 | norm 0.4503 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5769 | train 3.097551 | val 3.428548 | tok/s 2627826.47 | norm 0.4293 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5770 | train 3.042242 | val 3.428548 | tok/s 2622364.20 | norm 0.4249 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5771 | train 3.140792 | val 3.428548 | tok/s 2629527.43 | norm 0.4481 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5772 | train 3.030604 | val 3.428548 | tok/s 2630599.50 | norm 0.4759 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5773 | train 3.044375 | val 3.428548 | tok/s 2629474.39 | norm 0.4609 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5774 | train 2.967874 | val 3.428548 | tok/s 2634581.42 | norm 0.4522 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5775 | train 3.030871 | val 3.428548 | tok/s 2629838.51 | norm 0.4542 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5776 | train 3.123888 | val 3.428548 | tok/s 2630448.21 | norm 0.4680 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5777 | train 3.070748 | val 3.428548 | tok/s 2630521.71 | norm 0.4537 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5778 | train 3.155638 | val 3.428548 | tok/s 2629021.94 | norm 0.4859 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5779 | train 3.051285 | val 3.428548 | tok/s 2627522.87 | norm 0.4413 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5780 | train 3.062529 | val 3.428548 | tok/s 2629406.43 | norm 0.4238 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5781 | train 3.178932 | val 3.428548 | tok/s 2628788.04 | norm 0.4783 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5782 | train 3.092303 | val 3.428548 | tok/s 2627971.26 | norm 0.4616 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5783 | train 3.117187 | val 3.428548 | tok/s 2629647.63 | norm 0.4356 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5784 | train 3.099533 | val 3.428548 | tok/s 2618255.19 | norm 0.4466 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5785 | train 3.018678 | val 3.428548 | tok/s 2625290.95 | norm 0.4479 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5786 | train 3.082034 | val 3.428548 | tok/s 2630721.49 | norm 0.4212 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5787 | train 3.064424 | val 3.428548 | tok/s 2632076.39 | norm 0.3881 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5788 | train 2.982933 | val 3.428548 | tok/s 2622667.42 | norm 0.3660 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5789 | train 3.039495 | val 3.428548 | tok/s 2626740.37 | norm 0.3981 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5790 | train 3.115567 | val 3.428548 | tok/s 2629585.06 | norm 0.4345 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5791 | train 3.101619 | val 3.428548 | tok/s 2561663.02 | norm 0.4736 | dt 204.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5792 | train 2.979996 | val 3.428548 | tok/s 2584277.76 | norm 0.4021 | dt 202.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5793 | train 3.135900 | val 3.428548 | tok/s 2626397.94 | norm 0.4101 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5794 | train 3.101558 | val 3.428548 | tok/s 2630222.67 | norm 0.4308 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5795 | train 2.953739 | val 3.428548 | tok/s 2628024.15 | norm 0.4538 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5796 | train 3.062548 | val 3.428548 | tok/s 2617514.98 | norm 0.4652 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5797 | train 3.018980 | val 3.428548 | tok/s 2627205.99 | norm 0.4371 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5798 | train 3.013727 | val 3.428548 | tok/s 2630388.95 | norm 0.4386 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5799 | train 3.067776 | val 3.428548 | tok/s 2625097.40 | norm 0.4286 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5800 | train 3.022014 | val 3.428548 | tok/s 2628156.91 | norm 0.3941 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5801 | train 3.063330 | val 3.428548 | tok/s 2628379.75 | norm 0.4204 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5802 | train 3.022471 | val 3.428548 | tok/s 2627852.76 | norm 0.4381 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5803 | train 3.093885 | val 3.428548 | tok/s 2624278.79 | norm 0.4541 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5804 | train 3.173408 | val 3.428548 | tok/s 2630896.85 | norm 0.4836 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5805 | train 3.043421 | val 3.428548 | tok/s 2570671.39 | norm 0.4527 | dt 203.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5806 | train 3.016410 | val 3.428548 | tok/s 2630713.46 | norm 0.4369 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5807 | train 3.127402 | val 3.428548 | tok/s 2622666.53 | norm 0.4286 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5808 | train 3.157592 | val 3.428548 | tok/s 2624402.77 | norm 0.4109 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5809 | train 3.050010 | val 3.428548 | tok/s 2631892.11 | norm 0.4564 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5810 | train 3.043400 | val 3.428548 | tok/s 2628777.44 | norm 0.4515 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5811 | train 3.035962 | val 3.428548 | tok/s 2628935.08 | norm 0.4489 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5812 | train 3.052967 | val 3.428548 | tok/s 2621634.38 | norm 0.4693 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5813 | train 3.117126 | val 3.428548 | tok/s 2622707.57 | norm 0.4062 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5814 | train 3.070910 | val 3.428548 | tok/s 2630200.03 | norm 0.4282 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5815 | train 3.104529 | val 3.428548 | tok/s 2625549.31 | norm 0.4925 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5816 | train 3.135071 | val 3.428548 | tok/s 2629518.37 | norm 0.4367 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5817 | train 2.997974 | val 3.428548 | tok/s 2627859.15 | norm 0.6303 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5818 | train 3.087222 | val 3.428548 | tok/s 2627026.49 | norm 0.4694 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5819 | train 3.026506 | val 3.428548 | tok/s 2629645.18 | norm 0.4584 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5820 | train 3.049708 | val 3.428548 | tok/s 2626033.92 | norm 0.4422 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5821 | train 3.014207 | val 3.428548 | tok/s 2631811.83 | norm 0.4340 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5822 | train 3.047235 | val 3.428548 | tok/s 2629501.37 | norm 0.4333 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5823 | train 3.083388 | val 3.428548 | tok/s 2627042.96 | norm 0.4438 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5824 | train 3.135038 | val 3.428548 | tok/s 2626531.07 | norm 0.4028 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5825 | train 2.967413 | val 3.428548 | tok/s 2624956.81 | norm 0.4445 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5826 | train 2.968299 | val 3.428548 | tok/s 2628559.48 | norm 0.4371 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5827 | train 3.135285 | val 3.428548 | tok/s 2628308.77 | norm 0.4340 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5828 | train 2.995100 | val 3.428548 | tok/s 2625480.86 | norm 0.4317 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5829 | train 3.057333 | val 3.428548 | tok/s 2631099.93 | norm 0.4591 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5830 | train 3.077154 | val 3.428548 | tok/s 2576467.67 | norm 0.4081 | dt 203.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5831 | train 3.087657 | val 3.428548 | tok/s 2631753.94 | norm 0.4033 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5832 | train 3.065022 | val 3.428548 | tok/s 2628025.23 | norm 0.4635 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5833 | train 3.091470 | val 3.428548 | tok/s 2601398.01 | norm 0.4483 | dt 201.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5834 | train 3.008556 | val 3.428548 | tok/s 2631828.22 | norm 0.4270 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5835 | train 3.079156 | val 3.428548 | tok/s 2625186.78 | norm 0.4520 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5836 | train 3.080535 | val 3.428548 | tok/s 2624927.31 | norm 0.4593 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5837 | train 2.986951 | val 3.428548 | tok/s 2626416.94 | norm 0.4147 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5838 | train 3.000365 | val 3.428548 | tok/s 2630424.28 | norm 0.4500 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5839 | train 3.058417 | val 3.428548 | tok/s 2624307.54 | norm 0.4628 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5840 | train 3.081236 | val 3.428548 | tok/s 2574918.71 | norm 0.4011 | dt 203.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5841 | train 3.037574 | val 3.428548 | tok/s 2576046.42 | norm 0.4295 | dt 203.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5842 | train 3.142899 | val 3.428548 | tok/s 2628793.96 | norm 0.4787 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5843 | train 3.033777 | val 3.428548 | tok/s 2630511.60 | norm 0.4151 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5844 | train 2.933625 | val 3.428548 | tok/s 2627924.22 | norm 0.3807 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5845 | train 3.035736 | val 3.428548 | tok/s 2627384.22 | norm 0.3926 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5846 | train 3.107929 | val 3.428548 | tok/s 2628788.88 | norm 0.4286 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5847 | train 2.900294 | val 3.428548 | tok/s 2628627.50 | norm 0.4027 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5848 | train 3.040686 | val 3.428548 | tok/s 2628259.19 | norm 0.4074 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5849 | train 2.989218 | val 3.428548 | tok/s 2630663.43 | norm 0.4378 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5850 | train 2.988266 | val 3.428548 | tok/s 2631264.85 | norm 0.4640 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5851 | train 3.108942 | val 3.428548 | tok/s 2629178.41 | norm 0.4198 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5852 | train 3.059893 | val 3.428548 | tok/s 2627867.17 | norm 0.4321 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5853 | train 3.083421 | val 3.428548 | tok/s 2629026.11 | norm 0.4656 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5854 | train 2.976285 | val 3.428548 | tok/s 2624567.61 | norm 0.4617 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5855 | train 3.099527 | val 3.428548 | tok/s 2629515.81 | norm 0.4441 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5856 | train 3.089122 | val 3.428548 | tok/s 2631192.74 | norm 0.3924 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5857 | train 3.356362 | val 3.428548 | tok/s 2625216.72 | norm 0.4274 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5858 | train 3.078110 | val 3.428548 | tok/s 2629119.51 | norm 0.4698 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5859 | train 2.913873 | val 3.428548 | tok/s 2627740.13 | norm 0.4228 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5860 | train 3.000825 | val 3.428548 | tok/s 2628098.02 | norm 0.4680 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5861 | train 2.974294 | val 3.428548 | tok/s 2631088.16 | norm 0.4741 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5862 | train 3.068566 | val 3.428548 | tok/s 2628711.79 | norm 0.4242 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5863 | train 3.107068 | val 3.428548 | tok/s 2626190.56 | norm 0.4069 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5864 | train 3.122643 | val 3.428548 | tok/s 2616907.42 | norm 0.4704 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5865 | train 3.027477 | val 3.428548 | tok/s 2627777.54 | norm 0.4435 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5866 | train 3.120030 | val 3.428548 | tok/s 2621195.88 | norm 0.4432 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5867 | train 3.002452 | val 3.428548 | tok/s 2626390.43 | norm 0.4527 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5868 | train 3.019482 | val 3.428548 | tok/s 2619773.61 | norm 0.4511 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5869 | train 3.042071 | val 3.428548 | tok/s 2626821.67 | norm 0.4525 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5870 | train 2.963823 | val 3.428548 | tok/s 2629437.13 | norm 0.4363 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5871 | train 2.990817 | val 3.428548 | tok/s 2624744.20 | norm 0.4613 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5872 | train 3.136934 | val 3.428548 | tok/s 2627829.37 | norm 0.4061 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5873 | train 3.151908 | val 3.428548 | tok/s 2626265.42 | norm 0.4286 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5874 | train 3.030737 | val 3.428548 | tok/s 2622595.48 | norm 0.4496 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5875 | train 3.089386 | val 3.428548 | tok/s 2626946.64 | norm 0.4674 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5876 | train 2.981867 | val 3.428548 | tok/s 2630984.45 | norm 0.4581 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5877 | train 3.087359 | val 3.428548 | tok/s 2601822.63 | norm 0.4167 | dt 201.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5878 | train 3.104988 | val 3.428548 | tok/s 2634637.97 | norm 0.4181 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5879 | train 3.054151 | val 3.428548 | tok/s 2624523.24 | norm 0.4600 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5880 | train 3.048539 | val 3.428548 | tok/s 2630283.88 | norm 0.4637 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5881 | train 3.015384 | val 3.428548 | tok/s 2416160.29 | norm 0.4537 | dt 216.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5882 | train 3.056477 | val 3.428548 | tok/s 2604936.21 | norm 0.4112 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5883 | train 2.993150 | val 3.428548 | tok/s 2630113.89 | norm 0.3956 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5884 | train 3.050592 | val 3.428548 | tok/s 2625427.63 | norm 0.4001 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5885 | train 3.134309 | val 3.428548 | tok/s 2628988.10 | norm 0.4017 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5886 | train 3.063436 | val 3.428548 | tok/s 2622864.65 | norm 0.4228 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5887 | train 3.177162 | val 3.428548 | tok/s 2624330.38 | norm 0.4300 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5888 | train 3.066538 | val 3.428548 | tok/s 2512951.41 | norm 0.4052 | dt 208.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5889 | train 2.902762 | val 3.428548 | tok/s 2628565.73 | norm 0.4475 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5890 | train 3.097628 | val 3.428548 | tok/s 2630511.73 | norm 0.4744 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5891 | train 3.162419 | val 3.428548 | tok/s 2628456.60 | norm 0.4863 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5892 | train 3.038791 | val 3.428548 | tok/s 2629004.45 | norm 0.4688 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5893 | train 3.115417 | val 3.428548 | tok/s 2633010.21 | norm 0.4162 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5894 | train 3.113756 | val 3.428548 | tok/s 2629561.25 | norm 0.4245 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5895 | train 3.051783 | val 3.428548 | tok/s 2612882.40 | norm 0.4782 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5896 | train 2.988757 | val 3.428548 | tok/s 2609698.42 | norm 0.4751 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5897 | train 3.021912 | val 3.428548 | tok/s 2621220.79 | norm 0.4660 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5898 | train 2.950177 | val 3.428548 | tok/s 2630736.14 | norm 0.4355 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5899 | train 3.082902 | val 3.428548 | tok/s 2627075.05 | norm 0.4409 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5900 | train 3.029748 | val 3.428548 | tok/s 2630664.39 | norm 0.4423 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5901 | train 3.166853 | val 3.428548 | tok/s 2626928.60 | norm 0.4387 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5902 | train 3.054339 | val 3.428548 | tok/s 2631634.81 | norm 0.4244 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5903 | train 3.133915 | val 3.428548 | tok/s 2599422.05 | norm 0.4094 | dt 201.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5904 | train 3.081099 | val 3.428548 | tok/s 2628268.50 | norm 0.4489 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5905 | train 2.918040 | val 3.428548 | tok/s 2629561.33 | norm 0.4316 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5906 | train 3.196692 | val 3.428548 | tok/s 2626118.05 | norm 0.4534 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5907 | train 3.019422 | val 3.428548 | tok/s 2628562.91 | norm 0.4348 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5908 | train 2.993387 | val 3.428548 | tok/s 2630267.40 | norm 0.4090 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5909 | train 2.937285 | val 3.428548 | tok/s 2627497.96 | norm 0.4353 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5910 | train 3.069367 | val 3.428548 | tok/s 2632672.55 | norm 0.4499 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5911 | train 2.981461 | val 3.428548 | tok/s 2630200.26 | norm 0.4623 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5912 | train 3.035039 | val 3.428548 | tok/s 2625579.41 | norm 0.4555 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5913 | train 3.001926 | val 3.428548 | tok/s 2629278.88 | norm 0.4242 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5914 | train 3.019719 | val 3.428548 | tok/s 2618113.73 | norm 0.4444 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5915 | train 3.062314 | val 3.428548 | tok/s 2630737.87 | norm 0.4664 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5916 | train 3.006031 | val 3.428548 | tok/s 2628758.25 | norm 0.4229 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5917 | train 3.069584 | val 3.428548 | tok/s 2629900.22 | norm 0.4195 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5918 | train 3.069674 | val 3.428548 | tok/s 2626802.17 | norm 0.3768 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5919 | train 3.088683 | val 3.428548 | tok/s 2623221.82 | norm 0.3664 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5920 | train 2.984508 | val 3.428548 | tok/s 2628136.51 | norm 0.3962 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5921 | train 3.064790 | val 3.428548 | tok/s 2624715.84 | norm 0.4042 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5922 | train 3.037344 | val 3.428548 | tok/s 2627856.40 | norm 0.4398 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5923 | train 2.972154 | val 3.428548 | tok/s 2626207.03 | norm 0.4390 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5924 | train 3.186902 | val 3.428548 | tok/s 2628194.25 | norm 0.4360 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5925 | train 3.081931 | val 3.428548 | tok/s 2629396.46 | norm 0.4245 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5926 | train 2.849880 | val 3.428548 | tok/s 2626393.50 | norm 0.4016 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5927 | train 3.065995 | val 3.428548 | tok/s 2628197.53 | norm 0.4232 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5928 | train 3.060873 | val 3.428548 | tok/s 2628093.52 | norm 0.4449 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5929 | train 2.963754 | val 3.428548 | tok/s 2472324.85 | norm 0.4547 | dt 212.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5930 | train 3.007034 | val 3.428548 | tok/s 2627588.44 | norm 0.4481 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5931 | train 3.139048 | val 3.428548 | tok/s 2631480.27 | norm 0.4891 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5932 | train 3.051773 | val 3.428548 | tok/s 2625582.75 | norm 0.4775 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5933 | train 3.090687 | val 3.428548 | tok/s 2626938.17 | norm 0.4644 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5934 | train 3.006659 | val 3.428548 | tok/s 2604784.98 | norm 0.4633 | dt 201.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5935 | train 3.010011 | val 3.428548 | tok/s 2630611.94 | norm 0.4612 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5936 | train 2.974138 | val 3.428548 | tok/s 2626328.53 | norm 0.4246 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5937 | train 3.068916 | val 3.428548 | tok/s 2599955.60 | norm 0.4672 | dt 201.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5938 | train 3.057639 | val 3.428548 | tok/s 2624007.12 | norm 0.4050 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5939 | train 3.050218 | val 3.428548 | tok/s 2619860.76 | norm 0.4237 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5940 | train 3.025850 | val 3.428548 | tok/s 2631976.25 | norm 0.4719 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5941 | train 3.038785 | val 3.428548 | tok/s 2627278.15 | norm 0.4546 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5942 | train 3.068924 | val 3.428548 | tok/s 2626172.13 | norm 0.4320 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5943 | train 3.107895 | val 3.428548 | tok/s 2625328.33 | norm 0.4365 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5944 | train 3.122140 | val 3.428548 | tok/s 2627370.87 | norm 0.4164 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5945 | train 3.024500 | val 3.428548 | tok/s 2631068.92 | norm 0.4201 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5946 | train 3.037320 | val 3.428548 | tok/s 2629047.08 | norm 0.4332 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5947 | train 3.062198 | val 3.428548 | tok/s 2628468.75 | norm 0.4662 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5948 | train 3.004612 | val 3.428548 | tok/s 2630035.28 | norm 0.4736 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5949 | train 3.014346 | val 3.428548 | tok/s 2606952.98 | norm 0.4964 | dt 201.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5950 | train 3.039175 | val 3.428548 | tok/s 2628080.31 | norm 0.4633 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5951 | train 3.112937 | val 3.428548 | tok/s 2625205.10 | norm 0.4223 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5952 | train 3.011120 | val 3.428548 | tok/s 2629973.29 | norm 0.4085 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5953 | train 3.103748 | val 3.428548 | tok/s 2622740.43 | norm 0.3963 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5954 | train 3.081293 | val 3.428548 | tok/s 2628334.67 | norm 0.3909 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5955 | train 2.990173 | val 3.428548 | tok/s 2628229.28 | norm 0.3808 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5956 | train 3.004939 | val 3.428548 | tok/s 2629496.46 | norm 0.3917 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5957 | train 2.891435 | val 3.428548 | tok/s 2627642.60 | norm 0.4381 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5958 | train 2.991409 | val 3.428548 | tok/s 2628684.11 | norm 0.4561 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5959 | train 3.067076 | val 3.428548 | tok/s 2625824.22 | norm 0.4728 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5960 | train 3.121296 | val 3.428548 | tok/s 2618864.10 | norm 0.4577 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5961 | train 2.982132 | val 3.428548 | tok/s 2579841.87 | norm 0.4813 | dt 203.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5962 | train 3.035122 | val 3.428548 | tok/s 2628211.35 | norm 0.4569 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5963 | train 3.104298 | val 3.428548 | tok/s 2629586.60 | norm 0.4602 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5964 | train 2.965974 | val 3.428548 | tok/s 2630785.67 | norm 0.4543 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5965 | train 3.089499 | val 3.428548 | tok/s 2624011.30 | norm 0.4631 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5966 | train 2.913983 | val 3.428548 | tok/s 2451574.13 | norm 0.4683 | dt 213.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5967 | train 3.064380 | val 3.428548 | tok/s 2596588.82 | norm 0.4686 | dt 201.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5968 | train 3.013151 | val 3.428548 | tok/s 2625125.16 | norm 0.4402 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5969 | train 3.087089 | val 3.428548 | tok/s 2628500.45 | norm 0.4069 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5970 | train 3.076927 | val 3.428548 | tok/s 2623793.46 | norm 0.4026 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5971 | train 3.037521 | val 3.428548 | tok/s 2624520.18 | norm 0.4280 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5972 | train 2.914336 | val 3.428548 | tok/s 2627867.13 | norm 0.4953 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5973 | train 2.965642 | val 3.428548 | tok/s 2628407.50 | norm 0.4859 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5974 | train 3.086545 | val 3.428548 | tok/s 2633270.55 | norm 0.4237 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5975 | train 3.104852 | val 3.428548 | tok/s 2626362.46 | norm 0.4148 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5976 | train 3.102009 | val 3.428548 | tok/s 2627560.36 | norm 0.4014 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5977 | train 3.223078 | val 3.428548 | tok/s 2631324.81 | norm 0.4774 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5978 | train 3.111255 | val 3.428548 | tok/s 2618013.61 | norm 0.4442 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5979 | train 3.145023 | val 3.428548 | tok/s 2630786.29 | norm 0.3800 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5980 | train 3.235613 | val 3.428548 | tok/s 2628594.52 | norm 0.4219 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5981 | train 3.081412 | val 3.428548 | tok/s 2593699.35 | norm 0.3990 | dt 202.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5982 | train 3.015431 | val 3.428548 | tok/s 2591364.94 | norm 0.4346 | dt 202.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5983 | train 3.120380 | val 3.428548 | tok/s 2630736.81 | norm 0.4998 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5984 | train 3.178182 | val 3.428548 | tok/s 2628906.77 | norm 0.4674 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5985 | train 3.219380 | val 3.428548 | tok/s 2626543.27 | norm 0.4392 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5986 | train 3.034814 | val 3.428548 | tok/s 2630735.43 | norm 0.4806 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5987 | train 3.025887 | val 3.428548 | tok/s 2629015.51 | norm 0.4336 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5988 | train 3.100196 | val 3.428548 | tok/s 2630785.01 | norm 0.4039 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5989 | train 2.935503 | val 3.428548 | tok/s 2629654.21 | norm 0.4394 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5990 | train 3.025878 | val 3.428548 | tok/s 2631214.32 | norm 0.4197 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5991 | train 2.989952 | val 3.428548 | tok/s 2625655.46 | norm 0.4377 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5992 | train 3.075695 | val 3.428548 | tok/s 2630465.68 | norm 0.3897 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5993 | train 3.024392 | val 3.428548 | tok/s 2631900.69 | norm 0.3961 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5994 | train 3.012997 | val 3.428548 | tok/s 2628056.72 | norm 0.3930 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5995 | train 3.083910 | val 3.428548 | tok/s 2626972.35 | norm 0.4417 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5996 | train 3.096534 | val 3.428548 | tok/s 2629789.00 | norm 0.4733 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5997 | train 3.123845 | val 3.428548 | tok/s 2630669.47 | norm 0.4460 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5998 | train 3.074233 | val 3.428548 | tok/s 2628685.88 | norm 0.4489 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 5999 | train 3.032006 | val 3.428548 | tok/s 2625269.76 | norm 0.4913 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 6000 | val_loss 3.418432 step 6000 | train 3.099335 | val 3.418432 | tok/s 2607827.35 | norm 0.4800 | dt 201.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6001 | train 3.093570 | val 3.418432 | tok/s 2626767.58 | norm 0.4875 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6002 | train 3.139461 | val 3.418432 | tok/s 2625744.21 | norm 0.4580 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6003 | train 3.047172 | val 3.418432 | tok/s 2629134.31 | norm 0.4329 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6004 | train 3.059192 | val 3.418432 | tok/s 2628981.63 | norm 0.4849 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6005 | train 2.937225 | val 3.418432 | tok/s 2626690.02 | norm 0.4068 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6006 | train 3.053108 | val 3.418432 | tok/s 2629061.32 | norm 0.4249 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6007 | train 3.009885 | val 3.418432 | tok/s 2627489.20 | norm 0.5091 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6008 | train 3.068054 | val 3.418432 | tok/s 2628051.30 | norm 0.4486 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6009 | train 3.104441 | val 3.418432 | tok/s 2630181.96 | norm 0.4469 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6010 | train 3.032299 | val 3.418432 | tok/s 2629329.69 | norm 0.4547 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6011 | train 3.054739 | val 3.418432 | tok/s 2624636.83 | norm 0.4535 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6012 | train 3.094714 | val 3.418432 | tok/s 2628284.49 | norm 0.4981 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6013 | train 2.964581 | val 3.418432 | tok/s 2632812.64 | norm 0.4727 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6014 | train 3.069688 | val 3.418432 | tok/s 2627076.66 | norm 0.4722 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6015 | train 2.978293 | val 3.418432 | tok/s 2631843.06 | norm 0.4361 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6016 | train 3.152279 | val 3.418432 | tok/s 2626060.65 | norm 0.4222 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6017 | train 3.197430 | val 3.418432 | tok/s 2626596.18 | norm 0.4779 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6018 | train 3.175826 | val 3.418432 | tok/s 2623255.28 | norm 0.3964 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6019 | train 3.121684 | val 3.418432 | tok/s 2625235.61 | norm 0.3576 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6020 | train 3.174348 | val 3.418432 | tok/s 2627987.25 | norm 0.4292 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6021 | train 3.036716 | val 3.418432 | tok/s 2626618.76 | norm 0.4286 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6022 | train 2.970416 | val 3.418432 | tok/s 2626537.82 | norm 0.4191 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6023 | train 3.058292 | val 3.418432 | tok/s 2513480.56 | norm 0.4147 | dt 208.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6024 | train 3.164716 | val 3.418432 | tok/s 2627369.99 | norm 0.4067 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6025 | train 2.992318 | val 3.418432 | tok/s 2629108.44 | norm 0.3758 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6026 | train 3.092286 | val 3.418432 | tok/s 2626795.59 | norm 0.4134 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6027 | train 2.967706 | val 3.418432 | tok/s 2623756.96 | norm 0.4476 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6028 | train 3.093121 | val 3.418432 | tok/s 2626970.09 | norm 0.4552 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6029 | train 3.048428 | val 3.418432 | tok/s 2625966.75 | norm 0.4260 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6030 | train 3.549191 | val 3.418432 | tok/s 2629929.12 | norm 0.6969 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6031 | train 3.179199 | val 3.418432 | tok/s 2629624.92 | norm 0.4715 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6032 | train 2.975467 | val 3.418432 | tok/s 2627623.94 | norm 0.3850 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6033 | train 3.171625 | val 3.418432 | tok/s 2623322.36 | norm 0.4557 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6034 | train 3.049833 | val 3.418432 | tok/s 2626471.70 | norm 0.4000 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6035 | train 3.021035 | val 3.418432 | tok/s 2629305.87 | norm 0.4506 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6036 | train 2.979055 | val 3.418432 | tok/s 2629196.56 | norm 0.4558 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6037 | train 3.230722 | val 3.418432 | tok/s 2629617.45 | norm 0.4473 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6038 | train 2.951962 | val 3.418432 | tok/s 2625304.83 | norm 0.4102 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6039 | train 2.949609 | val 3.418432 | tok/s 2625662.21 | norm 0.4286 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6040 | train 3.009054 | val 3.418432 | tok/s 2627300.81 | norm 0.4119 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6041 | train 3.059557 | val 3.418432 | tok/s 2625843.80 | norm 0.3877 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6042 | train 3.107722 | val 3.418432 | tok/s 2629259.37 | norm 0.4460 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6043 | train 2.977794 | val 3.418432 | tok/s 2620249.08 | norm 0.4880 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6044 | train 3.005250 | val 3.418432 | tok/s 2628112.82 | norm 0.4331 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6045 | train 3.215939 | val 3.418432 | tok/s 2630466.91 | norm 0.4534 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6046 | train 3.133608 | val 3.418432 | tok/s 2630422.16 | norm 0.4433 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6047 | train 3.019828 | val 3.418432 | tok/s 2628093.22 | norm 0.3906 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6048 | train 3.050294 | val 3.418432 | tok/s 2627334.44 | norm 0.3585 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6049 | train 3.114609 | val 3.418432 | tok/s 2626515.32 | norm 0.4158 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6050 | train 3.077887 | val 3.418432 | tok/s 2628561.21 | norm 0.4380 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6051 | train 3.093337 | val 3.418432 | tok/s 2623823.30 | norm 0.3980 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6052 | train 3.101474 | val 3.418432 | tok/s 2629044.54 | norm 0.3925 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6053 | train 2.974736 | val 3.418432 | tok/s 2630830.49 | norm 0.4309 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6054 | train 3.058391 | val 3.418432 | tok/s 2623689.16 | norm 0.4058 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6055 | train 3.082663 | val 3.418432 | tok/s 2626042.88 | norm 0.4088 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6056 | train 3.096308 | val 3.418432 | tok/s 2628988.19 | norm 0.4330 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6057 | train 3.006784 | val 3.418432 | tok/s 2627485.24 | norm 0.4156 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6058 | train 3.094646 | val 3.418432 | tok/s 2625878.27 | norm 0.4918 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6059 | train 3.095230 | val 3.418432 | tok/s 2621153.77 | norm 0.4731 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6060 | train 2.934078 | val 3.418432 | tok/s 2630220.43 | norm 0.4626 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6061 | train 3.100601 | val 3.418432 | tok/s 2626800.68 | norm 0.4478 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6062 | train 3.022312 | val 3.418432 | tok/s 2628442.12 | norm 0.4615 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6063 | train 3.187759 | val 3.418432 | tok/s 2626452.65 | norm 0.5294 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6064 | train 3.084597 | val 3.418432 | tok/s 2629790.49 | norm 0.4405 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6065 | train 3.014664 | val 3.418432 | tok/s 2626867.23 | norm 0.4013 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6066 | train 3.139902 | val 3.418432 | tok/s 2624947.27 | norm 0.4271 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6067 | train 3.108164 | val 3.418432 | tok/s 2629256.40 | norm 0.4161 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6068 | train 3.193012 | val 3.418432 | tok/s 2629209.77 | norm 0.4223 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6069 | train 3.061486 | val 3.418432 | tok/s 2625999.44 | norm 0.4525 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6070 | train 2.990375 | val 3.418432 | tok/s 2629866.67 | norm 0.4340 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6071 | train 3.107967 | val 3.418432 | tok/s 2626347.26 | norm 0.4294 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6072 | train 3.082685 | val 3.418432 | tok/s 2628131.26 | norm 0.3901 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6073 | train 3.175143 | val 3.418432 | tok/s 2622063.04 | norm 0.4098 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6074 | train 3.023884 | val 3.418432 | tok/s 2629700.61 | norm 0.4177 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6075 | train 3.062134 | val 3.418432 | tok/s 2621221.61 | norm 0.4367 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6076 | train 3.122761 | val 3.418432 | tok/s 2622863.39 | norm 0.4521 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6077 | train 3.026761 | val 3.418432 | tok/s 2628997.46 | norm 0.4432 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6078 | train 3.062329 | val 3.418432 | tok/s 2627384.64 | norm 0.4484 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6079 | train 3.080141 | val 3.418432 | tok/s 2627504.48 | norm 0.4162 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6080 | train 3.105167 | val 3.418432 | tok/s 2628605.33 | norm 0.4193 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6081 | train 3.017408 | val 3.418432 | tok/s 2627177.10 | norm 0.4570 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6082 | train 3.078119 | val 3.418432 | tok/s 2628937.74 | norm 0.4848 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6083 | train 3.076538 | val 3.418432 | tok/s 2627005.34 | norm 0.4645 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6084 | train 3.051687 | val 3.418432 | tok/s 2622655.96 | norm 0.4021 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6085 | train 3.106389 | val 3.418432 | tok/s 2546984.98 | norm 0.4347 | dt 205.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6086 | train 3.128971 | val 3.418432 | tok/s 2627991.66 | norm 0.4971 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6087 | train 3.054142 | val 3.418432 | tok/s 2630078.31 | norm 0.4582 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6088 | train 3.062104 | val 3.418432 | tok/s 2622771.65 | norm 0.4325 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6089 | train 3.216903 | val 3.418432 | tok/s 2623940.98 | norm 0.4613 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6090 | train 3.223870 | val 3.418432 | tok/s 2627023.84 | norm 0.4707 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6091 | train 3.123581 | val 3.418432 | tok/s 2627908.87 | norm 0.4549 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6092 | train 3.047340 | val 3.418432 | tok/s 2629028.48 | norm 0.4372 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6093 | train 2.950994 | val 3.418432 | tok/s 2627867.18 | norm 0.4285 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6094 | train 3.013630 | val 3.418432 | tok/s 2627879.72 | norm 0.4311 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6095 | train 3.065111 | val 3.418432 | tok/s 2627817.81 | norm 0.4165 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6096 | train 3.268042 | val 3.418432 | tok/s 2630355.18 | norm 0.4863 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6097 | train 2.978283 | val 3.418432 | tok/s 2630937.78 | norm 0.4541 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6098 | train 3.038874 | val 3.418432 | tok/s 2622633.59 | norm 0.5160 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6099 | train 3.032107 | val 3.418432 | tok/s 2628327.11 | norm 0.5414 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6100 | train 3.075588 | val 3.418432 | tok/s 2621496.74 | norm 0.4571 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6101 | train 3.073067 | val 3.418432 | tok/s 2625125.15 | norm 0.4572 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6102 | train 3.013785 | val 3.418432 | tok/s 2632651.78 | norm 0.4869 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6103 | train 3.087636 | val 3.418432 | tok/s 2562688.36 | norm 0.4838 | dt 204.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6104 | train 3.098681 | val 3.418432 | tok/s 2627688.35 | norm 0.4554 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6105 | train 3.164075 | val 3.418432 | tok/s 2625440.59 | norm 0.4468 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6106 | train 2.936435 | val 3.418432 | tok/s 2599685.93 | norm 0.4464 | dt 201.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6107 | train 2.956972 | val 3.418432 | tok/s 2630039.44 | norm 0.4180 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6108 | train 2.976815 | val 3.418432 | tok/s 2624673.47 | norm 0.4448 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6109 | train 3.091871 | val 3.418432 | tok/s 2629598.53 | norm 0.4534 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6110 | train 3.108847 | val 3.418432 | tok/s 2622702.15 | norm 0.4631 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6111 | train 3.045572 | val 3.418432 | tok/s 2594949.11 | norm 0.4524 | dt 202.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6112 | train 3.012889 | val 3.418432 | tok/s 2629339.48 | norm 0.4577 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6113 | train 3.175687 | val 3.418432 | tok/s 2625369.94 | norm 0.4101 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6114 | train 3.082049 | val 3.418432 | tok/s 2628541.11 | norm 0.4209 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6115 | train 3.100866 | val 3.418432 | tok/s 2626347.82 | norm 0.4060 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6116 | train 3.011791 | val 3.418432 | tok/s 2627573.68 | norm 0.3940 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6117 | train 2.953711 | val 3.418432 | tok/s 2629686.39 | norm 0.4346 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6118 | train 2.972118 | val 3.418432 | tok/s 2628001.07 | norm 0.4661 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6119 | train 3.098347 | val 3.418432 | tok/s 2630819.71 | norm 0.4489 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6120 | train 3.078814 | val 3.418432 | tok/s 2625333.92 | norm 0.4419 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6121 | train 3.019762 | val 3.418432 | tok/s 2628024.88 | norm 0.4583 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6122 | train 3.124181 | val 3.418432 | tok/s 2626711.81 | norm 0.4075 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6123 | train 3.026041 | val 3.418432 | tok/s 2625532.04 | norm 0.4267 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6124 | train 3.006992 | val 3.418432 | tok/s 2629105.30 | norm 0.4606 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6125 | train 3.067916 | val 3.418432 | tok/s 2627798.34 | norm 0.4505 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6126 | train 3.102366 | val 3.418432 | tok/s 2627831.83 | norm 0.4746 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6127 | train 2.997560 | val 3.418432 | tok/s 2629911.77 | norm 0.4192 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6128 | train 3.154549 | val 3.418432 | tok/s 2623873.35 | norm 0.3925 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6129 | train 3.119982 | val 3.418432 | tok/s 2624290.44 | norm 0.4860 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6130 | train 3.053747 | val 3.418432 | tok/s 2628155.86 | norm 0.5065 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6131 | train 2.971690 | val 3.418432 | tok/s 2629352.10 | norm 0.4043 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6132 | train 3.005926 | val 3.418432 | tok/s 2624966.56 | norm 0.4067 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6133 | train 3.071276 | val 3.418432 | tok/s 2621919.42 | norm 0.4402 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6134 | train 2.970711 | val 3.418432 | tok/s 2624195.54 | norm 0.4630 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6135 | train 3.024762 | val 3.418432 | tok/s 2626928.09 | norm 0.4775 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6136 | train 3.059346 | val 3.418432 | tok/s 2631006.54 | norm 0.4662 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6137 | train 3.106107 | val 3.418432 | tok/s 2627152.53 | norm 0.4246 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6138 | train 2.960276 | val 3.418432 | tok/s 2625912.64 | norm 0.4263 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6139 | train 3.057746 | val 3.418432 | tok/s 2631747.03 | norm 0.4559 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6140 | train 3.112672 | val 3.418432 | tok/s 2628388.09 | norm 0.4124 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6141 | train 2.907955 | val 3.418432 | tok/s 2630698.32 | norm 0.4263 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6142 | train 3.134115 | val 3.418432 | tok/s 2622043.34 | norm 0.4396 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6143 | train 3.051662 | val 3.418432 | tok/s 2621433.50 | norm 0.4162 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6144 | train 3.025652 | val 3.418432 | tok/s 2631146.35 | norm 0.4477 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6145 | train 3.169910 | val 3.418432 | tok/s 2625458.62 | norm 0.4668 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6146 | train 3.035491 | val 3.418432 | tok/s 2625519.61 | norm 0.4370 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6147 | train 3.100897 | val 3.418432 | tok/s 2622564.78 | norm 0.4308 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6148 | train 3.010500 | val 3.418432 | tok/s 2626443.94 | norm 0.4002 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6149 | train 3.054448 | val 3.418432 | tok/s 2630868.70 | norm 0.4090 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6150 | train 3.020090 | val 3.418432 | tok/s 2628309.86 | norm 0.4070 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6151 | train 3.074672 | val 3.418432 | tok/s 2632885.92 | norm 0.3701 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6152 | train 3.135370 | val 3.418432 | tok/s 2474619.35 | norm 0.4045 | dt 211.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6153 | train 2.951201 | val 3.418432 | tok/s 2627720.41 | norm 0.4371 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6154 | train 3.069486 | val 3.418432 | tok/s 2632095.50 | norm 0.4595 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6155 | train 3.020377 | val 3.418432 | tok/s 2630561.60 | norm 0.4669 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6156 | train 2.964113 | val 3.418432 | tok/s 2630449.16 | norm 0.5385 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6157 | train 3.060702 | val 3.418432 | tok/s 2627324.51 | norm 0.4831 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6158 | train 3.172894 | val 3.418432 | tok/s 2627962.83 | norm 0.4578 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6159 | train 2.907043 | val 3.418432 | tok/s 2627460.35 | norm 0.4566 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6160 | train 3.162148 | val 3.418432 | tok/s 2626658.15 | norm 0.4957 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6161 | train 3.152536 | val 3.418432 | tok/s 2630844.81 | norm 0.4400 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6162 | train 3.041833 | val 3.418432 | tok/s 2630912.60 | norm 0.3961 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6163 | train 3.090390 | val 3.418432 | tok/s 2631016.43 | norm 0.4156 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6164 | train 3.009843 | val 3.418432 | tok/s 2627950.95 | norm 0.4218 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6165 | train 3.075698 | val 3.418432 | tok/s 2627987.06 | norm 0.4025 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6166 | train 3.075225 | val 3.418432 | tok/s 2627649.31 | norm 0.4091 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6167 | train 3.071912 | val 3.418432 | tok/s 2628395.33 | norm 0.4064 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6168 | train 3.024482 | val 3.418432 | tok/s 2627192.48 | norm 0.4635 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6169 | train 3.060297 | val 3.418432 | tok/s 2627130.08 | norm 0.4307 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6170 | train 2.892596 | val 3.418432 | tok/s 2624283.29 | norm 0.4086 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6171 | train 3.128936 | val 3.418432 | tok/s 2625819.85 | norm 0.4228 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6172 | train 3.077297 | val 3.418432 | tok/s 2623206.10 | norm 0.4344 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6173 | train 3.273193 | val 3.418432 | tok/s 2627213.24 | norm 0.4532 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6174 | train 3.106063 | val 3.418432 | tok/s 2629597.74 | norm 0.4518 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6175 | train 3.024018 | val 3.418432 | tok/s 2621759.32 | norm 0.4122 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6176 | train 2.943119 | val 3.418432 | tok/s 2627101.52 | norm 0.4102 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6177 | train 3.324934 | val 3.418432 | tok/s 2626133.37 | norm 0.4725 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6178 | train 3.050369 | val 3.418432 | tok/s 2626035.00 | norm 0.4544 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6179 | train 3.048166 | val 3.418432 | tok/s 2626536.56 | norm 0.3588 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6180 | train 2.953418 | val 3.418432 | tok/s 2623953.98 | norm 0.3979 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6181 | train 2.924750 | val 3.418432 | tok/s 2631285.86 | norm 0.4430 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6182 | train 3.159731 | val 3.418432 | tok/s 2629063.46 | norm 0.4681 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6183 | train 2.968542 | val 3.418432 | tok/s 2627687.58 | norm 0.4780 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6184 | train 3.185937 | val 3.418432 | tok/s 2627265.90 | norm 0.4568 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6185 | train 2.973830 | val 3.418432 | tok/s 2626301.88 | norm 0.3846 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6186 | train 3.036667 | val 3.418432 | tok/s 2632657.64 | norm 0.4239 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6187 | train 3.201145 | val 3.418432 | tok/s 2628525.45 | norm 0.4823 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6188 | train 3.052006 | val 3.418432 | tok/s 2630447.85 | norm 0.4674 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6189 | train 3.291668 | val 3.418432 | tok/s 2624027.49 | norm 0.5818 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6190 | train 3.045181 | val 3.418432 | tok/s 2624835.08 | norm 0.4063 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6191 | train 2.975328 | val 3.418432 | tok/s 2630451.26 | norm 0.3818 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6192 | train 3.055247 | val 3.418432 | tok/s 2629037.72 | norm 0.4834 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6193 | train 2.971425 | val 3.418432 | tok/s 2631299.23 | norm 0.4641 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6194 | train 3.148987 | val 3.418432 | tok/s 2619722.67 | norm 0.4083 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6195 | train 3.127515 | val 3.418432 | tok/s 2625280.22 | norm 0.4118 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6196 | train 2.965976 | val 3.418432 | tok/s 2631709.72 | norm 0.4385 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6197 | train 3.004619 | val 3.418432 | tok/s 2629447.56 | norm 0.3814 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6198 | train 3.043469 | val 3.418432 | tok/s 2631804.82 | norm 0.3717 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6199 | train 3.003278 | val 3.418432 | tok/s 2623527.63 | norm 0.4139 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6200 | train 3.111346 | val 3.418432 | tok/s 2627885.86 | norm 0.4245 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6201 | train 2.979145 | val 3.418432 | tok/s 2630465.84 | norm 0.4412 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6202 | train 2.962285 | val 3.418432 | tok/s 2629337.04 | norm 0.4438 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6203 | train 3.060383 | val 3.418432 | tok/s 2630297.38 | norm 0.4486 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6204 | train 3.050641 | val 3.418432 | tok/s 2496844.41 | norm 0.4504 | dt 209.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6205 | train 3.102699 | val 3.418432 | tok/s 2626781.63 | norm 0.4432 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6206 | train 3.014699 | val 3.418432 | tok/s 2618549.95 | norm 0.4470 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6207 | train 3.029670 | val 3.418432 | tok/s 2625105.44 | norm 0.3969 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6208 | train 3.104084 | val 3.418432 | tok/s 2627395.24 | norm 0.4240 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6209 | train 3.065721 | val 3.418432 | tok/s 2628757.88 | norm 0.4766 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6210 | train 3.009535 | val 3.418432 | tok/s 2622525.44 | norm 0.5208 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6211 | train 3.095567 | val 3.418432 | tok/s 2624721.12 | norm 0.4528 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6212 | train 3.058736 | val 3.418432 | tok/s 2629945.82 | norm 0.3908 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6213 | train 3.052967 | val 3.418432 | tok/s 2629027.20 | norm 0.4318 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6214 | train 3.040612 | val 3.418432 | tok/s 2625808.22 | norm 0.4867 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6215 | train 3.015233 | val 3.418432 | tok/s 2567773.09 | norm 0.4569 | dt 204.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6216 | train 2.956123 | val 3.418432 | tok/s 2629869.77 | norm 0.4405 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6217 | train 3.107056 | val 3.418432 | tok/s 2627876.21 | norm 0.4165 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6218 | train 3.099087 | val 3.418432 | tok/s 2626877.95 | norm 0.4048 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6219 | train 2.985087 | val 3.418432 | tok/s 2629371.64 | norm 0.4162 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6220 | train 3.075722 | val 3.418432 | tok/s 2628275.12 | norm 0.4380 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6221 | train 3.019267 | val 3.418432 | tok/s 2613954.26 | norm 0.4238 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6222 | train 3.193958 | val 3.418432 | tok/s 2629072.80 | norm 0.4307 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6223 | train 3.154948 | val 3.418432 | tok/s 2629789.36 | norm 0.4302 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6224 | train 2.977161 | val 3.418432 | tok/s 2628784.59 | norm 0.4141 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6225 | train 3.093013 | val 3.418432 | tok/s 2630248.44 | norm 0.4482 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6226 | train 3.110113 | val 3.418432 | tok/s 2630760.72 | norm 0.4480 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6227 | train 3.153643 | val 3.418432 | tok/s 2622782.47 | norm 0.4120 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6228 | train 3.028381 | val 3.418432 | tok/s 2629509.20 | norm 0.4266 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6229 | train 3.025489 | val 3.418432 | tok/s 2625705.34 | norm 0.4482 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6230 | train 3.031049 | val 3.418432 | tok/s 2626379.30 | norm 0.5106 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6231 | train 2.857209 | val 3.418432 | tok/s 2624619.16 | norm 0.4660 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6232 | train 3.096852 | val 3.418432 | tok/s 2625479.05 | norm 0.4224 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6233 | train 3.061790 | val 3.418432 | tok/s 2630457.33 | norm 0.4466 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6234 | train 3.175648 | val 3.418432 | tok/s 2626123.36 | norm 0.4642 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6235 | train 3.026928 | val 3.418432 | tok/s 2627668.65 | norm 0.4266 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6236 | train 2.923494 | val 3.418432 | tok/s 2623490.85 | norm 0.4091 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6237 | train 3.090364 | val 3.418432 | tok/s 2622784.26 | norm 0.4108 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6238 | train 3.040368 | val 3.418432 | tok/s 2630423.12 | norm 0.3973 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6239 | train 3.039754 | val 3.418432 | tok/s 2630519.74 | norm 0.4105 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6240 | train 2.973427 | val 3.418432 | tok/s 2631642.21 | norm 0.4145 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6241 | train 3.020271 | val 3.418432 | tok/s 2629892.94 | norm 0.4251 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6242 | train 3.014154 | val 3.418432 | tok/s 2627637.35 | norm 0.4426 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6243 | train 3.079148 | val 3.418432 | tok/s 2629429.53 | norm 0.4475 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6244 | train 2.970881 | val 3.418432 | tok/s 2622269.00 | norm 0.5094 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6245 | train 2.979985 | val 3.418432 | tok/s 2625727.01 | norm 0.5258 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6246 | train 2.948032 | val 3.418432 | tok/s 2627198.07 | norm 0.4534 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6247 | train 3.080508 | val 3.418432 | tok/s 2629212.91 | norm 0.5033 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6248 | train 2.990167 | val 3.418432 | tok/s 2627716.37 | norm 0.4673 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6249 | train 3.032592 | val 3.418432 | tok/s 2630482.67 | norm 0.4299 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 6250 | val_loss 3.412166 step 6250 | train 3.204320 | val 3.412166 | tok/s 2612107.05 | norm 0.4416 | dt 200.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6251 | train 3.066458 | val 3.412166 | tok/s 2628123.87 | norm 0.4425 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6252 | train 2.889837 | val 3.412166 | tok/s 2626531.66 | norm 0.4391 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6253 | train 3.104771 | val 3.412166 | tok/s 2627250.66 | norm 0.4134 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6254 | train 2.982296 | val 3.412166 | tok/s 2626106.25 | norm 0.3932 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6255 | train 3.037081 | val 3.412166 | tok/s 2625982.49 | norm 0.4301 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6256 | train 2.974935 | val 3.412166 | tok/s 2629639.07 | norm 0.4636 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6257 | train 3.084946 | val 3.412166 | tok/s 2629095.10 | norm 0.4100 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6258 | train 3.017293 | val 3.412166 | tok/s 2626657.40 | norm 0.4283 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6259 | train 3.004182 | val 3.412166 | tok/s 2628120.76 | norm 0.4707 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6260 | train 3.201702 | val 3.412166 | tok/s 2628172.27 | norm 0.4313 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6261 | train 3.008446 | val 3.412166 | tok/s 2627504.10 | norm 0.4104 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6262 | train 3.144537 | val 3.412166 | tok/s 2623630.71 | norm 0.4511 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6263 | train 2.957219 | val 3.412166 | tok/s 2628517.41 | norm 0.4019 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6264 | train 3.085295 | val 3.412166 | tok/s 2548067.39 | norm 0.3827 | dt 205.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6265 | train 2.862342 | val 3.412166 | tok/s 2626449.49 | norm 0.4293 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6266 | train 3.106394 | val 3.412166 | tok/s 2626409.34 | norm 0.4829 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6267 | train 3.067718 | val 3.412166 | tok/s 2626138.11 | norm 0.4418 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6268 | train 2.984219 | val 3.412166 | tok/s 2629179.48 | norm 0.4120 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6269 | train 3.019649 | val 3.412166 | tok/s 2631716.58 | norm 0.4576 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6270 | train 3.080684 | val 3.412166 | tok/s 2629001.88 | norm 0.4279 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6271 | train 2.985545 | val 3.412166 | tok/s 2475974.79 | norm 0.4599 | dt 211.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6272 | train 3.005460 | val 3.412166 | tok/s 2628909.57 | norm 0.4981 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6273 | train 3.093308 | val 3.412166 | tok/s 2590557.91 | norm 0.4829 | dt 202.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6274 | train 3.010939 | val 3.412166 | tok/s 2631437.59 | norm 0.4328 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6275 | train 3.000751 | val 3.412166 | tok/s 2629050.30 | norm 0.4520 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6276 | train 3.051411 | val 3.412166 | tok/s 2629908.95 | norm 0.4785 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6277 | train 3.239547 | val 3.412166 | tok/s 2627235.42 | norm 0.5385 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6278 | train 2.903337 | val 3.412166 | tok/s 2627235.19 | norm 0.4151 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6279 | train 3.010836 | val 3.412166 | tok/s 2628403.72 | norm 0.3942 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6280 | train 2.988075 | val 3.412166 | tok/s 2620631.44 | norm 0.4293 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6281 | train 3.049975 | val 3.412166 | tok/s 2631289.72 | norm 0.4293 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6282 | train 3.116653 | val 3.412166 | tok/s 2624748.35 | norm 0.5301 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6283 | train 2.931202 | val 3.412166 | tok/s 2627760.37 | norm 0.4061 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6284 | train 2.993582 | val 3.412166 | tok/s 2631174.04 | norm 0.3915 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6285 | train 2.956942 | val 3.412166 | tok/s 2628026.72 | norm 0.3968 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6286 | train 3.166566 | val 3.412166 | tok/s 2629492.20 | norm 0.4175 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6287 | train 3.023724 | val 3.412166 | tok/s 2621189.64 | norm 0.4527 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6288 | train 3.130090 | val 3.412166 | tok/s 2626648.43 | norm 0.4166 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6289 | train 3.088833 | val 3.412166 | tok/s 2624581.06 | norm 0.4345 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6290 | train 3.259798 | val 3.412166 | tok/s 2626273.20 | norm 0.5122 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6291 | train 2.989418 | val 3.412166 | tok/s 2630698.90 | norm 0.4511 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6292 | train 3.024905 | val 3.412166 | tok/s 2628629.35 | norm 0.3941 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6293 | train 3.019264 | val 3.412166 | tok/s 2627476.28 | norm 0.4578 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6294 | train 2.996312 | val 3.412166 | tok/s 2626792.77 | norm 0.4429 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6295 | train 3.024356 | val 3.412166 | tok/s 2628557.09 | norm 0.4378 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6296 | train 3.085803 | val 3.412166 | tok/s 2583914.63 | norm 0.4564 | dt 202.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6297 | train 2.900998 | val 3.412166 | tok/s 2627965.73 | norm 0.4785 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6298 | train 3.363993 | val 3.412166 | tok/s 2604479.25 | norm 0.5636 | dt 201.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6299 | train 3.023324 | val 3.412166 | tok/s 2623241.87 | norm 0.4392 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6300 | train 3.114262 | val 3.412166 | tok/s 2628505.46 | norm 0.3587 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6301 | train 2.991210 | val 3.412166 | tok/s 2630276.70 | norm 0.3752 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6302 | train 3.012504 | val 3.412166 | tok/s 2626120.14 | norm 0.3654 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6303 | train 3.119540 | val 3.412166 | tok/s 2628102.53 | norm 0.3928 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6304 | train 2.991932 | val 3.412166 | tok/s 2427427.85 | norm 0.4025 | dt 215.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6305 | train 3.052815 | val 3.412166 | tok/s 2626747.21 | norm 0.4646 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6306 | train 3.031761 | val 3.412166 | tok/s 2631614.92 | norm 0.4462 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6307 | train 3.168782 | val 3.412166 | tok/s 2627721.45 | norm 0.4087 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6308 | train 3.038023 | val 3.412166 | tok/s 2630482.78 | norm 0.3896 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6309 | train 3.057592 | val 3.412166 | tok/s 2633403.13 | norm 0.3755 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6310 | train 3.036255 | val 3.412166 | tok/s 2502329.24 | norm 0.4006 | dt 209.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6311 | train 2.903200 | val 3.412166 | tok/s 2474479.88 | norm 0.4315 | dt 211.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6312 | train 3.125329 | val 3.412166 | tok/s 2630191.59 | norm 0.4678 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6313 | train 3.226937 | val 3.412166 | tok/s 2629213.34 | norm 0.5513 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6314 | train 3.077269 | val 3.412166 | tok/s 2630371.33 | norm 0.4662 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6315 | train 2.992710 | val 3.412166 | tok/s 2629394.93 | norm 0.4284 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6316 | train 2.878873 | val 3.412166 | tok/s 2631196.61 | norm 0.4832 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6317 | train 3.103048 | val 3.412166 | tok/s 2526490.58 | norm 0.4503 | dt 207.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6318 | train 3.049331 | val 3.412166 | tok/s 2630316.00 | norm 0.4451 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6319 | train 3.079108 | val 3.412166 | tok/s 2629510.23 | norm 0.4343 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6320 | train 3.037771 | val 3.412166 | tok/s 2629366.90 | norm 0.3932 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6321 | train 3.125206 | val 3.412166 | tok/s 2629522.32 | norm 0.4625 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6322 | train 3.312905 | val 3.412166 | tok/s 2625515.51 | norm 0.4152 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6323 | train 3.078161 | val 3.412166 | tok/s 2629990.85 | norm 0.3867 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6324 | train 2.990112 | val 3.412166 | tok/s 2627251.80 | norm 0.3738 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6325 | train 3.165015 | val 3.412166 | tok/s 2628935.33 | norm 0.4844 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6326 | train 3.195357 | val 3.412166 | tok/s 2630371.19 | norm 0.4081 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6327 | train 2.991734 | val 3.412166 | tok/s 2627276.14 | norm 0.3764 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6328 | train 3.153528 | val 3.412166 | tok/s 2629550.15 | norm 0.4588 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6329 | train 3.083084 | val 3.412166 | tok/s 2632565.39 | norm 0.3953 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6330 | train 2.933261 | val 3.412166 | tok/s 2630426.69 | norm 0.4193 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6331 | train 2.929854 | val 3.412166 | tok/s 2628891.36 | norm 0.4409 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6332 | train 3.075022 | val 3.412166 | tok/s 2628010.22 | norm 0.4372 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6333 | train 3.071262 | val 3.412166 | tok/s 2627758.79 | norm 0.4498 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6334 | train 2.968592 | val 3.412166 | tok/s 2630059.68 | norm 0.4222 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6335 | train 3.081148 | val 3.412166 | tok/s 2630423.66 | norm 0.4049 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6336 | train 3.114186 | val 3.412166 | tok/s 2628544.22 | norm 0.4327 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6337 | train 3.089025 | val 3.412166 | tok/s 2626161.64 | norm 0.4172 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6338 | train 3.119372 | val 3.412166 | tok/s 2628841.62 | norm 0.4204 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6339 | train 3.179958 | val 3.412166 | tok/s 2623595.03 | norm 0.4537 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6340 | train 3.127332 | val 3.412166 | tok/s 2629490.69 | norm 0.4487 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6341 | train 3.039738 | val 3.412166 | tok/s 2626288.90 | norm 0.4218 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6342 | train 2.925684 | val 3.412166 | tok/s 2626415.47 | norm 0.4116 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6343 | train 3.033111 | val 3.412166 | tok/s 2629159.96 | norm 0.4424 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6344 | train 2.996170 | val 3.412166 | tok/s 2632576.03 | norm 0.4592 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6345 | train 3.018123 | val 3.412166 | tok/s 2625782.82 | norm 0.4155 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6346 | train 3.004874 | val 3.412166 | tok/s 2623722.42 | norm 0.4232 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6347 | train 3.020663 | val 3.412166 | tok/s 2619547.95 | norm 0.4182 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6348 | train 3.178133 | val 3.412166 | tok/s 2623429.87 | norm 0.4561 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6349 | train 3.034467 | val 3.412166 | tok/s 2625490.28 | norm 0.4094 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6350 | train 3.143522 | val 3.412166 | tok/s 2627930.15 | norm 0.3974 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6351 | train 3.292077 | val 3.412166 | tok/s 2626432.48 | norm 0.4919 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6352 | train 2.999395 | val 3.412166 | tok/s 2629904.95 | norm 0.4102 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6353 | train 3.017140 | val 3.412166 | tok/s 2630902.15 | norm 0.4153 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6354 | train 3.099396 | val 3.412166 | tok/s 2628492.07 | norm 0.4432 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6355 | train 2.969677 | val 3.412166 | tok/s 2630848.15 | norm 0.4716 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6356 | train 3.032148 | val 3.412166 | tok/s 2628704.70 | norm 0.4661 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6357 | train 2.994417 | val 3.412166 | tok/s 2630912.52 | norm 0.4505 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6358 | train 2.960356 | val 3.412166 | tok/s 2633746.93 | norm 0.4684 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6359 | train 2.941008 | val 3.412166 | tok/s 2628106.57 | norm 0.4518 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6360 | train 3.066396 | val 3.412166 | tok/s 2629079.95 | norm 0.4896 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6361 | train 2.968195 | val 3.412166 | tok/s 2626571.99 | norm 0.5025 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6362 | train 2.917030 | val 3.412166 | tok/s 2591331.73 | norm 0.4622 | dt 202.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6363 | train 2.985606 | val 3.412166 | tok/s 2627531.05 | norm 0.4327 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6364 | train 3.066078 | val 3.412166 | tok/s 2628119.51 | norm 0.4257 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6365 | train 3.053744 | val 3.412166 | tok/s 2630668.42 | norm 0.4593 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6366 | train 3.128475 | val 3.412166 | tok/s 2543454.68 | norm 0.4441 | dt 206.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6367 | train 2.983242 | val 3.412166 | tok/s 2575740.87 | norm 0.4018 | dt 203.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6368 | train 3.100695 | val 3.412166 | tok/s 2629973.03 | norm 0.3764 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6369 | train 3.187443 | val 3.412166 | tok/s 2628268.68 | norm 0.4319 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6370 | train 3.086318 | val 3.412166 | tok/s 2632065.44 | norm 0.4367 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6371 | train 3.095213 | val 3.412166 | tok/s 2628104.94 | norm 0.4375 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6372 | train 3.023523 | val 3.412166 | tok/s 2629455.01 | norm 0.4430 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6373 | train 3.070027 | val 3.412166 | tok/s 2625397.35 | norm 0.4232 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6374 | train 3.114302 | val 3.412166 | tok/s 2622964.77 | norm 0.4507 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6375 | train 3.032361 | val 3.412166 | tok/s 2629673.88 | norm 0.4218 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6376 | train 2.965262 | val 3.412166 | tok/s 2631845.74 | norm 0.3943 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6377 | train 2.968244 | val 3.412166 | tok/s 2627087.32 | norm 0.4162 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6378 | train 3.051591 | val 3.412166 | tok/s 2628889.27 | norm 0.4419 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6379 | train 3.072900 | val 3.412166 | tok/s 2623323.32 | norm 0.4055 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6380 | train 3.072917 | val 3.412166 | tok/s 2627711.89 | norm 0.3804 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6381 | train 3.019575 | val 3.412166 | tok/s 2632720.06 | norm 0.4588 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6382 | train 3.054075 | val 3.412166 | tok/s 2631323.95 | norm 0.4548 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6383 | train 2.952343 | val 3.412166 | tok/s 2618409.56 | norm 0.4353 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6384 | train 2.976982 | val 3.412166 | tok/s 2625013.10 | norm 0.4414 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6385 | train 3.029768 | val 3.412166 | tok/s 2629381.22 | norm 0.4348 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6386 | train 2.995679 | val 3.412166 | tok/s 2625539.70 | norm 0.4158 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6387 | train 2.972123 | val 3.412166 | tok/s 2619525.34 | norm 0.4268 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6388 | train 3.096148 | val 3.412166 | tok/s 2631427.13 | norm 0.4312 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6389 | train 3.090849 | val 3.412166 | tok/s 2625789.91 | norm 0.4045 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6390 | train 2.954678 | val 3.412166 | tok/s 2627996.75 | norm 0.3912 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6391 | train 3.098473 | val 3.412166 | tok/s 2619309.24 | norm 0.4262 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6392 | train 2.992181 | val 3.412166 | tok/s 2632446.84 | norm 0.4578 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6393 | train 3.124677 | val 3.412166 | tok/s 2630626.44 | norm 0.4684 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6394 | train 3.052879 | val 3.412166 | tok/s 2627649.71 | norm 0.4156 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6395 | train 3.093865 | val 3.412166 | tok/s 2633790.17 | norm 0.4212 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6396 | train 3.094287 | val 3.412166 | tok/s 2628462.28 | norm 0.4446 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6397 | train 3.000388 | val 3.412166 | tok/s 2634778.66 | norm 0.3922 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6398 | train 3.161589 | val 3.412166 | tok/s 2630341.03 | norm 0.3852 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6399 | train 3.031126 | val 3.412166 | tok/s 2627877.25 | norm 0.4100 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6400 | train 3.057048 | val 3.412166 | tok/s 2633134.83 | norm 0.4074 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6401 | train 3.027270 | val 3.412166 | tok/s 2628649.53 | norm 0.4427 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6402 | train 2.995321 | val 3.412166 | tok/s 2629403.73 | norm 0.4461 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6403 | train 3.233554 | val 3.412166 | tok/s 2627081.28 | norm 0.4847 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6404 | train 3.500266 | val 3.412166 | tok/s 2627161.81 | norm 0.3974 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6405 | train 3.082890 | val 3.412166 | tok/s 2629999.97 | norm 0.4420 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6406 | train 3.037138 | val 3.412166 | tok/s 2629292.57 | norm 0.4242 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6407 | train 3.054922 | val 3.412166 | tok/s 2626020.26 | norm 0.4626 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6408 | train 2.979843 | val 3.412166 | tok/s 2538567.19 | norm 0.4227 | dt 206.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6409 | train 2.953062 | val 3.412166 | tok/s 2624608.18 | norm 0.4209 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6410 | train 3.116712 | val 3.412166 | tok/s 2630931.52 | norm 0.4304 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6411 | train 2.983821 | val 3.412166 | tok/s 2628241.09 | norm 0.4000 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6412 | train 3.139164 | val 3.412166 | tok/s 2627804.23 | norm 0.4176 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6413 | train 3.183986 | val 3.412166 | tok/s 2627116.37 | norm 0.4584 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6414 | train 3.006924 | val 3.412166 | tok/s 2627786.47 | norm 0.4546 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6415 | train 2.904608 | val 3.412166 | tok/s 2622582.21 | norm 0.5178 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6416 | train 3.101923 | val 3.412166 | tok/s 2631174.50 | norm 0.4801 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6417 | train 3.207697 | val 3.412166 | tok/s 2630421.81 | norm 0.4304 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6418 | train 3.020491 | val 3.412166 | tok/s 2627065.62 | norm 0.4371 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6419 | train 3.193791 | val 3.412166 | tok/s 2630201.74 | norm 0.5333 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6420 | train 3.090098 | val 3.412166 | tok/s 2632968.31 | norm 0.5337 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6421 | train 2.968921 | val 3.412166 | tok/s 2632178.84 | norm 0.4562 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6422 | train 3.209906 | val 3.412166 | tok/s 2506517.14 | norm 0.5006 | dt 209.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6423 | train 3.113322 | val 3.412166 | tok/s 2579337.30 | norm 0.4938 | dt 203.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6424 | train 3.007919 | val 3.412166 | tok/s 2631461.45 | norm 0.3714 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6425 | train 3.057414 | val 3.412166 | tok/s 2632116.64 | norm 0.3630 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6426 | train 3.081747 | val 3.412166 | tok/s 2628400.20 | norm 0.4050 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6427 | train 2.997999 | val 3.412166 | tok/s 2631444.05 | norm 0.4293 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6428 | train 3.030862 | val 3.412166 | tok/s 2629542.29 | norm 0.4139 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6429 | train 2.911517 | val 3.412166 | tok/s 2628873.22 | norm 0.4732 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6430 | train 3.058208 | val 3.412166 | tok/s 2626263.45 | norm 0.4695 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6431 | train 3.005229 | val 3.412166 | tok/s 2626713.03 | norm 0.4444 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6432 | train 3.152295 | val 3.412166 | tok/s 2627988.59 | norm 0.4365 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6433 | train 3.028581 | val 3.412166 | tok/s 2628632.91 | norm 0.4075 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6434 | train 2.942994 | val 3.412166 | tok/s 2630260.62 | norm 0.4321 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6435 | train 3.243555 | val 3.412166 | tok/s 2623158.30 | norm 0.4279 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6436 | train 3.367543 | val 3.412166 | tok/s 2622870.67 | norm 0.5516 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6437 | train 3.061908 | val 3.412166 | tok/s 2627274.23 | norm 0.5123 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6438 | train 2.973243 | val 3.412166 | tok/s 2625159.13 | norm 0.4329 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6439 | train 3.089447 | val 3.412166 | tok/s 2627000.41 | norm 0.4295 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6440 | train 3.023911 | val 3.412166 | tok/s 2625474.13 | norm 0.4295 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6441 | train 3.153479 | val 3.412166 | tok/s 2627337.14 | norm 0.4135 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6442 | train 3.066794 | val 3.412166 | tok/s 2631078.64 | norm 0.3997 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6443 | train 3.069564 | val 3.412166 | tok/s 2632098.01 | norm 0.4124 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6444 | train 2.978479 | val 3.412166 | tok/s 2629325.11 | norm 0.4003 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6445 | train 3.031889 | val 3.412166 | tok/s 2626564.01 | norm 0.3981 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6446 | train 3.119796 | val 3.412166 | tok/s 2629091.47 | norm 0.4301 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6447 | train 3.105827 | val 3.412166 | tok/s 2631449.43 | norm 0.4402 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6448 | train 3.034785 | val 3.412166 | tok/s 2629124.46 | norm 0.4361 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6449 | train 2.945472 | val 3.412166 | tok/s 2628779.24 | norm 0.4126 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6450 | train 3.030662 | val 3.412166 | tok/s 2623254.18 | norm 0.4048 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6451 | train 3.036909 | val 3.412166 | tok/s 2625934.43 | norm 0.4455 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6452 | train 3.030878 | val 3.412166 | tok/s 2629203.56 | norm 0.4511 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6453 | train 3.143965 | val 3.412166 | tok/s 2627443.97 | norm 0.4124 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6454 | train 3.096059 | val 3.412166 | tok/s 2628384.38 | norm 0.4298 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6455 | train 3.096927 | val 3.412166 | tok/s 2625694.28 | norm 0.4334 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6456 | train 2.946669 | val 3.412166 | tok/s 2621903.12 | norm 0.4115 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6457 | train 3.116932 | val 3.412166 | tok/s 2632301.43 | norm 0.4179 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6458 | train 3.018022 | val 3.412166 | tok/s 2627867.59 | norm 0.3994 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6459 | train 2.975951 | val 3.412166 | tok/s 2626099.32 | norm 0.4463 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6460 | train 3.091009 | val 3.412166 | tok/s 2629213.80 | norm 0.4913 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6461 | train 3.025228 | val 3.412166 | tok/s 2626983.49 | norm 0.4296 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6462 | train 3.127207 | val 3.412166 | tok/s 2627640.98 | norm 0.4340 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6463 | train 2.993518 | val 3.412166 | tok/s 2630113.85 | norm 0.4714 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6464 | train 2.941960 | val 3.412166 | tok/s 2599478.55 | norm 0.4266 | dt 201.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6465 | train 3.061588 | val 3.412166 | tok/s 2623429.58 | norm 0.3931 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6466 | train 2.988503 | val 3.412166 | tok/s 2584719.47 | norm 0.4036 | dt 202.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6467 | train 2.964812 | val 3.412166 | tok/s 2478652.50 | norm 0.4575 | dt 211.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6468 | train 2.940939 | val 3.412166 | tok/s 2626077.22 | norm 0.4480 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6469 | train 3.142203 | val 3.412166 | tok/s 2627946.34 | norm 0.4264 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6470 | train 3.007538 | val 3.412166 | tok/s 2629308.15 | norm 0.4379 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6471 | train 3.158755 | val 3.412166 | tok/s 2631994.44 | norm 0.4139 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6472 | train 3.019105 | val 3.412166 | tok/s 2631534.03 | norm 0.4286 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6473 | train 3.007770 | val 3.412166 | tok/s 2623385.28 | norm 0.4578 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6474 | train 3.010167 | val 3.412166 | tok/s 2628240.80 | norm 0.4151 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6475 | train 3.095634 | val 3.412166 | tok/s 2631510.86 | norm 0.4753 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6476 | train 3.156787 | val 3.412166 | tok/s 2626346.99 | norm 0.4234 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6477 | train 3.057211 | val 3.412166 | tok/s 2629572.29 | norm 0.4105 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6478 | train 3.009284 | val 3.412166 | tok/s 2629402.59 | norm 0.4705 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6479 | train 3.010938 | val 3.412166 | tok/s 2630865.44 | norm 0.4398 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6480 | train 3.357727 | val 3.412166 | tok/s 2630275.82 | norm 0.4409 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6481 | train 2.997847 | val 3.412166 | tok/s 2629889.69 | norm 0.4153 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6482 | train 3.077214 | val 3.412166 | tok/s 2626613.15 | norm 0.4121 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6483 | train 3.008945 | val 3.412166 | tok/s 2623075.70 | norm 0.4227 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6484 | train 3.062043 | val 3.412166 | tok/s 2632382.44 | norm 0.4292 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6485 | train 3.082634 | val 3.412166 | tok/s 2631992.62 | norm 0.4056 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6486 | train 3.083881 | val 3.412166 | tok/s 2629303.52 | norm 0.4069 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6487 | train 3.027792 | val 3.412166 | tok/s 2628122.36 | norm 0.4368 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6488 | train 3.081589 | val 3.412166 | tok/s 2628720.53 | norm 0.4463 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6489 | train 3.070665 | val 3.412166 | tok/s 2627845.02 | norm 0.4741 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6490 | train 3.021369 | val 3.412166 | tok/s 2629342.59 | norm 0.4846 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6491 | train 2.833978 | val 3.412166 | tok/s 2626356.16 | norm 0.5009 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6492 | train 3.019588 | val 3.412166 | tok/s 2628194.86 | norm 0.4153 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6493 | train 3.142491 | val 3.412166 | tok/s 2626723.58 | norm 0.3913 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6494 | train 3.028979 | val 3.412166 | tok/s 2629453.43 | norm 0.4193 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6495 | train 3.121179 | val 3.412166 | tok/s 2624329.91 | norm 0.4275 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6496 | train 2.990488 | val 3.412166 | tok/s 2628827.85 | norm 0.3929 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6497 | train 3.075646 | val 3.412166 | tok/s 2625010.16 | norm 0.3787 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6498 | train 3.373907 | val 3.412166 | tok/s 2626853.46 | norm 0.4968 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6499 | train 2.975638 | val 3.412166 | tok/s 2626057.75 | norm 0.4745 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 6500 | val_loss 3.410660 step 6500 | train 3.035968 | val 3.410660 | tok/s 2522140.61 | norm 0.4243 | dt 207.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6501 | train 2.976349 | val 3.410660 | tok/s 2626313.24 | norm 0.4519 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6502 | train 2.988319 | val 3.410660 | tok/s 2630277.75 | norm 0.4448 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6503 | train 2.932254 | val 3.410660 | tok/s 2625087.70 | norm 0.4308 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6504 | train 2.967318 | val 3.410660 | tok/s 2623002.97 | norm 0.5013 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6505 | train 3.060756 | val 3.410660 | tok/s 2630867.40 | norm 0.4462 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6506 | train 2.992138 | val 3.410660 | tok/s 2627997.73 | norm 0.4134 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6507 | train 3.027323 | val 3.410660 | tok/s 2631351.80 | norm 0.4217 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6508 | train 3.161314 | val 3.410660 | tok/s 2630554.37 | norm 0.4534 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6509 | train 2.975564 | val 3.410660 | tok/s 2625497.04 | norm 0.4091 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6510 | train 2.999372 | val 3.410660 | tok/s 2629869.09 | norm 0.3573 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6511 | train 2.943492 | val 3.410660 | tok/s 2631777.86 | norm 0.4167 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6512 | train 3.077479 | val 3.410660 | tok/s 2630510.97 | norm 0.4296 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6513 | train 3.092298 | val 3.410660 | tok/s 2622286.84 | norm 0.4535 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6514 | train 2.906929 | val 3.410660 | tok/s 2618890.20 | norm 0.4604 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6515 | train 2.961815 | val 3.410660 | tok/s 2630805.52 | norm 0.4868 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6516 | train 3.112936 | val 3.410660 | tok/s 2631048.48 | norm 0.4752 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6517 | train 3.210239 | val 3.410660 | tok/s 2631810.73 | norm 0.4239 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6518 | train 3.063504 | val 3.410660 | tok/s 2624093.12 | norm 0.3923 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6519 | train 3.021958 | val 3.410660 | tok/s 2626244.08 | norm 0.4208 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6520 | train 2.970789 | val 3.410660 | tok/s 2630437.60 | norm 0.4330 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6521 | train 2.947817 | val 3.410660 | tok/s 2626128.89 | norm 0.4531 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6522 | train 3.094701 | val 3.410660 | tok/s 2629447.38 | norm 0.4497 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6523 | train 3.091110 | val 3.410660 | tok/s 2625276.62 | norm 0.4207 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6524 | train 2.964624 | val 3.410660 | tok/s 2610258.88 | norm 0.4370 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6525 | train 2.964917 | val 3.410660 | tok/s 2627576.80 | norm 0.4501 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6526 | train 3.216179 | val 3.410660 | tok/s 2630953.14 | norm 0.4355 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6527 | train 3.008319 | val 3.410660 | tok/s 2629106.61 | norm 0.4434 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6528 | train 3.006688 | val 3.410660 | tok/s 2623823.02 | norm 0.4136 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6529 | train 2.986497 | val 3.410660 | tok/s 2627741.78 | norm 0.3938 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6530 | train 2.874650 | val 3.410660 | tok/s 2629160.60 | norm 0.4355 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6531 | train 3.086255 | val 3.410660 | tok/s 2627099.52 | norm 0.4423 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6532 | train 3.120731 | val 3.410660 | tok/s 2485939.49 | norm 0.4479 | dt 210.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6533 | train 3.004262 | val 3.410660 | tok/s 2623799.78 | norm 0.4369 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6534 | train 3.010028 | val 3.410660 | tok/s 2630808.70 | norm 0.4226 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6535 | train 3.127822 | val 3.410660 | tok/s 2629426.71 | norm 0.4227 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6536 | train 3.016757 | val 3.410660 | tok/s 2625801.48 | norm 0.4479 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6537 | train 3.067638 | val 3.410660 | tok/s 2630386.92 | norm 0.4307 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6538 | train 3.136176 | val 3.410660 | tok/s 2610385.12 | norm 0.4136 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6539 | train 3.142259 | val 3.410660 | tok/s 2619511.40 | norm 0.3986 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6540 | train 3.070377 | val 3.410660 | tok/s 2626825.06 | norm 0.4135 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6541 | train 2.952677 | val 3.410660 | tok/s 2627215.68 | norm 0.4236 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6542 | train 3.077718 | val 3.410660 | tok/s 2628523.83 | norm 0.3937 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6543 | train 2.983559 | val 3.410660 | tok/s 2626395.21 | norm 0.4220 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6544 | train 3.057690 | val 3.410660 | tok/s 2629820.12 | norm 0.4233 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6545 | train 3.110691 | val 3.410660 | tok/s 2630448.29 | norm 0.3792 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6546 | train 3.041364 | val 3.410660 | tok/s 2626636.56 | norm 0.3654 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6547 | train 3.068622 | val 3.410660 | tok/s 2633074.15 | norm 0.3721 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6548 | train 3.057268 | val 3.410660 | tok/s 2624841.76 | norm 0.4021 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6549 | train 3.040032 | val 3.410660 | tok/s 2630144.74 | norm 0.4019 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6550 | train 2.948527 | val 3.410660 | tok/s 2626615.71 | norm 0.3982 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6551 | train 2.974277 | val 3.410660 | tok/s 2626526.66 | norm 0.4302 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6552 | train 2.956308 | val 3.410660 | tok/s 2628193.88 | norm 0.4381 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6553 | train 3.007587 | val 3.410660 | tok/s 2625961.43 | norm 0.4788 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6554 | train 2.952939 | val 3.410660 | tok/s 2628751.10 | norm 0.4899 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6555 | train 3.031262 | val 3.410660 | tok/s 2634056.15 | norm 0.4631 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6556 | train 3.048250 | val 3.410660 | tok/s 2627951.64 | norm 0.4689 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6557 | train 3.045546 | val 3.410660 | tok/s 2632093.03 | norm 0.4544 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6558 | train 3.143261 | val 3.410660 | tok/s 2628022.99 | norm 0.4488 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6559 | train 2.980550 | val 3.410660 | tok/s 2624977.05 | norm 0.4498 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6560 | train 3.027090 | val 3.410660 | tok/s 2630408.47 | norm 0.4276 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6561 | train 3.117445 | val 3.410660 | tok/s 2625763.33 | norm 0.4455 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6562 | train 2.950857 | val 3.410660 | tok/s 2630874.50 | norm 0.4399 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6563 | train 3.008348 | val 3.410660 | tok/s 2627759.51 | norm 0.4174 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6564 | train 3.065613 | val 3.410660 | tok/s 2630311.12 | norm 0.4545 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6565 | train 3.082201 | val 3.410660 | tok/s 2624204.61 | norm 0.4545 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6566 | train 3.163481 | val 3.410660 | tok/s 2629470.26 | norm 0.4654 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6567 | train 3.067694 | val 3.410660 | tok/s 2619628.29 | norm 0.4542 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6568 | train 3.080211 | val 3.410660 | tok/s 2630937.18 | norm 0.3958 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6569 | train 3.048445 | val 3.410660 | tok/s 2632965.95 | norm 0.4127 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6570 | train 3.056103 | val 3.410660 | tok/s 2631190.39 | norm 0.4302 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6571 | train 3.052540 | val 3.410660 | tok/s 2629315.37 | norm 0.4394 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6572 | train 3.092021 | val 3.410660 | tok/s 2632022.51 | norm 0.4235 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6573 | train 3.079118 | val 3.410660 | tok/s 2631765.51 | norm 0.4087 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6574 | train 3.078747 | val 3.410660 | tok/s 2632372.16 | norm 0.4083 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6575 | train 3.169213 | val 3.410660 | tok/s 2625737.28 | norm 0.4038 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6576 | train 2.929493 | val 3.410660 | tok/s 2630575.09 | norm 0.3997 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6577 | train 3.249283 | val 3.410660 | tok/s 2627268.03 | norm 0.4575 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6578 | train 2.965255 | val 3.410660 | tok/s 2632396.22 | norm 0.5329 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6579 | train 3.027625 | val 3.410660 | tok/s 2631592.21 | norm 0.4666 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6580 | train 3.044959 | val 3.410660 | tok/s 2624178.82 | norm 0.4926 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6581 | train 3.083802 | val 3.410660 | tok/s 2626743.22 | norm 0.5744 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6582 | train 3.011214 | val 3.410660 | tok/s 2631418.55 | norm 0.4394 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6583 | train 3.113271 | val 3.410660 | tok/s 2629469.81 | norm 0.4112 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6584 | train 3.041892 | val 3.410660 | tok/s 2633282.31 | norm 0.4483 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6585 | train 3.117416 | val 3.410660 | tok/s 2627026.79 | norm 0.4208 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6586 | train 3.094902 | val 3.410660 | tok/s 2625995.64 | norm 0.4142 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6587 | train 3.034200 | val 3.410660 | tok/s 2627938.12 | norm 0.4180 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6588 | train 3.188960 | val 3.410660 | tok/s 2628349.85 | norm 0.4206 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6589 | train 3.154682 | val 3.410660 | tok/s 2629860.76 | norm 0.4135 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6590 | train 2.993227 | val 3.410660 | tok/s 2625810.81 | norm 0.4507 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6591 | train 3.124217 | val 3.410660 | tok/s 2627956.40 | norm 0.4530 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6592 | train 3.026498 | val 3.410660 | tok/s 2627066.14 | norm 0.4197 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6593 | train 3.058333 | val 3.410660 | tok/s 2627790.24 | norm 0.4105 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6594 | train 3.067047 | val 3.410660 | tok/s 2483390.26 | norm 0.4414 | dt 211.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6595 | train 2.978430 | val 3.410660 | tok/s 2631941.08 | norm 0.4564 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6596 | train 3.044127 | val 3.410660 | tok/s 2628523.82 | norm 0.4532 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6597 | train 3.044094 | val 3.410660 | tok/s 2629721.37 | norm 0.4284 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6598 | train 3.041228 | val 3.410660 | tok/s 2626952.50 | norm 0.4049 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6599 | train 2.986188 | val 3.410660 | tok/s 2620789.19 | norm 0.3955 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6600 | train 2.968243 | val 3.410660 | tok/s 2627807.12 | norm 0.4506 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6601 | train 3.201169 | val 3.410660 | tok/s 2630436.95 | norm 0.5029 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6602 | train 2.878095 | val 3.410660 | tok/s 2631466.12 | norm 0.4824 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6603 | train 2.969831 | val 3.410660 | tok/s 2627250.76 | norm 0.3811 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6604 | train 3.097176 | val 3.410660 | tok/s 2630226.53 | norm 0.4179 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6605 | train 3.003541 | val 3.410660 | tok/s 2634018.96 | norm 0.4220 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6606 | train 3.087767 | val 3.410660 | tok/s 2634095.18 | norm 0.4124 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6607 | train 3.012723 | val 3.410660 | tok/s 2626227.71 | norm 0.4188 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6608 | train 3.071083 | val 3.410660 | tok/s 2622039.81 | norm 0.4235 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6609 | train 3.033942 | val 3.410660 | tok/s 2627450.28 | norm 0.4000 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6610 | train 3.126816 | val 3.410660 | tok/s 2615693.44 | norm 0.4187 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6611 | train 3.145936 | val 3.410660 | tok/s 2628819.94 | norm 0.4359 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6612 | train 3.056403 | val 3.410660 | tok/s 2625891.31 | norm 0.4303 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6613 | train 3.026367 | val 3.410660 | tok/s 2628873.27 | norm 0.4494 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6614 | train 3.081219 | val 3.410660 | tok/s 2628592.23 | norm 0.4527 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6615 | train 2.987129 | val 3.410660 | tok/s 2632500.31 | norm 0.4659 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6616 | train 3.202228 | val 3.410660 | tok/s 2625544.38 | norm 0.4525 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6617 | train 3.089105 | val 3.410660 | tok/s 2625050.69 | norm 0.4167 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6618 | train 3.033196 | val 3.410660 | tok/s 2627835.27 | norm 0.4376 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6619 | train 2.981649 | val 3.410660 | tok/s 2630208.97 | norm 0.4818 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6620 | train 2.995421 | val 3.410660 | tok/s 2630191.37 | norm 0.3934 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6621 | train 2.946367 | val 3.410660 | tok/s 2632239.24 | norm 0.4447 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6622 | train 3.060696 | val 3.410660 | tok/s 2635374.69 | norm 0.4740 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6623 | train 3.043864 | val 3.410660 | tok/s 2629318.71 | norm 0.4297 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6624 | train 2.967150 | val 3.410660 | tok/s 2632387.30 | norm 0.4134 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6625 | train 3.178164 | val 3.410660 | tok/s 2630386.84 | norm 0.4717 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6626 | train 3.122374 | val 3.410660 | tok/s 2628266.13 | norm 0.4229 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6627 | train 3.084556 | val 3.410660 | tok/s 2618043.15 | norm 0.4330 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6628 | train 2.990891 | val 3.410660 | tok/s 2625992.58 | norm 0.4656 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6629 | train 3.099855 | val 3.410660 | tok/s 2631032.15 | norm 0.4174 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6630 | train 2.998143 | val 3.410660 | tok/s 2630990.51 | norm 0.4462 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6631 | train 2.876562 | val 3.410660 | tok/s 2625763.48 | norm 0.4157 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6632 | train 2.907329 | val 3.410660 | tok/s 2625590.36 | norm 0.4387 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6633 | train 3.048630 | val 3.410660 | tok/s 2626730.54 | norm 0.4188 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6634 | train 3.158792 | val 3.410660 | tok/s 2634092.89 | norm 0.4130 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6635 | train 3.031933 | val 3.410660 | tok/s 2630244.96 | norm 0.3742 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6636 | train 2.968332 | val 3.410660 | tok/s 2631927.80 | norm 0.4255 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6637 | train 3.045296 | val 3.410660 | tok/s 2622398.66 | norm 0.4226 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6638 | train 3.213678 | val 3.410660 | tok/s 2628087.79 | norm 0.5456 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6639 | train 3.047325 | val 3.410660 | tok/s 2630550.58 | norm 0.4162 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6640 | train 3.092028 | val 3.410660 | tok/s 2629924.65 | norm 0.3766 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6641 | train 2.962904 | val 3.410660 | tok/s 2627907.62 | norm 0.3854 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6642 | train 2.954185 | val 3.410660 | tok/s 2624112.17 | norm 0.3870 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6643 | train 3.079688 | val 3.410660 | tok/s 2585496.33 | norm 0.3980 | dt 202.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6644 | train 3.081132 | val 3.410660 | tok/s 2629545.64 | norm 0.4637 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6645 | train 2.979769 | val 3.410660 | tok/s 2629852.98 | norm 0.4232 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6646 | train 3.035306 | val 3.410660 | tok/s 2631792.62 | norm 0.3769 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6647 | train 2.917609 | val 3.410660 | tok/s 2625338.04 | norm 0.4234 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6648 | train 2.975389 | val 3.410660 | tok/s 2621153.23 | norm 0.4462 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6649 | train 3.161541 | val 3.410660 | tok/s 2623699.70 | norm 0.4019 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6650 | train 3.014850 | val 3.410660 | tok/s 2626274.09 | norm 0.4440 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6651 | train 3.105673 | val 3.410660 | tok/s 2477091.56 | norm 0.5315 | dt 211.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6652 | train 3.016111 | val 3.410660 | tok/s 2523042.55 | norm 0.4161 | dt 207.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6653 | train 3.018533 | val 3.410660 | tok/s 2630564.22 | norm 0.3925 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6654 | train 2.924166 | val 3.410660 | tok/s 2631557.54 | norm 0.4301 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6655 | train 3.103582 | val 3.410660 | tok/s 2628740.91 | norm 0.4301 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6656 | train 3.044986 | val 3.410660 | tok/s 2628802.01 | norm 0.4462 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6657 | train 3.016182 | val 3.410660 | tok/s 2629296.50 | norm 0.4339 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6658 | train 3.005517 | val 3.410660 | tok/s 2629522.30 | norm 0.3986 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6659 | train 3.000897 | val 3.410660 | tok/s 2630875.25 | norm 0.3831 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6660 | train 2.988472 | val 3.410660 | tok/s 2627669.09 | norm 0.3941 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6661 | train 2.996190 | val 3.410660 | tok/s 2633527.17 | norm 0.3979 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6662 | train 3.085027 | val 3.410660 | tok/s 2625067.53 | norm 0.4133 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6663 | train 3.020454 | val 3.410660 | tok/s 2629418.76 | norm 0.4487 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6664 | train 3.061607 | val 3.410660 | tok/s 2626967.12 | norm 0.4801 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6665 | train 3.003090 | val 3.410660 | tok/s 2629336.88 | norm 0.4213 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6666 | train 3.038424 | val 3.410660 | tok/s 2627954.44 | norm 0.4117 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6667 | train 3.041862 | val 3.410660 | tok/s 2627746.67 | norm 0.4262 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6668 | train 3.044909 | val 3.410660 | tok/s 2626761.29 | norm 0.4353 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6669 | train 3.054727 | val 3.410660 | tok/s 2619579.56 | norm 0.4290 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6670 | train 3.004847 | val 3.410660 | tok/s 2625993.70 | norm 0.4482 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6671 | train 3.058965 | val 3.410660 | tok/s 2631943.72 | norm 0.4921 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6672 | train 3.095647 | val 3.410660 | tok/s 2629243.48 | norm 0.4329 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6673 | train 3.117302 | val 3.410660 | tok/s 2629111.13 | norm 0.4304 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6674 | train 3.044746 | val 3.410660 | tok/s 2627167.00 | norm 0.3952 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6675 | train 2.947411 | val 3.410660 | tok/s 2626571.36 | norm 0.4008 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6676 | train 3.144403 | val 3.410660 | tok/s 2630952.49 | norm 0.4444 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6677 | train 2.939301 | val 3.410660 | tok/s 2629628.57 | norm 0.4330 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6678 | train 3.051035 | val 3.410660 | tok/s 2632034.93 | norm 0.4231 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6679 | train 2.983565 | val 3.410660 | tok/s 2629890.58 | norm 0.4048 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6680 | train 3.003883 | val 3.410660 | tok/s 2460695.47 | norm 0.4198 | dt 213.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6681 | train 3.060529 | val 3.410660 | tok/s 2627993.94 | norm 0.4100 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6682 | train 3.038356 | val 3.410660 | tok/s 2629930.66 | norm 0.3995 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6683 | train 3.072437 | val 3.410660 | tok/s 2627450.33 | norm 0.4017 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6684 | train 3.005502 | val 3.410660 | tok/s 2625038.39 | norm 0.4515 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6685 | train 2.933311 | val 3.410660 | tok/s 2628713.65 | norm 0.4523 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6686 | train 3.070947 | val 3.410660 | tok/s 2624517.09 | norm 0.4337 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6687 | train 3.002476 | val 3.410660 | tok/s 2628948.96 | norm 0.4818 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6688 | train 3.077713 | val 3.410660 | tok/s 2614879.45 | norm 0.4327 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6689 | train 3.353053 | val 3.410660 | tok/s 2621638.14 | norm 0.4215 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6690 | train 2.995921 | val 3.410660 | tok/s 2632024.33 | norm 0.4031 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6691 | train 2.977632 | val 3.410660 | tok/s 2626139.19 | norm 0.3803 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6692 | train 3.117287 | val 3.410660 | tok/s 2626096.08 | norm 0.4066 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6693 | train 3.088499 | val 3.410660 | tok/s 2621832.02 | norm 0.3704 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6694 | train 3.040703 | val 3.410660 | tok/s 2627726.91 | norm 0.3937 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6695 | train 3.063722 | val 3.410660 | tok/s 2628161.46 | norm 0.4567 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6696 | train 3.030826 | val 3.410660 | tok/s 2630217.71 | norm 0.4646 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6697 | train 3.047688 | val 3.410660 | tok/s 2629617.24 | norm 0.4440 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6698 | train 2.983278 | val 3.410660 | tok/s 2629188.39 | norm 0.4565 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6699 | train 3.053116 | val 3.410660 | tok/s 2624419.87 | norm 0.4584 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6700 | train 2.950367 | val 3.410660 | tok/s 2632473.83 | norm 0.4329 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6701 | train 2.941230 | val 3.410660 | tok/s 2628663.50 | norm 0.4219 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6702 | train 3.078898 | val 3.410660 | tok/s 2630844.06 | norm 0.4426 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6703 | train 3.031148 | val 3.410660 | tok/s 2620530.70 | norm 0.4297 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6704 | train 2.899624 | val 3.410660 | tok/s 2625583.06 | norm 0.4447 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6705 | train 2.984000 | val 3.410660 | tok/s 2631654.36 | norm 0.4185 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6706 | train 3.145683 | val 3.410660 | tok/s 2627026.66 | norm 0.3898 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6707 | train 2.892565 | val 3.410660 | tok/s 2629591.77 | norm 0.4234 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6708 | train 2.998647 | val 3.410660 | tok/s 2622748.53 | norm 0.4014 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6709 | train 2.983161 | val 3.410660 | tok/s 2626930.92 | norm 0.4052 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6710 | train 3.019954 | val 3.410660 | tok/s 2620152.52 | norm 0.4412 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6711 | train 3.034815 | val 3.410660 | tok/s 2628682.10 | norm 0.4558 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6712 | train 2.958451 | val 3.410660 | tok/s 2624827.04 | norm 0.4610 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6713 | train 3.049540 | val 3.410660 | tok/s 2628923.34 | norm 0.4680 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6714 | train 3.179802 | val 3.410660 | tok/s 2625186.35 | norm 0.4500 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6715 | train 3.059946 | val 3.410660 | tok/s 2632377.30 | norm 0.4602 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6716 | train 2.961180 | val 3.410660 | tok/s 2624876.40 | norm 0.4218 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6717 | train 2.959377 | val 3.410660 | tok/s 2629655.81 | norm 0.4287 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6718 | train 3.007081 | val 3.410660 | tok/s 2627171.48 | norm 0.4938 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6719 | train 3.051169 | val 3.410660 | tok/s 2627531.52 | norm 0.4760 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6720 | train 2.995806 | val 3.410660 | tok/s 2632582.34 | norm 0.4312 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6721 | train 2.923905 | val 3.410660 | tok/s 2627718.58 | norm 0.4873 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6722 | train 3.035485 | val 3.410660 | tok/s 2623086.66 | norm 0.4676 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6723 | train 2.975981 | val 3.410660 | tok/s 2629763.75 | norm 0.4371 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6724 | train 3.025615 | val 3.410660 | tok/s 2629620.75 | norm 0.4369 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6725 | train 3.274543 | val 3.410660 | tok/s 2631117.79 | norm 0.4589 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6726 | train 2.969164 | val 3.410660 | tok/s 2630917.67 | norm 0.4151 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6727 | train 3.087161 | val 3.410660 | tok/s 2632088.11 | norm 0.4052 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6728 | train 3.181109 | val 3.410660 | tok/s 2628568.80 | norm 0.4098 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6729 | train 3.000043 | val 3.410660 | tok/s 2629374.89 | norm 0.3875 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6730 | train 3.054174 | val 3.410660 | tok/s 2633356.02 | norm 0.3654 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6731 | train 2.979957 | val 3.410660 | tok/s 2628367.21 | norm 0.4169 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6732 | train 3.043046 | val 3.410660 | tok/s 2627850.39 | norm 0.4371 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6733 | train 3.189957 | val 3.410660 | tok/s 2625052.62 | norm 0.4039 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6734 | train 3.022547 | val 3.410660 | tok/s 2613524.07 | norm 0.4153 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6735 | train 3.115371 | val 3.410660 | tok/s 2630159.90 | norm 0.4166 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6736 | train 3.075106 | val 3.410660 | tok/s 2623219.75 | norm 0.3798 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6737 | train 3.015928 | val 3.410660 | tok/s 2631626.99 | norm 0.3945 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6738 | train 3.144860 | val 3.410660 | tok/s 2630144.50 | norm 0.4281 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6739 | train 3.138246 | val 3.410660 | tok/s 2632021.03 | norm 0.4394 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6740 | train 2.963825 | val 3.410660 | tok/s 2626419.53 | norm 0.4276 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6741 | train 2.980114 | val 3.410660 | tok/s 2625256.01 | norm 0.4352 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6742 | train 2.959236 | val 3.410660 | tok/s 2632360.78 | norm 0.4174 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6743 | train 3.063969 | val 3.410660 | tok/s 2628536.11 | norm 0.4102 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6744 | train 3.140576 | val 3.410660 | tok/s 2630002.99 | norm 0.3953 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6745 | train 3.062686 | val 3.410660 | tok/s 2492830.87 | norm 0.4422 | dt 210.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6746 | train 2.983168 | val 3.410660 | tok/s 2628296.03 | norm 0.4277 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6747 | train 3.059790 | val 3.410660 | tok/s 2630094.75 | norm 0.4230 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6748 | train 2.962489 | val 3.410660 | tok/s 2629820.84 | norm 0.4065 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6749 | train 2.979932 | val 3.410660 | tok/s 2627961.09 | norm 0.3759 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 6750 | val_loss 3.400678 step 6750 | train 3.052624 | val 3.400678 | tok/s 2564726.40 | norm 0.4272 | dt 204.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6751 | train 2.952286 | val 3.400678 | tok/s 2626044.24 | norm 0.4373 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6752 | train 2.990759 | val 3.400678 | tok/s 2623407.92 | norm 0.3912 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6753 | train 3.008966 | val 3.400678 | tok/s 2630085.42 | norm 0.4044 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6754 | train 3.030126 | val 3.400678 | tok/s 2631420.72 | norm 0.4232 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6755 | train 3.090470 | val 3.400678 | tok/s 2630415.60 | norm 0.4009 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6756 | train 2.970581 | val 3.400678 | tok/s 2626681.78 | norm 0.3942 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6757 | train 3.092643 | val 3.400678 | tok/s 2625659.40 | norm 0.4509 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6758 | train 3.059877 | val 3.400678 | tok/s 2607243.20 | norm 0.4440 | dt 201.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6759 | train 3.188737 | val 3.400678 | tok/s 2613928.17 | norm 0.4600 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6760 | train 2.993787 | val 3.400678 | tok/s 2631976.78 | norm 0.4506 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6761 | train 3.032661 | val 3.400678 | tok/s 2616799.63 | norm 0.3956 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6762 | train 3.150349 | val 3.400678 | tok/s 2624809.72 | norm 0.4468 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6763 | train 3.071723 | val 3.400678 | tok/s 2632469.48 | norm 0.4537 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6764 | train 2.942634 | val 3.400678 | tok/s 2633230.33 | norm 0.4013 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6765 | train 3.145973 | val 3.400678 | tok/s 2631356.81 | norm 0.4754 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6766 | train 2.941079 | val 3.400678 | tok/s 2631042.74 | norm 0.4471 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6767 | train 3.002874 | val 3.400678 | tok/s 2628545.35 | norm 0.3818 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6768 | train 3.114356 | val 3.400678 | tok/s 2632867.88 | norm 0.3712 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6769 | train 3.009017 | val 3.400678 | tok/s 2626081.51 | norm 0.3983 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6770 | train 3.055367 | val 3.400678 | tok/s 2627293.95 | norm 0.3792 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6771 | train 3.028725 | val 3.400678 | tok/s 2622839.61 | norm 0.3926 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6772 | train 3.028042 | val 3.400678 | tok/s 2623015.59 | norm 0.4084 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6773 | train 3.100493 | val 3.400678 | tok/s 2631099.08 | norm 0.4272 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6774 | train 2.972541 | val 3.400678 | tok/s 2628769.46 | norm 0.4364 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6775 | train 2.979934 | val 3.400678 | tok/s 2627114.45 | norm 0.4635 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6776 | train 3.079782 | val 3.400678 | tok/s 2628793.00 | norm 0.4024 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6777 | train 3.023835 | val 3.400678 | tok/s 2627151.81 | norm 0.4068 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6778 | train 3.096187 | val 3.400678 | tok/s 2632467.19 | norm 0.4547 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6779 | train 2.992092 | val 3.400678 | tok/s 2627976.79 | norm 0.4448 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6780 | train 3.284101 | val 3.400678 | tok/s 2629408.51 | norm 0.4993 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6781 | train 3.022495 | val 3.400678 | tok/s 2627192.87 | norm 0.4715 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6782 | train 3.001860 | val 3.400678 | tok/s 2629575.93 | norm 0.4013 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6783 | train 3.103310 | val 3.400678 | tok/s 2631297.18 | norm 0.4136 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6784 | train 3.033851 | val 3.400678 | tok/s 2627236.34 | norm 0.4195 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6785 | train 2.995921 | val 3.400678 | tok/s 2626211.13 | norm 0.3957 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6786 | train 3.033892 | val 3.400678 | tok/s 2628490.89 | norm 0.3965 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6787 | train 2.961793 | val 3.400678 | tok/s 2630484.62 | norm 0.4159 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6788 | train 2.919308 | val 3.400678 | tok/s 2631938.26 | norm 0.4238 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6789 | train 2.977415 | val 3.400678 | tok/s 2631729.39 | norm 0.4463 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6790 | train 3.015360 | val 3.400678 | tok/s 2630552.88 | norm 0.4660 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6791 | train 3.059661 | val 3.400678 | tok/s 2628540.11 | norm 0.4377 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6792 | train 3.019870 | val 3.400678 | tok/s 2630959.32 | norm 0.3871 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6793 | train 3.092915 | val 3.400678 | tok/s 2597058.80 | norm 0.4076 | dt 201.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6794 | train 3.085729 | val 3.400678 | tok/s 2544453.95 | norm 0.4091 | dt 206.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6795 | train 3.021792 | val 3.400678 | tok/s 2628237.63 | norm 0.3889 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6796 | train 3.016172 | val 3.400678 | tok/s 2623418.83 | norm 0.4080 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6797 | train 3.032811 | val 3.400678 | tok/s 2628538.00 | norm 0.4513 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6798 | train 3.242810 | val 3.400678 | tok/s 2629506.42 | norm 0.4726 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6799 | train 3.175928 | val 3.400678 | tok/s 2623985.49 | norm 0.4586 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6800 | train 3.118518 | val 3.400678 | tok/s 2631919.98 | norm 0.4731 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6801 | train 3.000113 | val 3.400678 | tok/s 2627604.90 | norm 0.4898 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6802 | train 2.940433 | val 3.400678 | tok/s 2626024.66 | norm 0.4793 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6803 | train 3.054607 | val 3.400678 | tok/s 2629327.89 | norm 0.4702 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6804 | train 3.145241 | val 3.400678 | tok/s 2622668.56 | norm 0.4679 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6805 | train 3.084730 | val 3.400678 | tok/s 2629732.31 | norm 0.4712 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6806 | train 3.064513 | val 3.400678 | tok/s 2625165.85 | norm 0.4216 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6807 | train 3.065433 | val 3.400678 | tok/s 2628746.02 | norm 0.4129 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6808 | train 3.043373 | val 3.400678 | tok/s 2619605.66 | norm 0.4122 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6809 | train 2.993289 | val 3.400678 | tok/s 2619035.71 | norm 0.4694 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6810 | train 3.074748 | val 3.400678 | tok/s 2628935.45 | norm 0.4954 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6811 | train 3.193752 | val 3.400678 | tok/s 2627595.64 | norm 0.4567 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6812 | train 3.066946 | val 3.400678 | tok/s 2631627.77 | norm 0.4326 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6813 | train 3.009345 | val 3.400678 | tok/s 2627467.62 | norm 0.4624 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6814 | train 3.044754 | val 3.400678 | tok/s 2625786.61 | norm 0.4357 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6815 | train 3.026130 | val 3.400678 | tok/s 2624055.95 | norm 0.4446 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6816 | train 2.986667 | val 3.400678 | tok/s 2623038.30 | norm 0.4294 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6817 | train 2.923625 | val 3.400678 | tok/s 2626650.78 | norm 0.4133 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6818 | train 2.909500 | val 3.400678 | tok/s 2613246.25 | norm 0.3933 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6819 | train 3.062181 | val 3.400678 | tok/s 2625244.37 | norm 0.4510 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6820 | train 3.052022 | val 3.400678 | tok/s 2630614.44 | norm 0.4346 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6821 | train 3.050685 | val 3.400678 | tok/s 2625870.76 | norm 0.3996 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6822 | train 3.083836 | val 3.400678 | tok/s 2629863.20 | norm 0.3819 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6823 | train 3.009660 | val 3.400678 | tok/s 2628608.90 | norm 0.4202 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6824 | train 3.083842 | val 3.400678 | tok/s 2628515.09 | norm 0.4287 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6825 | train 3.119874 | val 3.400678 | tok/s 2633699.66 | norm 0.3976 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6826 | train 3.018043 | val 3.400678 | tok/s 2629332.48 | norm 0.3899 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6827 | train 2.965458 | val 3.400678 | tok/s 2629683.20 | norm 0.4019 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6828 | train 3.165053 | val 3.400678 | tok/s 2626182.20 | norm 0.3948 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6829 | train 3.028326 | val 3.400678 | tok/s 2625329.90 | norm 0.4182 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6830 | train 3.086613 | val 3.400678 | tok/s 2630031.27 | norm 0.4420 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6831 | train 3.100335 | val 3.400678 | tok/s 2624044.76 | norm 0.4261 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6832 | train 3.071333 | val 3.400678 | tok/s 2628091.98 | norm 0.3913 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6833 | train 2.982313 | val 3.400678 | tok/s 2622813.37 | norm 0.4272 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6834 | train 3.076228 | val 3.400678 | tok/s 2624511.43 | norm 0.4073 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6835 | train 3.032015 | val 3.400678 | tok/s 2627616.07 | norm 0.3789 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6836 | train 3.045309 | val 3.400678 | tok/s 2629125.39 | norm 0.3875 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6837 | train 3.003546 | val 3.400678 | tok/s 2625523.15 | norm 0.3953 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6838 | train 3.059419 | val 3.400678 | tok/s 2621508.72 | norm 0.4258 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6839 | train 3.102034 | val 3.400678 | tok/s 2626408.65 | norm 0.4521 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6840 | train 3.048829 | val 3.400678 | tok/s 2630945.96 | norm 0.4025 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6841 | train 3.009066 | val 3.400678 | tok/s 2631521.01 | norm 0.3969 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6842 | train 3.021811 | val 3.400678 | tok/s 2622869.14 | norm 0.4070 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6843 | train 3.079131 | val 3.400678 | tok/s 2623792.30 | norm 0.4137 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6844 | train 3.017016 | val 3.400678 | tok/s 2627283.10 | norm 0.3918 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6845 | train 3.037508 | val 3.400678 | tok/s 2629012.69 | norm 0.4295 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6846 | train 2.990175 | val 3.400678 | tok/s 2627686.38 | norm 0.4129 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6847 | train 3.072320 | val 3.400678 | tok/s 2630467.29 | norm 0.4207 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6848 | train 3.010569 | val 3.400678 | tok/s 2629915.88 | norm 0.4002 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6849 | train 3.000993 | val 3.400678 | tok/s 2631546.40 | norm 0.4194 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6850 | train 3.053819 | val 3.400678 | tok/s 2633183.21 | norm 0.4254 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6851 | train 3.060892 | val 3.400678 | tok/s 2628367.92 | norm 0.4222 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6852 | train 3.092101 | val 3.400678 | tok/s 2628049.48 | norm 0.3910 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6853 | train 2.954497 | val 3.400678 | tok/s 2628221.17 | norm 0.4146 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6854 | train 2.981714 | val 3.400678 | tok/s 2629533.15 | norm 0.4023 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6855 | train 3.029525 | val 3.400678 | tok/s 2630196.66 | norm 0.4034 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6856 | train 2.959765 | val 3.400678 | tok/s 2625739.18 | norm 0.3834 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6857 | train 3.044749 | val 3.400678 | tok/s 2614322.04 | norm 0.3881 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6858 | train 2.943720 | val 3.400678 | tok/s 2630306.09 | norm 0.4186 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6859 | train 3.045425 | val 3.400678 | tok/s 2632827.89 | norm 0.4223 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6860 | train 2.928203 | val 3.400678 | tok/s 2627195.53 | norm 0.4332 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6861 | train 3.022580 | val 3.400678 | tok/s 2626200.08 | norm 0.4411 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6862 | train 3.066821 | val 3.400678 | tok/s 2631371.12 | norm 0.4674 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6863 | train 3.042306 | val 3.400678 | tok/s 2628758.11 | norm 0.3960 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6864 | train 3.102190 | val 3.400678 | tok/s 2630129.38 | norm 0.3898 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6865 | train 2.962591 | val 3.400678 | tok/s 2618895.27 | norm 0.4251 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6866 | train 3.015114 | val 3.400678 | tok/s 2623102.19 | norm 0.4501 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6867 | train 2.985226 | val 3.400678 | tok/s 2631855.71 | norm 0.4231 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6868 | train 3.209142 | val 3.400678 | tok/s 2627778.75 | norm 0.4383 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6869 | train 3.019822 | val 3.400678 | tok/s 2629400.33 | norm 0.4377 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6870 | train 3.037528 | val 3.400678 | tok/s 2626955.92 | norm 0.4021 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6871 | train 3.195172 | val 3.400678 | tok/s 2628219.31 | norm 0.4491 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6872 | train 3.115066 | val 3.400678 | tok/s 2631229.87 | norm 0.3968 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6873 | train 2.959262 | val 3.400678 | tok/s 2627277.82 | norm 0.4054 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6874 | train 3.031071 | val 3.400678 | tok/s 2626641.21 | norm 0.4974 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6875 | train 3.002600 | val 3.400678 | tok/s 2622590.57 | norm 0.4441 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6876 | train 2.968271 | val 3.400678 | tok/s 2627581.47 | norm 0.3772 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6877 | train 3.086945 | val 3.400678 | tok/s 2628248.41 | norm 0.3995 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6878 | train 3.068950 | val 3.400678 | tok/s 2630359.95 | norm 0.4108 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6879 | train 3.134256 | val 3.400678 | tok/s 2625260.77 | norm 0.4371 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6880 | train 3.008833 | val 3.400678 | tok/s 2607708.97 | norm 0.4535 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6881 | train 3.130765 | val 3.400678 | tok/s 2629489.71 | norm 0.4533 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6882 | train 3.056021 | val 3.400678 | tok/s 2616641.98 | norm 0.4127 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6883 | train 3.066585 | val 3.400678 | tok/s 2628195.65 | norm 0.4175 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6884 | train 2.905896 | val 3.400678 | tok/s 2630029.69 | norm 0.3996 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6885 | train 2.921896 | val 3.400678 | tok/s 2627583.11 | norm 0.4446 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6886 | train 2.974529 | val 3.400678 | tok/s 2626892.19 | norm 0.4637 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6887 | train 2.987539 | val 3.400678 | tok/s 2633560.53 | norm 0.4283 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6888 | train 2.946556 | val 3.400678 | tok/s 2629992.99 | norm 0.4031 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6889 | train 2.984190 | val 3.400678 | tok/s 2628751.95 | norm 0.4609 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6890 | train 3.089594 | val 3.400678 | tok/s 2627626.69 | norm 0.4773 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6891 | train 3.071796 | val 3.400678 | tok/s 2627542.63 | norm 0.4656 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6892 | train 3.323014 | val 3.400678 | tok/s 2612369.59 | norm 0.7011 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6893 | train 2.969842 | val 3.400678 | tok/s 2627348.17 | norm 0.4191 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6894 | train 2.977880 | val 3.400678 | tok/s 2629761.45 | norm 0.3560 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6895 | train 3.002621 | val 3.400678 | tok/s 2631991.92 | norm 0.4237 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6896 | train 3.119612 | val 3.400678 | tok/s 2627792.69 | norm 0.4182 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6897 | train 3.039783 | val 3.400678 | tok/s 2633359.75 | norm 0.5496 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6898 | train 2.944920 | val 3.400678 | tok/s 2187261.51 | norm 0.3914 | dt 239.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6899 | train 2.937922 | val 3.400678 | tok/s 2621921.70 | norm 0.3912 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6900 | train 3.015485 | val 3.400678 | tok/s 2628088.46 | norm 0.4306 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6901 | train 2.997269 | val 3.400678 | tok/s 2631317.68 | norm 0.4292 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6902 | train 3.094306 | val 3.400678 | tok/s 2632019.71 | norm 0.4126 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6903 | train 2.939829 | val 3.400678 | tok/s 2623532.15 | norm 0.3848 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6904 | train 3.038636 | val 3.400678 | tok/s 2629926.80 | norm 0.4134 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6905 | train 2.992492 | val 3.400678 | tok/s 2629758.19 | norm 0.4755 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6906 | train 3.147194 | val 3.400678 | tok/s 2208667.54 | norm 0.3888 | dt 237.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6907 | train 3.042469 | val 3.400678 | tok/s 2627639.82 | norm 0.3870 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6908 | train 3.001374 | val 3.400678 | tok/s 2624520.37 | norm 0.4317 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6909 | train 3.182670 | val 3.400678 | tok/s 2629443.75 | norm 0.4743 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6910 | train 3.088183 | val 3.400678 | tok/s 2630555.57 | norm 0.4168 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6911 | train 3.075582 | val 3.400678 | tok/s 2631781.07 | norm 0.4800 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6912 | train 3.154762 | val 3.400678 | tok/s 2625233.67 | norm 0.5022 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6913 | train 3.092521 | val 3.400678 | tok/s 2629766.18 | norm 0.4764 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6914 | train 3.092323 | val 3.400678 | tok/s 2633230.66 | norm 0.4232 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6915 | train 3.039970 | val 3.400678 | tok/s 2631905.79 | norm 0.4315 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6916 | train 3.066927 | val 3.400678 | tok/s 2631392.89 | norm 0.4342 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6917 | train 3.074325 | val 3.400678 | tok/s 2629735.70 | norm 0.4239 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6918 | train 2.998965 | val 3.400678 | tok/s 2626913.07 | norm 0.3613 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6919 | train 3.057613 | val 3.400678 | tok/s 2627216.87 | norm 0.4037 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6920 | train 3.075875 | val 3.400678 | tok/s 2626403.21 | norm 0.4241 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6921 | train 3.038418 | val 3.400678 | tok/s 2632159.36 | norm 0.3982 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6922 | train 3.070282 | val 3.400678 | tok/s 2632478.20 | norm 0.4029 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6923 | train 3.134926 | val 3.400678 | tok/s 2630575.17 | norm 0.4458 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6924 | train 3.141941 | val 3.400678 | tok/s 2635255.01 | norm 0.4733 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6925 | train 3.040029 | val 3.400678 | tok/s 2629119.95 | norm 0.4088 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6926 | train 2.936011 | val 3.400678 | tok/s 2630437.66 | norm 0.4470 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6927 | train 2.921748 | val 3.400678 | tok/s 2625344.42 | norm 0.4451 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6928 | train 3.031589 | val 3.400678 | tok/s 2626123.91 | norm 0.3862 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6929 | train 3.042286 | val 3.400678 | tok/s 2504596.13 | norm 0.3914 | dt 209.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6930 | train 3.047187 | val 3.400678 | tok/s 2621643.10 | norm 0.4061 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6931 | train 2.998432 | val 3.400678 | tok/s 2628200.88 | norm 0.3916 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6932 | train 2.975546 | val 3.400678 | tok/s 2626629.25 | norm 0.3938 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6933 | train 2.979663 | val 3.400678 | tok/s 2632446.94 | norm 0.4246 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6934 | train 3.023474 | val 3.400678 | tok/s 2620468.25 | norm 0.4432 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6935 | train 3.072809 | val 3.400678 | tok/s 2625997.70 | norm 0.4296 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6936 | train 3.084211 | val 3.400678 | tok/s 2586311.32 | norm 0.4259 | dt 202.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6937 | train 3.073100 | val 3.400678 | tok/s 2593984.25 | norm 0.4556 | dt 202.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6938 | train 2.995022 | val 3.400678 | tok/s 2632383.89 | norm 0.5156 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6939 | train 2.991223 | val 3.400678 | tok/s 2623481.76 | norm 0.4381 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6940 | train 3.026094 | val 3.400678 | tok/s 2627020.45 | norm 0.4058 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6941 | train 3.024295 | val 3.400678 | tok/s 2629555.40 | norm 0.4114 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6942 | train 3.102448 | val 3.400678 | tok/s 2623747.11 | norm 0.4158 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6943 | train 3.085761 | val 3.400678 | tok/s 2630487.43 | norm 0.4144 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6944 | train 3.044464 | val 3.400678 | tok/s 2627170.54 | norm 0.4252 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6945 | train 3.009853 | val 3.400678 | tok/s 2631765.05 | norm 0.4427 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6946 | train 3.030576 | val 3.400678 | tok/s 2625077.85 | norm 0.4539 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6947 | train 2.956963 | val 3.400678 | tok/s 2623551.26 | norm 0.4279 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6948 | train 3.093051 | val 3.400678 | tok/s 2627600.87 | norm 0.4432 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6949 | train 3.053401 | val 3.400678 | tok/s 2625570.15 | norm 0.4390 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6950 | train 3.078898 | val 3.400678 | tok/s 2619315.53 | norm 0.4131 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6951 | train 3.113365 | val 3.400678 | tok/s 2626256.82 | norm 0.4132 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6952 | train 3.021331 | val 3.400678 | tok/s 2626427.61 | norm 0.4365 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6953 | train 3.004699 | val 3.400678 | tok/s 2629423.65 | norm 0.4486 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6954 | train 3.121483 | val 3.400678 | tok/s 2624012.83 | norm 0.4330 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6955 | train 2.999474 | val 3.400678 | tok/s 2627579.27 | norm 0.4543 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6956 | train 3.024986 | val 3.400678 | tok/s 2627039.26 | norm 0.4371 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6957 | train 3.118839 | val 3.400678 | tok/s 2624797.42 | norm 0.4344 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6958 | train 3.063036 | val 3.400678 | tok/s 2630326.28 | norm 0.4405 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6959 | train 3.135243 | val 3.400678 | tok/s 2629475.27 | norm 0.4465 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6960 | train 2.972112 | val 3.400678 | tok/s 2627244.79 | norm 0.4477 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6961 | train 2.995795 | val 3.400678 | tok/s 2630358.30 | norm 0.4393 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6962 | train 3.081970 | val 3.400678 | tok/s 2628064.56 | norm 0.4150 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6963 | train 2.992245 | val 3.400678 | tok/s 2631739.87 | norm 0.4039 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6964 | train 3.105096 | val 3.400678 | tok/s 2631850.97 | norm 0.4371 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6965 | train 3.042929 | val 3.400678 | tok/s 2627858.12 | norm 0.4099 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6966 | train 3.037388 | val 3.400678 | tok/s 2630902.58 | norm 0.4024 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6967 | train 3.112695 | val 3.400678 | tok/s 2632787.70 | norm 0.4556 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6968 | train 3.028684 | val 3.400678 | tok/s 2628940.05 | norm 0.4351 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6969 | train 3.123772 | val 3.400678 | tok/s 2618289.44 | norm 0.4053 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6970 | train 3.029098 | val 3.400678 | tok/s 2631009.55 | norm 0.4504 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6971 | train 2.993534 | val 3.400678 | tok/s 2623606.02 | norm 0.4348 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6972 | train 3.094837 | val 3.400678 | tok/s 2620839.84 | norm 0.3900 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6973 | train 3.044674 | val 3.400678 | tok/s 2630249.19 | norm 0.4062 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6974 | train 2.948526 | val 3.400678 | tok/s 2624119.41 | norm 0.4211 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6975 | train 3.092365 | val 3.400678 | tok/s 2625390.24 | norm 0.4483 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6976 | train 3.227018 | val 3.400678 | tok/s 2631028.57 | norm 0.4586 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6977 | train 2.965937 | val 3.400678 | tok/s 2624469.77 | norm 0.4085 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6978 | train 2.960742 | val 3.400678 | tok/s 2628204.05 | norm 0.4294 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6979 | train 3.103186 | val 3.400678 | tok/s 2627082.86 | norm 0.4241 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6980 | train 2.987149 | val 3.400678 | tok/s 2629158.70 | norm 0.4116 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6981 | train 3.094594 | val 3.400678 | tok/s 2521058.58 | norm 0.4403 | dt 207.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6982 | train 2.996286 | val 3.400678 | tok/s 2455105.96 | norm 0.4378 | dt 213.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6983 | train 2.937397 | val 3.400678 | tok/s 2630461.11 | norm 0.4535 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6984 | train 2.925133 | val 3.400678 | tok/s 2628767.73 | norm 0.4540 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6985 | train 3.085595 | val 3.400678 | tok/s 2630760.36 | norm 0.4302 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6986 | train 3.136669 | val 3.400678 | tok/s 2626377.17 | norm 0.4505 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6987 | train 3.068984 | val 3.400678 | tok/s 2626467.33 | norm 0.4457 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6988 | train 3.019139 | val 3.400678 | tok/s 2630765.97 | norm 0.5041 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6989 | train 2.963615 | val 3.400678 | tok/s 2626886.03 | norm 0.4203 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6990 | train 3.077312 | val 3.400678 | tok/s 2632658.06 | norm 0.4166 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6991 | train 2.942155 | val 3.400678 | tok/s 2625078.74 | norm 0.4921 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6992 | train 3.112593 | val 3.400678 | tok/s 2629045.70 | norm 0.4579 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6993 | train 2.897268 | val 3.400678 | tok/s 2627056.74 | norm 0.4580 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6994 | train 3.080388 | val 3.400678 | tok/s 2624356.15 | norm 0.4330 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6995 | train 3.052547 | val 3.400678 | tok/s 2628679.71 | norm 0.4409 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6996 | train 2.866996 | val 3.400678 | tok/s 2625773.89 | norm 0.4491 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6997 | train 3.259410 | val 3.400678 | tok/s 2625567.21 | norm 0.4979 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6998 | train 3.103473 | val 3.400678 | tok/s 2628976.11 | norm 0.4284 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 6999 | train 2.989761 | val 3.400678 | tok/s 2625941.55 | norm 0.3660 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 7000 | val_loss 3.398238 step 7000 | train 2.972759 | val 3.398238 | tok/s 2607040.64 | norm 0.3861 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7001 | train 3.124030 | val 3.398238 | tok/s 2629681.07 | norm 0.4264 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7002 | train 2.938648 | val 3.398238 | tok/s 2628255.99 | norm 0.4392 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7003 | train 3.067863 | val 3.398238 | tok/s 2627758.92 | norm 0.4293 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7004 | train 3.050294 | val 3.398238 | tok/s 2630498.83 | norm 0.3870 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7005 | train 2.945777 | val 3.398238 | tok/s 2627535.73 | norm 0.3943 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7006 | train 2.956692 | val 3.398238 | tok/s 2630329.44 | norm 0.4416 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7007 | train 3.020687 | val 3.398238 | tok/s 2627575.82 | norm 0.4438 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7008 | train 2.977848 | val 3.398238 | tok/s 2610915.15 | norm 0.4072 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7009 | train 3.166028 | val 3.398238 | tok/s 2626573.68 | norm 0.4301 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7010 | train 3.013301 | val 3.398238 | tok/s 2630958.80 | norm 0.4087 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7011 | train 3.038929 | val 3.398238 | tok/s 2628526.22 | norm 0.4154 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7012 | train 2.934871 | val 3.398238 | tok/s 2628168.97 | norm 0.4141 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7013 | train 3.103042 | val 3.398238 | tok/s 2627582.53 | norm 0.3656 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7014 | train 3.098877 | val 3.398238 | tok/s 2628029.43 | norm 0.4119 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7015 | train 3.017245 | val 3.398238 | tok/s 2625187.93 | norm 0.4145 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7016 | train 2.952564 | val 3.398238 | tok/s 2629611.85 | norm 0.4062 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7017 | train 2.960603 | val 3.398238 | tok/s 2624975.05 | norm 0.4342 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7018 | train 3.015029 | val 3.398238 | tok/s 2633875.76 | norm 0.4220 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7019 | train 3.052518 | val 3.398238 | tok/s 2629136.65 | norm 0.4177 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7020 | train 3.120576 | val 3.398238 | tok/s 2629705.66 | norm 0.4176 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7021 | train 2.953569 | val 3.398238 | tok/s 2628535.57 | norm 0.3795 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7022 | train 3.085568 | val 3.398238 | tok/s 2632035.57 | norm 0.4136 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7023 | train 2.960342 | val 3.398238 | tok/s 2630892.73 | norm 0.4521 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7024 | train 2.965725 | val 3.398238 | tok/s 2629498.94 | norm 0.4353 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7025 | train 3.002317 | val 3.398238 | tok/s 2629754.01 | norm 0.4376 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7026 | train 3.017204 | val 3.398238 | tok/s 2632070.76 | norm 0.4524 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7027 | train 3.056693 | val 3.398238 | tok/s 2627583.66 | norm 0.4632 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7028 | train 3.076138 | val 3.398238 | tok/s 2628408.05 | norm 0.4394 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7029 | train 3.119994 | val 3.398238 | tok/s 2631405.29 | norm 0.4174 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7030 | train 2.985035 | val 3.398238 | tok/s 2413059.08 | norm 0.4125 | dt 217.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7031 | train 3.023239 | val 3.398238 | tok/s 2536114.16 | norm 0.4064 | dt 206.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7032 | train 2.991973 | val 3.398238 | tok/s 2607460.30 | norm 0.3949 | dt 201.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7033 | train 3.061687 | val 3.398238 | tok/s 2630227.95 | norm 0.4145 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7034 | train 3.083571 | val 3.398238 | tok/s 2627777.75 | norm 0.4075 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7035 | train 3.152528 | val 3.398238 | tok/s 2627826.31 | norm 0.4027 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7036 | train 3.097932 | val 3.398238 | tok/s 2629335.12 | norm 0.4436 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7037 | train 2.980933 | val 3.398238 | tok/s 2630982.64 | norm 0.5170 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7038 | train 3.169425 | val 3.398238 | tok/s 2507418.04 | norm 0.4784 | dt 209.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7039 | train 3.061248 | val 3.398238 | tok/s 2624544.02 | norm 0.4461 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7040 | train 3.020910 | val 3.398238 | tok/s 2625987.99 | norm 0.4483 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7041 | train 3.084532 | val 3.398238 | tok/s 2630794.76 | norm 0.4267 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7042 | train 3.103727 | val 3.398238 | tok/s 2628140.29 | norm 0.4585 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7043 | train 2.966175 | val 3.398238 | tok/s 2631995.18 | norm 0.4186 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7044 | train 3.073204 | val 3.398238 | tok/s 2625468.28 | norm 0.3907 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7045 | train 2.997790 | val 3.398238 | tok/s 2627400.88 | norm 0.4078 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7046 | train 3.088521 | val 3.398238 | tok/s 2621227.44 | norm 0.4034 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7047 | train 2.996085 | val 3.398238 | tok/s 2628889.88 | norm 0.4387 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7048 | train 3.011231 | val 3.398238 | tok/s 2631720.71 | norm 0.4359 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7049 | train 3.049974 | val 3.398238 | tok/s 2627636.77 | norm 0.4292 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7050 | train 2.917526 | val 3.398238 | tok/s 2627126.02 | norm 0.4353 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7051 | train 3.057827 | val 3.398238 | tok/s 2632257.72 | norm 0.4331 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7052 | train 3.149813 | val 3.398238 | tok/s 2624817.13 | norm 0.4140 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7053 | train 3.034676 | val 3.398238 | tok/s 2625245.75 | norm 0.3949 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7054 | train 2.981801 | val 3.398238 | tok/s 2622797.25 | norm 0.4014 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7055 | train 3.070335 | val 3.398238 | tok/s 2627044.80 | norm 0.4132 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7056 | train 3.056759 | val 3.398238 | tok/s 2625706.59 | norm 0.6555 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7057 | train 3.025040 | val 3.398238 | tok/s 2627798.18 | norm 0.3872 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7058 | train 3.097048 | val 3.398238 | tok/s 2633170.50 | norm 0.3834 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7059 | train 3.121765 | val 3.398238 | tok/s 2627206.03 | norm 0.4962 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7060 | train 2.994506 | val 3.398238 | tok/s 2625853.77 | norm 0.3341 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7061 | train 2.966556 | val 3.398238 | tok/s 2630566.83 | norm 0.3907 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7062 | train 3.182485 | val 3.398238 | tok/s 2627159.97 | norm 0.3996 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7063 | train 3.092205 | val 3.398238 | tok/s 2628242.01 | norm 0.4549 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7064 | train 3.045485 | val 3.398238 | tok/s 2622295.42 | norm 0.4442 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7065 | train 3.030791 | val 3.398238 | tok/s 2628460.21 | norm 0.4435 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7066 | train 3.111171 | val 3.398238 | tok/s 2627606.97 | norm 0.4338 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7067 | train 2.949472 | val 3.398238 | tok/s 2627159.10 | norm 0.4399 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7068 | train 3.006403 | val 3.398238 | tok/s 2629279.45 | norm 0.4152 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7069 | train 3.105193 | val 3.398238 | tok/s 2628369.79 | norm 0.4366 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7070 | train 3.038172 | val 3.398238 | tok/s 2628338.13 | norm 0.3943 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7071 | train 2.978964 | val 3.398238 | tok/s 2630246.68 | norm 0.3953 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7072 | train 3.014698 | val 3.398238 | tok/s 2623337.11 | norm 0.4100 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7073 | train 2.990151 | val 3.398238 | tok/s 2628318.20 | norm 0.4378 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7074 | train 3.141378 | val 3.398238 | tok/s 2627905.11 | norm 0.4386 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7075 | train 2.964230 | val 3.398238 | tok/s 2629255.58 | norm 0.3961 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7076 | train 3.026603 | val 3.398238 | tok/s 2624894.97 | norm 0.3729 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7077 | train 3.032080 | val 3.398238 | tok/s 2622889.53 | norm 0.4156 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7078 | train 3.036229 | val 3.398238 | tok/s 2631120.99 | norm 0.4315 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7079 | train 3.053377 | val 3.398238 | tok/s 2625131.41 | norm 0.4396 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7080 | train 3.050543 | val 3.398238 | tok/s 2627846.04 | norm 0.4182 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7081 | train 3.028785 | val 3.398238 | tok/s 2627485.74 | norm 0.4406 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7082 | train 3.097420 | val 3.398238 | tok/s 2625453.37 | norm 0.4506 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7083 | train 2.995242 | val 3.398238 | tok/s 2631564.22 | norm 0.3982 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7084 | train 2.945334 | val 3.398238 | tok/s 2632682.37 | norm 0.4359 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7085 | train 3.165524 | val 3.398238 | tok/s 2628905.82 | norm 0.4264 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7086 | train 3.035113 | val 3.398238 | tok/s 2620494.94 | norm 0.3690 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7087 | train 2.965356 | val 3.398238 | tok/s 2624699.71 | norm 0.3579 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7088 | train 2.958643 | val 3.398238 | tok/s 2628862.46 | norm 0.3999 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7089 | train 2.951119 | val 3.398238 | tok/s 2624085.71 | norm 0.4236 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7090 | train 3.023261 | val 3.398238 | tok/s 2484489.98 | norm 0.3962 | dt 211.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7091 | train 3.081189 | val 3.398238 | tok/s 2625745.63 | norm 0.4155 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7092 | train 3.014597 | val 3.398238 | tok/s 2627956.02 | norm 0.3587 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7093 | train 2.973698 | val 3.398238 | tok/s 2630994.44 | norm 0.4064 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7094 | train 3.028019 | val 3.398238 | tok/s 2628548.54 | norm 0.4489 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7095 | train 3.112875 | val 3.398238 | tok/s 2627742.02 | norm 0.4728 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7096 | train 3.038300 | val 3.398238 | tok/s 2623375.32 | norm 0.4199 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7097 | train 3.141481 | val 3.398238 | tok/s 2624718.22 | norm 0.3782 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7098 | train 3.025582 | val 3.398238 | tok/s 2626597.76 | norm 0.3586 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7099 | train 2.979646 | val 3.398238 | tok/s 2623015.68 | norm 0.3961 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7100 | train 2.977375 | val 3.398238 | tok/s 2626857.32 | norm 0.4294 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7101 | train 3.062484 | val 3.398238 | tok/s 2625431.35 | norm 0.4323 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7102 | train 3.048530 | val 3.398238 | tok/s 2626791.68 | norm 0.4597 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7103 | train 2.982250 | val 3.398238 | tok/s 2626752.46 | norm 0.4298 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7104 | train 3.095487 | val 3.398238 | tok/s 2620657.76 | norm 0.4690 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7105 | train 2.998397 | val 3.398238 | tok/s 2603523.59 | norm 0.4831 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7106 | train 3.044169 | val 3.398238 | tok/s 2621317.73 | norm 0.4205 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7107 | train 3.019519 | val 3.398238 | tok/s 2628592.06 | norm 0.4472 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7108 | train 3.072774 | val 3.398238 | tok/s 2627989.11 | norm 0.4575 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7109 | train 3.020506 | val 3.398238 | tok/s 2628492.90 | norm 0.4238 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7110 | train 2.965569 | val 3.398238 | tok/s 2631480.65 | norm 0.3886 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7111 | train 3.027366 | val 3.398238 | tok/s 2628850.23 | norm 0.4243 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7112 | train 2.958988 | val 3.398238 | tok/s 2624439.31 | norm 0.4736 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7113 | train 3.061267 | val 3.398238 | tok/s 2628669.16 | norm 0.4476 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7114 | train 3.038087 | val 3.398238 | tok/s 2625634.69 | norm 0.4274 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7115 | train 3.093204 | val 3.398238 | tok/s 2626756.96 | norm 0.4319 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7116 | train 2.950700 | val 3.398238 | tok/s 2625572.30 | norm 0.4588 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7117 | train 2.962075 | val 3.398238 | tok/s 2627268.05 | norm 0.4180 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7118 | train 2.896096 | val 3.398238 | tok/s 2629327.26 | norm 0.4488 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7119 | train 3.009657 | val 3.398238 | tok/s 2625754.68 | norm 0.5024 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7120 | train 2.992451 | val 3.398238 | tok/s 2627897.28 | norm 0.4546 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7121 | train 3.025040 | val 3.398238 | tok/s 2625685.00 | norm 0.4156 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7122 | train 3.021958 | val 3.398238 | tok/s 2627422.86 | norm 0.4198 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7123 | train 3.007851 | val 3.398238 | tok/s 2629352.50 | norm 0.4113 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7124 | train 3.018469 | val 3.398238 | tok/s 2627018.78 | norm 0.4372 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7125 | train 3.158200 | val 3.398238 | tok/s 2630197.91 | norm 0.4432 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7126 | train 3.111062 | val 3.398238 | tok/s 2628496.51 | norm 0.4295 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7127 | train 3.005653 | val 3.398238 | tok/s 2625126.28 | norm 0.4338 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7128 | train 3.084202 | val 3.398238 | tok/s 2625898.99 | norm 0.4425 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7129 | train 3.009188 | val 3.398238 | tok/s 2618179.45 | norm 0.4367 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7130 | train 2.995939 | val 3.398238 | tok/s 2572779.00 | norm 0.3977 | dt 203.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7131 | train 2.977392 | val 3.398238 | tok/s 2545279.35 | norm 0.4386 | dt 205.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7132 | train 3.037398 | val 3.398238 | tok/s 2627392.78 | norm 0.4680 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7133 | train 3.025528 | val 3.398238 | tok/s 2617729.05 | norm 0.4138 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7134 | train 2.949160 | val 3.398238 | tok/s 2625191.60 | norm 0.4098 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7135 | train 2.973499 | val 3.398238 | tok/s 2631323.58 | norm 0.4569 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7136 | train 3.136055 | val 3.398238 | tok/s 2627484.41 | norm 0.4937 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7137 | train 3.056065 | val 3.398238 | tok/s 2486076.89 | norm 0.4339 | dt 210.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7138 | train 3.238933 | val 3.398238 | tok/s 2540232.44 | norm 0.4677 | dt 206.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7139 | train 3.067129 | val 3.398238 | tok/s 2627014.26 | norm 0.3921 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7140 | train 3.033557 | val 3.398238 | tok/s 2631153.30 | norm 0.3923 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7141 | train 2.973735 | val 3.398238 | tok/s 2629251.39 | norm 0.3829 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7142 | train 3.085007 | val 3.398238 | tok/s 2629117.11 | norm 0.3968 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7143 | train 3.110199 | val 3.398238 | tok/s 2624973.82 | norm 0.4239 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7144 | train 2.952533 | val 3.398238 | tok/s 2627443.56 | norm 0.3903 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7145 | train 3.083615 | val 3.398238 | tok/s 2592713.22 | norm 0.3995 | dt 202.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7146 | train 3.067646 | val 3.398238 | tok/s 2626721.29 | norm 0.4095 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7147 | train 3.055783 | val 3.398238 | tok/s 2632551.34 | norm 0.4271 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7148 | train 2.956795 | val 3.398238 | tok/s 2631897.24 | norm 0.4748 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7149 | train 3.001520 | val 3.398238 | tok/s 2630359.52 | norm 0.4942 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7150 | train 3.044369 | val 3.398238 | tok/s 2629921.44 | norm 0.5155 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7151 | train 2.892895 | val 3.398238 | tok/s 2624148.53 | norm 0.3819 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7152 | train 3.035546 | val 3.398238 | tok/s 2628536.38 | norm 0.3700 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7153 | train 3.093212 | val 3.398238 | tok/s 2627398.36 | norm 0.4322 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7154 | train 3.049371 | val 3.398238 | tok/s 2631065.32 | norm 0.4942 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7155 | train 2.987849 | val 3.398238 | tok/s 2629557.30 | norm 0.5103 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7156 | train 3.049882 | val 3.398238 | tok/s 2629555.81 | norm 0.4965 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7157 | train 3.082114 | val 3.398238 | tok/s 2630998.25 | norm 0.4075 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7158 | train 2.968011 | val 3.398238 | tok/s 2624947.97 | norm 0.4188 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7159 | train 3.111828 | val 3.398238 | tok/s 2612858.67 | norm 0.5173 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7160 | train 3.063337 | val 3.398238 | tok/s 2629074.46 | norm 0.4184 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7161 | train 3.119068 | val 3.398238 | tok/s 2522709.22 | norm 0.3958 | dt 207.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7162 | train 2.903181 | val 3.398238 | tok/s 2627324.59 | norm 0.4056 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7163 | train 2.981614 | val 3.398238 | tok/s 2627773.20 | norm 0.4258 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7164 | train 3.128374 | val 3.398238 | tok/s 2630418.99 | norm 0.4489 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7165 | train 3.042392 | val 3.398238 | tok/s 2626496.11 | norm 0.4138 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7166 | train 3.031355 | val 3.398238 | tok/s 2628591.64 | norm 0.4153 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7167 | train 3.081281 | val 3.398238 | tok/s 2630253.60 | norm 0.4244 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7168 | train 3.035641 | val 3.398238 | tok/s 2630670.95 | norm 0.4170 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7169 | train 2.986347 | val 3.398238 | tok/s 2630272.13 | norm 0.3716 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7170 | train 3.000841 | val 3.398238 | tok/s 2628796.82 | norm 0.3956 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7171 | train 3.027665 | val 3.398238 | tok/s 2621390.51 | norm 0.4362 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7172 | train 3.060823 | val 3.398238 | tok/s 2630056.59 | norm 0.4132 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7173 | train 3.156191 | val 3.398238 | tok/s 2627852.47 | norm 0.3590 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7174 | train 3.113955 | val 3.398238 | tok/s 2627967.21 | norm 0.3935 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7175 | train 3.040920 | val 3.398238 | tok/s 2627645.86 | norm 0.4381 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7176 | train 3.085904 | val 3.398238 | tok/s 2594021.68 | norm 0.4427 | dt 202.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7177 | train 3.088140 | val 3.398238 | tok/s 2628851.24 | norm 0.4576 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7178 | train 2.981292 | val 3.398238 | tok/s 2541189.71 | norm 0.4618 | dt 206.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7179 | train 3.009878 | val 3.398238 | tok/s 2629644.12 | norm 0.4637 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7180 | train 2.959893 | val 3.398238 | tok/s 2625066.32 | norm 0.4681 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7181 | train 3.040307 | val 3.398238 | tok/s 2628161.79 | norm 0.4172 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7182 | train 2.987317 | val 3.398238 | tok/s 2632739.97 | norm 0.4362 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7183 | train 2.997423 | val 3.398238 | tok/s 2627169.55 | norm 0.4469 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7184 | train 3.074397 | val 3.398238 | tok/s 2627042.81 | norm 0.4343 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7185 | train 3.159899 | val 3.398238 | tok/s 2629890.38 | norm 0.3764 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7186 | train 2.955493 | val 3.398238 | tok/s 2630508.31 | norm 0.3945 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7187 | train 2.923269 | val 3.398238 | tok/s 2628301.79 | norm 0.4232 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7188 | train 2.999016 | val 3.398238 | tok/s 2626286.73 | norm 0.4218 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7189 | train 2.976000 | val 3.398238 | tok/s 2601773.53 | norm 0.4190 | dt 201.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7190 | train 3.122087 | val 3.398238 | tok/s 2622328.83 | norm 0.4151 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7191 | train 3.021320 | val 3.398238 | tok/s 2623263.63 | norm 0.4031 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7192 | train 3.147814 | val 3.398238 | tok/s 2628928.03 | norm 0.4450 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7193 | train 2.999488 | val 3.398238 | tok/s 2594138.24 | norm 0.4336 | dt 202.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7194 | train 3.119344 | val 3.398238 | tok/s 2610748.06 | norm 0.3830 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7195 | train 3.045470 | val 3.398238 | tok/s 2626617.57 | norm 0.4310 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7196 | train 2.883474 | val 3.398238 | tok/s 2628830.06 | norm 0.4048 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7197 | train 3.174830 | val 3.398238 | tok/s 2627535.44 | norm 0.3584 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7198 | train 3.041113 | val 3.398238 | tok/s 2626880.98 | norm 0.3822 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7199 | train 3.061116 | val 3.398238 | tok/s 2626258.66 | norm 0.3897 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7200 | train 3.052211 | val 3.398238 | tok/s 2628474.02 | norm 0.4534 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7201 | train 3.002653 | val 3.398238 | tok/s 2627445.53 | norm 0.4652 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7202 | train 2.940972 | val 3.398238 | tok/s 2631981.20 | norm 0.4026 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7203 | train 3.113870 | val 3.398238 | tok/s 2622884.30 | norm 0.4065 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7204 | train 3.085967 | val 3.398238 | tok/s 2631410.23 | norm 0.4481 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7205 | train 3.042890 | val 3.398238 | tok/s 2628173.42 | norm 0.3754 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7206 | train 2.918672 | val 3.398238 | tok/s 2628057.98 | norm 0.4094 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7207 | train 2.973444 | val 3.398238 | tok/s 2626078.48 | norm 0.4404 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7208 | train 3.089794 | val 3.398238 | tok/s 2616947.46 | norm 0.4896 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7209 | train 3.034244 | val 3.398238 | tok/s 2631251.73 | norm 0.4567 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7210 | train 2.971723 | val 3.398238 | tok/s 2627493.02 | norm 0.4071 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7211 | train 3.082149 | val 3.398238 | tok/s 2628340.95 | norm 0.3899 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7212 | train 3.134329 | val 3.398238 | tok/s 2622744.17 | norm 0.4472 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7213 | train 3.190396 | val 3.398238 | tok/s 2619380.37 | norm 0.4523 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7214 | train 3.023084 | val 3.398238 | tok/s 2624214.63 | norm 0.4677 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7215 | train 3.009922 | val 3.398238 | tok/s 2629355.98 | norm 0.4372 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7216 | train 3.104117 | val 3.398238 | tok/s 2629067.09 | norm 0.4208 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7217 | train 2.968223 | val 3.398238 | tok/s 2622301.14 | norm 0.4430 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7218 | train 3.112391 | val 3.398238 | tok/s 2625043.42 | norm 0.4766 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7219 | train 3.046131 | val 3.398238 | tok/s 2628552.32 | norm 0.4635 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7220 | train 3.022905 | val 3.398238 | tok/s 2625086.49 | norm 0.4255 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7221 | train 3.047585 | val 3.398238 | tok/s 2632216.64 | norm 0.4319 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7222 | train 2.988750 | val 3.398238 | tok/s 2633034.41 | norm 0.4198 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7223 | train 3.095458 | val 3.398238 | tok/s 2626003.64 | norm 0.4211 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7224 | train 2.964572 | val 3.398238 | tok/s 2628687.69 | norm 0.4443 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7225 | train 2.950349 | val 3.398238 | tok/s 2626211.83 | norm 0.4507 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7226 | train 2.883781 | val 3.398238 | tok/s 2629157.00 | norm 0.4191 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7227 | train 3.005390 | val 3.398238 | tok/s 2623138.76 | norm 0.4150 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7228 | train 2.983010 | val 3.398238 | tok/s 2622511.47 | norm 0.4350 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7229 | train 2.988328 | val 3.398238 | tok/s 2632123.39 | norm 0.4270 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7230 | train 3.057848 | val 3.398238 | tok/s 2629582.82 | norm 0.4076 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7231 | train 3.052788 | val 3.398238 | tok/s 2633402.68 | norm 0.3819 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7232 | train 3.003715 | val 3.398238 | tok/s 2622168.22 | norm 0.4026 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7233 | train 2.948015 | val 3.398238 | tok/s 2626027.91 | norm 0.4481 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7234 | train 3.157706 | val 3.398238 | tok/s 2632162.83 | norm 0.4186 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7235 | train 2.926073 | val 3.398238 | tok/s 2628289.87 | norm 0.4320 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7236 | train 3.087811 | val 3.398238 | tok/s 2536199.32 | norm 0.5201 | dt 206.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7237 | train 3.001527 | val 3.398238 | tok/s 2623857.67 | norm 0.4742 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7238 | train 3.115514 | val 3.398238 | tok/s 2623321.97 | norm 0.4498 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7239 | train 3.091995 | val 3.398238 | tok/s 2628030.19 | norm 0.4655 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7240 | train 3.017632 | val 3.398238 | tok/s 2623468.36 | norm 0.4025 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7241 | train 3.090879 | val 3.398238 | tok/s 2630598.80 | norm 0.3899 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7242 | train 3.195037 | val 3.398238 | tok/s 2626018.34 | norm 0.5371 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7243 | train 3.107573 | val 3.398238 | tok/s 2475522.57 | norm 0.4864 | dt 211.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7244 | train 2.991377 | val 3.398238 | tok/s 2633638.83 | norm 0.3687 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7245 | train 3.025514 | val 3.398238 | tok/s 2633238.28 | norm 0.3648 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7246 | train 3.040016 | val 3.398238 | tok/s 2626761.87 | norm 0.3654 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7247 | train 2.984568 | val 3.398238 | tok/s 2624207.63 | norm 0.4141 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7248 | train 3.048210 | val 3.398238 | tok/s 2631878.73 | norm 0.4585 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7249 | train 3.002133 | val 3.398238 | tok/s 2627901.15 | norm 0.4198 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 7250 | val_loss 3.393853 step 7250 | train 3.009930 | val 3.393853 | tok/s 2486780.39 | norm 0.4489 | dt 210.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7251 | train 3.010116 | val 3.393853 | tok/s 2628799.80 | norm 0.4649 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7252 | train 3.047723 | val 3.393853 | tok/s 2626498.15 | norm 0.4298 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7253 | train 3.075624 | val 3.393853 | tok/s 2623024.61 | norm 0.4184 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7254 | train 3.077372 | val 3.393853 | tok/s 2626676.24 | norm 0.3848 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7255 | train 2.981755 | val 3.393853 | tok/s 2631125.38 | norm 0.4190 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7256 | train 2.978011 | val 3.393853 | tok/s 2631772.32 | norm 0.4563 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7257 | train 2.958347 | val 3.393853 | tok/s 2585265.43 | norm 0.4274 | dt 202.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7258 | train 2.898506 | val 3.393853 | tok/s 2604652.18 | norm 0.4181 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7259 | train 3.041824 | val 3.393853 | tok/s 2622022.73 | norm 0.4257 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7260 | train 2.942339 | val 3.393853 | tok/s 2628681.04 | norm 0.4149 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7261 | train 3.009623 | val 3.393853 | tok/s 2626086.09 | norm 0.4211 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7262 | train 3.004530 | val 3.393853 | tok/s 2625916.28 | norm 0.4397 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7263 | train 3.008149 | val 3.393853 | tok/s 2627723.10 | norm 0.4278 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7264 | train 2.920834 | val 3.393853 | tok/s 2629073.99 | norm 0.4612 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7265 | train 2.970031 | val 3.393853 | tok/s 2628051.63 | norm 0.4284 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7266 | train 3.028241 | val 3.393853 | tok/s 2625098.76 | norm 0.3858 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7267 | train 3.089557 | val 3.393853 | tok/s 2629811.61 | norm 0.4036 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7268 | train 2.929009 | val 3.393853 | tok/s 2625834.40 | norm 0.4205 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7269 | train 3.054647 | val 3.393853 | tok/s 2628479.44 | norm 0.4095 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7270 | train 2.895765 | val 3.393853 | tok/s 2628471.57 | norm 0.4508 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7271 | train 2.979859 | val 3.393853 | tok/s 2626563.30 | norm 0.4432 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7272 | train 3.001532 | val 3.393853 | tok/s 2629051.64 | norm 0.4421 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7273 | train 3.040691 | val 3.393853 | tok/s 2627578.86 | norm 0.4651 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7274 | train 3.101782 | val 3.393853 | tok/s 2632675.03 | norm 0.5507 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7275 | train 3.001873 | val 3.393853 | tok/s 2624385.56 | norm 0.4162 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7276 | train 2.965951 | val 3.393853 | tok/s 2628048.20 | norm 0.4140 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7277 | train 3.019572 | val 3.393853 | tok/s 2626272.41 | norm 0.4394 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7278 | train 3.093614 | val 3.393853 | tok/s 2628125.91 | norm 0.3867 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7279 | train 3.102368 | val 3.393853 | tok/s 2631018.63 | norm 0.3700 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7280 | train 3.215599 | val 3.393853 | tok/s 2625626.98 | norm 0.4277 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7281 | train 3.118864 | val 3.393853 | tok/s 2628542.14 | norm 0.4202 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7282 | train 3.105913 | val 3.393853 | tok/s 2631326.47 | norm 0.3881 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7283 | train 2.972530 | val 3.393853 | tok/s 2627929.66 | norm 0.4204 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7284 | train 3.039598 | val 3.393853 | tok/s 2630692.49 | norm 0.4043 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7285 | train 2.992315 | val 3.393853 | tok/s 2625619.95 | norm 0.4111 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7286 | train 2.898321 | val 3.393853 | tok/s 2628778.71 | norm 0.4183 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7287 | train 3.174107 | val 3.393853 | tok/s 2629846.98 | norm 0.4533 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7288 | train 3.081414 | val 3.393853 | tok/s 2626987.46 | norm 0.4553 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7289 | train 3.078545 | val 3.393853 | tok/s 2629864.44 | norm 0.4429 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7290 | train 2.963670 | val 3.393853 | tok/s 2567146.76 | norm 0.4124 | dt 204.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7291 | train 3.156397 | val 3.393853 | tok/s 2625394.76 | norm 0.4332 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7292 | train 3.040134 | val 3.393853 | tok/s 2627866.96 | norm 0.4329 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7293 | train 3.039131 | val 3.393853 | tok/s 2629640.35 | norm 0.4636 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7294 | train 2.957857 | val 3.393853 | tok/s 2627698.39 | norm 0.4172 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7295 | train 3.029523 | val 3.393853 | tok/s 2621809.50 | norm 0.3657 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7296 | train 2.992956 | val 3.393853 | tok/s 2629736.53 | norm 0.4080 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7297 | train 3.047698 | val 3.393853 | tok/s 2626604.06 | norm 0.4360 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7298 | train 3.033152 | val 3.393853 | tok/s 2627892.46 | norm 0.4406 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7299 | train 3.185666 | val 3.393853 | tok/s 2622518.51 | norm 0.4637 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7300 | train 3.134807 | val 3.393853 | tok/s 2620235.98 | norm 0.4731 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7301 | train 2.988684 | val 3.393853 | tok/s 2624244.23 | norm 0.4212 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7302 | train 3.056166 | val 3.393853 | tok/s 2634161.07 | norm 0.3816 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7303 | train 3.029927 | val 3.393853 | tok/s 2627150.91 | norm 0.4103 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7304 | train 3.065022 | val 3.393853 | tok/s 2629648.07 | norm 0.4142 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7305 | train 3.059739 | val 3.393853 | tok/s 2629934.46 | norm 0.3978 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7306 | train 3.065237 | val 3.393853 | tok/s 2589031.85 | norm 0.4116 | dt 202.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7307 | train 3.099290 | val 3.393853 | tok/s 2613359.48 | norm 0.3918 | dt 200.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7308 | train 2.840100 | val 3.393853 | tok/s 2625910.30 | norm 0.4001 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7309 | train 2.988825 | val 3.393853 | tok/s 2630402.12 | norm 0.3973 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7310 | train 3.023002 | val 3.393853 | tok/s 2627819.94 | norm 0.3747 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7311 | train 3.119890 | val 3.393853 | tok/s 2628249.80 | norm 0.3885 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7312 | train 3.043514 | val 3.393853 | tok/s 2628561.76 | norm 0.4050 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7313 | train 2.969809 | val 3.393853 | tok/s 2625635.00 | norm 0.5022 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7314 | train 3.161462 | val 3.393853 | tok/s 2629083.28 | norm 0.4557 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7315 | train 2.998595 | val 3.393853 | tok/s 2626920.23 | norm 0.3508 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7316 | train 3.073844 | val 3.393853 | tok/s 2628879.23 | norm 0.3767 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7317 | train 3.016946 | val 3.393853 | tok/s 2624559.49 | norm 0.4429 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7318 | train 2.993567 | val 3.393853 | tok/s 2627269.21 | norm 0.4120 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7319 | train 3.010667 | val 3.393853 | tok/s 2628033.93 | norm 0.4452 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7320 | train 3.023614 | val 3.393853 | tok/s 2626619.87 | norm 0.4476 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7321 | train 3.128817 | val 3.393853 | tok/s 2628834.20 | norm 0.5121 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7322 | train 3.007613 | val 3.393853 | tok/s 2624979.85 | norm 0.4593 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7323 | train 3.128772 | val 3.393853 | tok/s 2628909.65 | norm 0.4458 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7324 | train 3.088744 | val 3.393853 | tok/s 2629665.40 | norm 0.4223 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7325 | train 3.153595 | val 3.393853 | tok/s 2626793.04 | norm 0.4255 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7326 | train 3.303068 | val 3.393853 | tok/s 2631275.67 | norm 0.5552 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7327 | train 3.023927 | val 3.393853 | tok/s 2622468.19 | norm 0.4652 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7328 | train 3.005967 | val 3.393853 | tok/s 2627841.45 | norm 0.3998 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7329 | train 3.085109 | val 3.393853 | tok/s 2629775.71 | norm 0.4322 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7330 | train 2.915790 | val 3.393853 | tok/s 2632299.34 | norm 0.4302 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7331 | train 3.075705 | val 3.393853 | tok/s 2630178.25 | norm 0.4177 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7332 | train 3.159017 | val 3.393853 | tok/s 2626394.60 | norm 0.4083 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7333 | train 3.072274 | val 3.393853 | tok/s 2626079.53 | norm 0.4120 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7334 | train 3.026367 | val 3.393853 | tok/s 2629874.92 | norm 0.4142 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7335 | train 3.165324 | val 3.393853 | tok/s 2629420.40 | norm 0.4988 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7336 | train 2.950476 | val 3.393853 | tok/s 2629823.93 | norm 0.4579 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7337 | train 3.055634 | val 3.393853 | tok/s 2626793.30 | norm 0.3597 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7338 | train 3.091348 | val 3.393853 | tok/s 2627565.39 | norm 0.4617 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7339 | train 3.112646 | val 3.393853 | tok/s 2631347.66 | norm 0.4722 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7340 | train 2.947776 | val 3.393853 | tok/s 2625801.97 | norm 0.5513 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7341 | train 3.078005 | val 3.393853 | tok/s 2626649.30 | norm 0.4121 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7342 | train 3.047216 | val 3.393853 | tok/s 2628654.92 | norm 0.4049 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7343 | train 3.027342 | val 3.393853 | tok/s 2630570.38 | norm 0.4303 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7344 | train 3.006149 | val 3.393853 | tok/s 2629811.11 | norm 0.4167 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7345 | train 2.947530 | val 3.393853 | tok/s 2626866.30 | norm 0.3981 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7346 | train 3.057139 | val 3.393853 | tok/s 2603026.13 | norm 0.4064 | dt 201.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7347 | train 3.067480 | val 3.393853 | tok/s 2622843.01 | norm 0.3614 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7348 | train 3.020334 | val 3.393853 | tok/s 2603895.15 | norm 0.4238 | dt 201.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7349 | train 3.076464 | val 3.393853 | tok/s 2629215.38 | norm 0.4043 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7350 | train 3.014843 | val 3.393853 | tok/s 2626518.96 | norm 0.3947 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7351 | train 3.044097 | val 3.393853 | tok/s 2629731.83 | norm 0.4601 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7352 | train 3.114274 | val 3.393853 | tok/s 2627655.46 | norm 0.5113 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7353 | train 2.988689 | val 3.393853 | tok/s 2627972.60 | norm 0.4197 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7354 | train 2.987720 | val 3.393853 | tok/s 2629715.56 | norm 0.4211 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7355 | train 2.933190 | val 3.393853 | tok/s 2627679.03 | norm 0.3925 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7356 | train 2.956650 | val 3.393853 | tok/s 2624708.88 | norm 0.3706 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7357 | train 3.000998 | val 3.393853 | tok/s 2626808.34 | norm 0.3846 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7358 | train 3.036304 | val 3.393853 | tok/s 2631942.34 | norm 0.3954 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7359 | train 3.178877 | val 3.393853 | tok/s 2626944.46 | norm 0.4248 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7360 | train 2.981915 | val 3.393853 | tok/s 2626271.20 | norm 0.4313 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7361 | train 2.983138 | val 3.393853 | tok/s 2631484.68 | norm 0.4396 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7362 | train 2.971557 | val 3.393853 | tok/s 2624254.64 | norm 0.4214 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7363 | train 3.021883 | val 3.393853 | tok/s 2625875.92 | norm 0.4475 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7364 | train 3.070879 | val 3.393853 | tok/s 2624307.13 | norm 0.4134 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7365 | train 2.910441 | val 3.393853 | tok/s 2624220.01 | norm 0.3965 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7366 | train 3.044154 | val 3.393853 | tok/s 2631094.29 | norm 0.4159 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7367 | train 2.969862 | val 3.393853 | tok/s 2627940.76 | norm 0.3823 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7368 | train 3.063717 | val 3.393853 | tok/s 2627654.31 | norm 0.3673 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7369 | train 3.016395 | val 3.393853 | tok/s 2629582.98 | norm 0.4047 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7370 | train 3.081852 | val 3.393853 | tok/s 2628990.32 | norm 0.4306 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7371 | train 3.013179 | val 3.393853 | tok/s 2629110.80 | norm 0.4799 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7372 | train 3.064704 | val 3.393853 | tok/s 2624311.01 | norm 0.4242 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7373 | train 3.113573 | val 3.393853 | tok/s 2626935.68 | norm 0.3902 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7374 | train 3.005015 | val 3.393853 | tok/s 2627753.43 | norm 0.4099 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7375 | train 2.995586 | val 3.393853 | tok/s 2630690.22 | norm 0.4180 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7376 | train 3.012869 | val 3.393853 | tok/s 2628932.22 | norm 0.4163 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7377 | train 3.009761 | val 3.393853 | tok/s 2625634.69 | norm 0.4324 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7378 | train 3.043941 | val 3.393853 | tok/s 2628887.03 | norm 0.4218 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7379 | train 2.985774 | val 3.393853 | tok/s 2629341.24 | norm 0.4272 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7380 | train 3.034927 | val 3.393853 | tok/s 2628713.64 | norm 0.4118 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7381 | train 3.023914 | val 3.393853 | tok/s 2630914.92 | norm 0.4106 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7382 | train 3.025257 | val 3.393853 | tok/s 2624634.77 | norm 0.3947 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7383 | train 3.111867 | val 3.393853 | tok/s 2629480.75 | norm 0.3891 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7384 | train 3.154424 | val 3.393853 | tok/s 2622550.07 | norm 0.3763 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7385 | train 3.109347 | val 3.393853 | tok/s 2627029.94 | norm 0.3501 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7386 | train 3.057488 | val 3.393853 | tok/s 2629089.89 | norm 0.3547 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7387 | train 3.021823 | val 3.393853 | tok/s 2630438.48 | norm 0.3885 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7388 | train 2.993156 | val 3.393853 | tok/s 2630027.42 | norm 0.4186 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7389 | train 2.967190 | val 3.393853 | tok/s 2624708.84 | norm 0.4349 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7390 | train 3.112103 | val 3.393853 | tok/s 2629414.87 | norm 0.4350 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7391 | train 2.979325 | val 3.393853 | tok/s 2629295.11 | norm 0.4513 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7392 | train 2.980529 | val 3.393853 | tok/s 2627573.64 | norm 0.4565 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7393 | train 3.009828 | val 3.393853 | tok/s 2627871.57 | norm 0.4783 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7394 | train 3.049159 | val 3.393853 | tok/s 2617192.28 | norm 0.4310 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7395 | train 3.157336 | val 3.393853 | tok/s 2624099.84 | norm 0.4415 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7396 | train 2.931058 | val 3.393853 | tok/s 2629046.29 | norm 0.4261 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7397 | train 2.993076 | val 3.393853 | tok/s 2627341.42 | norm 0.4230 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7398 | train 3.058320 | val 3.393853 | tok/s 2628031.44 | norm 0.4446 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7399 | train 3.009087 | val 3.393853 | tok/s 2623546.18 | norm 0.4148 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7400 | train 3.035292 | val 3.393853 | tok/s 2626878.80 | norm 0.3951 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7401 | train 3.058580 | val 3.393853 | tok/s 2627481.68 | norm 0.4347 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7402 | train 3.056653 | val 3.393853 | tok/s 2624662.90 | norm 0.4044 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7403 | train 3.078485 | val 3.393853 | tok/s 2627434.93 | norm 0.3947 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7404 | train 3.028580 | val 3.393853 | tok/s 2629038.39 | norm 0.4372 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7405 | train 2.984347 | val 3.393853 | tok/s 2629121.60 | norm 0.4366 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7406 | train 3.119214 | val 3.393853 | tok/s 2632184.44 | norm 0.4182 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7407 | train 3.058814 | val 3.393853 | tok/s 2630771.61 | norm 0.4300 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7408 | train 3.035587 | val 3.393853 | tok/s 2632484.40 | norm 0.4167 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7409 | train 2.999274 | val 3.393853 | tok/s 2621836.08 | norm 0.4515 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7410 | train 3.071280 | val 3.393853 | tok/s 2624637.14 | norm 0.4000 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7411 | train 2.968897 | val 3.393853 | tok/s 2631526.56 | norm 0.3691 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7412 | train 3.065470 | val 3.393853 | tok/s 2625817.74 | norm 0.4081 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7413 | train 2.949900 | val 3.393853 | tok/s 2632978.77 | norm 0.5299 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7414 | train 3.065887 | val 3.393853 | tok/s 2624399.85 | norm 0.3859 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7415 | train 3.069263 | val 3.393853 | tok/s 2627846.39 | norm 0.4375 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7416 | train 2.926517 | val 3.393853 | tok/s 2627563.91 | norm 0.4106 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7417 | train 3.149909 | val 3.393853 | tok/s 2628142.75 | norm 0.4287 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7418 | train 3.052519 | val 3.393853 | tok/s 2632677.95 | norm 0.4148 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7419 | train 2.958349 | val 3.393853 | tok/s 2627953.21 | norm 0.4170 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7420 | train 2.967465 | val 3.393853 | tok/s 2622788.66 | norm 0.4469 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7421 | train 3.081893 | val 3.393853 | tok/s 2631757.36 | norm 0.4208 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7422 | train 3.059188 | val 3.393853 | tok/s 2626007.26 | norm 0.3902 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7423 | train 2.971993 | val 3.393853 | tok/s 2631070.38 | norm 0.4338 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7424 | train 3.009716 | val 3.393853 | tok/s 2509189.46 | norm 0.4501 | dt 208.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7425 | train 3.099246 | val 3.393853 | tok/s 2626024.90 | norm 0.3992 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7426 | train 2.995176 | val 3.393853 | tok/s 2622227.86 | norm 0.4191 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7427 | train 2.887224 | val 3.393853 | tok/s 2618883.03 | norm 0.4266 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7428 | train 3.079066 | val 3.393853 | tok/s 2629549.20 | norm 0.4220 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7429 | train 2.983056 | val 3.393853 | tok/s 2627606.03 | norm 0.4089 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7430 | train 2.981479 | val 3.393853 | tok/s 2629048.78 | norm 0.3977 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7431 | train 3.119537 | val 3.393853 | tok/s 2626480.23 | norm 0.4789 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7432 | train 3.097843 | val 3.393853 | tok/s 2626926.24 | norm 0.5141 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7433 | train 2.974535 | val 3.393853 | tok/s 2623202.91 | norm 0.4275 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7434 | train 2.995359 | val 3.393853 | tok/s 2626123.52 | norm 0.4426 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7435 | train 3.077625 | val 3.393853 | tok/s 2629050.30 | norm 0.3798 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7436 | train 3.009317 | val 3.393853 | tok/s 2624723.16 | norm 0.3633 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7437 | train 3.145709 | val 3.393853 | tok/s 2626699.76 | norm 0.4216 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7438 | train 3.075418 | val 3.393853 | tok/s 2629656.60 | norm 0.4096 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7439 | train 2.994148 | val 3.393853 | tok/s 2627059.70 | norm 0.4622 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7440 | train 2.954303 | val 3.393853 | tok/s 2629895.63 | norm 0.4902 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7441 | train 3.044262 | val 3.393853 | tok/s 2625530.69 | norm 0.4513 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7442 | train 3.089441 | val 3.393853 | tok/s 2628395.80 | norm 0.4487 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7443 | train 3.116163 | val 3.393853 | tok/s 2630291.43 | norm 0.4318 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7444 | train 3.020021 | val 3.393853 | tok/s 2623376.34 | norm 0.3749 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7445 | train 2.952657 | val 3.393853 | tok/s 2632668.69 | norm 0.4065 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7446 | train 2.943659 | val 3.393853 | tok/s 2623599.69 | norm 0.4465 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7447 | train 2.914783 | val 3.393853 | tok/s 2629806.15 | norm 0.3957 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7448 | train 3.127162 | val 3.393853 | tok/s 2628707.63 | norm 0.3999 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7449 | train 2.963831 | val 3.393853 | tok/s 2625243.95 | norm 0.3986 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7450 | train 3.032163 | val 3.393853 | tok/s 2628348.44 | norm 0.3802 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7451 | train 2.981825 | val 3.393853 | tok/s 2609645.32 | norm 0.4129 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7452 | train 3.124434 | val 3.393853 | tok/s 2626953.66 | norm 0.4113 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7453 | train 3.100529 | val 3.393853 | tok/s 2630883.36 | norm 0.4636 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7454 | train 3.029256 | val 3.393853 | tok/s 2631961.20 | norm 0.4681 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7455 | train 2.834302 | val 3.393853 | tok/s 2629541.09 | norm 0.4188 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7456 | train 2.926150 | val 3.393853 | tok/s 2628193.80 | norm 0.4510 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7457 | train 3.105732 | val 3.393853 | tok/s 2629536.50 | norm 0.4691 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7458 | train 3.036681 | val 3.393853 | tok/s 2630582.59 | norm 0.4449 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7459 | train 3.028612 | val 3.393853 | tok/s 2627661.33 | norm 0.4075 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7460 | train 2.990180 | val 3.393853 | tok/s 2627612.31 | norm 0.4269 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7461 | train 2.956108 | val 3.393853 | tok/s 2628466.31 | norm 0.4141 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7462 | train 3.217761 | val 3.393853 | tok/s 2626063.63 | norm 0.4020 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7463 | train 2.887391 | val 3.393853 | tok/s 2633231.88 | norm 0.3741 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7464 | train 3.005533 | val 3.393853 | tok/s 2631110.19 | norm 0.3979 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7465 | train 3.037469 | val 3.393853 | tok/s 2632442.77 | norm 0.4252 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7466 | train 3.050063 | val 3.393853 | tok/s 2626371.16 | norm 0.4354 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7467 | train 3.052626 | val 3.393853 | tok/s 2614743.71 | norm 0.4621 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7468 | train 2.967659 | val 3.393853 | tok/s 2586870.61 | norm 0.4826 | dt 202.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7469 | train 3.175876 | val 3.393853 | tok/s 2621800.48 | norm 0.5040 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7470 | train 2.943593 | val 3.393853 | tok/s 2629465.89 | norm 0.4476 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7471 | train 3.042496 | val 3.393853 | tok/s 2627111.07 | norm 0.4253 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7472 | train 3.098772 | val 3.393853 | tok/s 2631010.83 | norm 0.4422 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7473 | train 3.038369 | val 3.393853 | tok/s 2631599.43 | norm 0.4020 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7474 | train 3.038720 | val 3.393853 | tok/s 2628771.21 | norm 0.3935 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7475 | train 2.949594 | val 3.393853 | tok/s 2631543.96 | norm 0.4023 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7476 | train 3.086590 | val 3.393853 | tok/s 2629953.34 | norm 0.3803 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7477 | train 3.052291 | val 3.393853 | tok/s 2631599.91 | norm 0.4150 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7478 | train 3.082364 | val 3.393853 | tok/s 2628920.16 | norm 0.4107 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7479 | train 3.007979 | val 3.393853 | tok/s 2623304.36 | norm 0.4089 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7480 | train 3.028624 | val 3.393853 | tok/s 2630305.85 | norm 0.3851 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7481 | train 3.083458 | val 3.393853 | tok/s 2626514.16 | norm 0.4386 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7482 | train 3.166286 | val 3.393853 | tok/s 2505638.34 | norm 0.4226 | dt 209.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7483 | train 2.975846 | val 3.393853 | tok/s 2628737.59 | norm 0.4231 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7484 | train 2.971958 | val 3.393853 | tok/s 2623692.05 | norm 0.3858 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7485 | train 3.063359 | val 3.393853 | tok/s 2625816.32 | norm 0.3632 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7486 | train 3.026877 | val 3.393853 | tok/s 2625476.76 | norm 0.4060 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7487 | train 3.086294 | val 3.393853 | tok/s 2627761.83 | norm 0.4544 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7488 | train 3.044245 | val 3.393853 | tok/s 2597773.27 | norm 0.4292 | dt 201.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7489 | train 2.946383 | val 3.393853 | tok/s 2625540.11 | norm 0.4342 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7490 | train 3.120602 | val 3.393853 | tok/s 2634929.65 | norm 0.4570 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7491 | train 3.057411 | val 3.393853 | tok/s 2630142.67 | norm 0.4483 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7492 | train 3.028972 | val 3.393853 | tok/s 2629057.50 | norm 0.4301 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7493 | train 3.009305 | val 3.393853 | tok/s 2630816.77 | norm 0.4459 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7494 | train 2.932122 | val 3.393853 | tok/s 2631582.18 | norm 0.4353 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7495 | train 3.003464 | val 3.393853 | tok/s 2631336.87 | norm 0.4262 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7496 | train 3.027886 | val 3.393853 | tok/s 2625222.47 | norm 0.4278 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7497 | train 2.979304 | val 3.393853 | tok/s 2628973.52 | norm 0.4100 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7498 | train 2.988503 | val 3.393853 | tok/s 2626145.36 | norm 0.4259 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7499 | train 3.174047 | val 3.393853 | tok/s 2624878.34 | norm 0.4736 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 7500 | val_loss 3.388238 step 7500 | train 3.086674 | val 3.388238 | tok/s 2602943.15 | norm 0.4087 | dt 201.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7501 | train 3.027654 | val 3.388238 | tok/s 2621447.43 | norm 0.4026 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7502 | train 2.977934 | val 3.388238 | tok/s 2627856.48 | norm 0.4326 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7503 | train 3.066205 | val 3.388238 | tok/s 2631934.13 | norm 0.4236 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7504 | train 3.066886 | val 3.388238 | tok/s 2622596.08 | norm 0.4100 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7505 | train 2.947994 | val 3.388238 | tok/s 2623696.23 | norm 0.4007 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7506 | train 2.933552 | val 3.388238 | tok/s 2627538.00 | norm 0.3802 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7507 | train 3.036270 | val 3.388238 | tok/s 2626975.55 | norm 0.3583 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7508 | train 3.022852 | val 3.388238 | tok/s 2630303.17 | norm 0.4053 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7509 | train 2.968354 | val 3.388238 | tok/s 2627405.95 | norm 0.4622 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7510 | train 3.040629 | val 3.388238 | tok/s 2628490.20 | norm 0.3935 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7511 | train 3.166838 | val 3.388238 | tok/s 2630931.97 | norm 0.4212 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7512 | train 3.029363 | val 3.388238 | tok/s 2631026.05 | norm 0.4834 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7513 | train 2.998490 | val 3.388238 | tok/s 2632390.92 | norm 0.4465 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7514 | train 3.068243 | val 3.388238 | tok/s 2623704.39 | norm 0.4178 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7515 | train 2.898837 | val 3.388238 | tok/s 2625798.29 | norm 0.4725 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7516 | train 2.993415 | val 3.388238 | tok/s 2630836.19 | norm 0.4168 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7517 | train 2.976100 | val 3.388238 | tok/s 2623301.13 | norm 0.4194 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7518 | train 2.894849 | val 3.388238 | tok/s 2627765.47 | norm 0.4643 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7519 | train 3.010185 | val 3.388238 | tok/s 2624404.54 | norm 0.4480 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7520 | train 3.095597 | val 3.388238 | tok/s 2627896.28 | norm 0.4410 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7521 | train 2.993173 | val 3.388238 | tok/s 2628956.96 | norm 0.4121 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7522 | train 3.149381 | val 3.388238 | tok/s 2514031.52 | norm 0.3819 | dt 208.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7523 | train 3.053615 | val 3.388238 | tok/s 2627806.32 | norm 0.4263 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7524 | train 3.079719 | val 3.388238 | tok/s 2584679.23 | norm 0.4926 | dt 202.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7525 | train 3.066059 | val 3.388238 | tok/s 2626957.27 | norm 0.4753 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7526 | train 3.115065 | val 3.388238 | tok/s 2626308.68 | norm 0.4609 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7527 | train 3.149492 | val 3.388238 | tok/s 2624732.46 | norm 0.4243 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7528 | train 3.153238 | val 3.388238 | tok/s 2628499.70 | norm 0.4235 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7529 | train 3.035820 | val 3.388238 | tok/s 2629986.07 | norm 0.4821 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7530 | train 2.888023 | val 3.388238 | tok/s 2629098.27 | norm 0.4125 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7531 | train 2.972713 | val 3.388238 | tok/s 2627093.70 | norm 0.4008 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7532 | train 3.063385 | val 3.388238 | tok/s 2623476.89 | norm 0.4163 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7533 | train 2.956188 | val 3.388238 | tok/s 2627745.91 | norm 0.4345 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7534 | train 2.959397 | val 3.388238 | tok/s 2625486.66 | norm 0.4265 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7535 | train 2.945031 | val 3.388238 | tok/s 2628778.04 | norm 0.4298 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7536 | train 3.030039 | val 3.388238 | tok/s 2628399.08 | norm 0.4056 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7537 | train 3.021706 | val 3.388238 | tok/s 2623564.77 | norm 0.4064 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7538 | train 2.989621 | val 3.388238 | tok/s 2627305.22 | norm 0.4208 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7539 | train 2.989820 | val 3.388238 | tok/s 2478743.18 | norm 0.4245 | dt 211.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7540 | train 2.996827 | val 3.388238 | tok/s 2628620.30 | norm 0.4264 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7541 | train 3.032254 | val 3.388238 | tok/s 2631071.15 | norm 0.4555 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7542 | train 3.098590 | val 3.388238 | tok/s 2630780.37 | norm 0.4387 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7543 | train 2.983510 | val 3.388238 | tok/s 2630945.80 | norm 0.4162 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7544 | train 3.043438 | val 3.388238 | tok/s 2627190.49 | norm 0.4616 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7545 | train 3.030849 | val 3.388238 | tok/s 2629814.29 | norm 0.3938 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7546 | train 3.102333 | val 3.388238 | tok/s 2629160.49 | norm 0.4168 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7547 | train 3.080047 | val 3.388238 | tok/s 2627048.87 | norm 0.4433 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7548 | train 3.093042 | val 3.388238 | tok/s 2629505.13 | norm 0.4556 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7549 | train 3.027313 | val 3.388238 | tok/s 2631541.41 | norm 0.4622 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7550 | train 3.063657 | val 3.388238 | tok/s 2627448.99 | norm 0.4374 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7551 | train 2.932458 | val 3.388238 | tok/s 2620268.75 | norm 0.4269 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7552 | train 2.948424 | val 3.388238 | tok/s 2627681.16 | norm 0.4245 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7553 | train 3.016963 | val 3.388238 | tok/s 2635320.25 | norm 0.3947 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7554 | train 3.084861 | val 3.388238 | tok/s 2631599.90 | norm 0.3568 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7555 | train 2.970578 | val 3.388238 | tok/s 2631156.82 | norm 0.3721 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7556 | train 3.097365 | val 3.388238 | tok/s 2627726.74 | norm 0.3709 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7557 | train 2.947498 | val 3.388238 | tok/s 2629789.46 | norm 0.4231 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7558 | train 3.010960 | val 3.388238 | tok/s 2632053.94 | norm 0.3863 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7559 | train 2.955892 | val 3.388238 | tok/s 2628168.44 | norm 0.4050 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7560 | train 3.020878 | val 3.388238 | tok/s 2627457.09 | norm 0.4603 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7561 | train 3.067668 | val 3.388238 | tok/s 2628608.89 | norm 0.4300 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7562 | train 2.950600 | val 3.388238 | tok/s 2630558.20 | norm 0.4508 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7563 | train 3.051866 | val 3.388238 | tok/s 2632648.92 | norm 0.4716 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7564 | train 2.982300 | val 3.388238 | tok/s 2627020.76 | norm 0.4467 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7565 | train 2.963317 | val 3.388238 | tok/s 2627002.57 | norm 0.4153 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7566 | train 3.029360 | val 3.388238 | tok/s 2603370.64 | norm 0.4061 | dt 201.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7567 | train 3.007699 | val 3.388238 | tok/s 2626786.60 | norm 0.4256 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7568 | train 2.969861 | val 3.388238 | tok/s 2623467.27 | norm 0.4424 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7569 | train 3.043516 | val 3.388238 | tok/s 2600134.83 | norm 0.4331 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7570 | train 3.021516 | val 3.388238 | tok/s 2630683.78 | norm 0.4160 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7571 | train 3.005310 | val 3.388238 | tok/s 2627799.17 | norm 0.3950 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7572 | train 3.009689 | val 3.388238 | tok/s 2629644.23 | norm 0.4195 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7573 | train 3.001544 | val 3.388238 | tok/s 2628607.00 | norm 0.4082 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7574 | train 3.002492 | val 3.388238 | tok/s 2610068.40 | norm 0.4100 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7575 | train 3.113481 | val 3.388238 | tok/s 2630717.92 | norm 0.3909 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7576 | train 3.036474 | val 3.388238 | tok/s 2626447.31 | norm 0.3777 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7577 | train 3.042960 | val 3.388238 | tok/s 2633198.01 | norm 0.4106 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7578 | train 2.930593 | val 3.388238 | tok/s 2629813.92 | norm 0.4065 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7579 | train 2.970945 | val 3.388238 | tok/s 2621820.52 | norm 0.3752 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7580 | train 3.020804 | val 3.388238 | tok/s 2630028.07 | norm 0.3961 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7581 | train 2.990407 | val 3.388238 | tok/s 2633795.61 | norm 0.4258 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7582 | train 3.015027 | val 3.388238 | tok/s 2634726.42 | norm 0.4403 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7583 | train 2.983225 | val 3.388238 | tok/s 2628071.81 | norm 0.4081 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7584 | train 2.984699 | val 3.388238 | tok/s 2627850.76 | norm 0.3636 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7585 | train 3.083858 | val 3.388238 | tok/s 2486422.11 | norm 0.3933 | dt 210.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7586 | train 2.961383 | val 3.388238 | tok/s 2629742.73 | norm 0.4166 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7587 | train 3.029114 | val 3.388238 | tok/s 2630613.99 | norm 0.3972 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7588 | train 3.081177 | val 3.388238 | tok/s 2632027.84 | norm 0.4235 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7589 | train 3.014577 | val 3.388238 | tok/s 2628455.22 | norm 0.4294 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7590 | train 3.034863 | val 3.388238 | tok/s 2632240.74 | norm 0.4328 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7591 | train 3.207473 | val 3.388238 | tok/s 2628592.81 | norm 0.3934 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7592 | train 3.139197 | val 3.388238 | tok/s 2629405.45 | norm 0.4028 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7593 | train 3.109710 | val 3.388238 | tok/s 2616717.86 | norm 0.4311 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7594 | train 3.045329 | val 3.388238 | tok/s 2630470.58 | norm 0.4321 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7595 | train 3.118651 | val 3.388238 | tok/s 2633005.81 | norm 0.4270 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7596 | train 3.025132 | val 3.388238 | tok/s 2633117.28 | norm 0.4247 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7597 | train 2.981894 | val 3.388238 | tok/s 2632190.02 | norm 0.3961 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7598 | train 2.919803 | val 3.388238 | tok/s 2608062.26 | norm 0.4286 | dt 201.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7599 | train 2.965281 | val 3.388238 | tok/s 2623497.65 | norm 0.4556 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7600 | train 2.923311 | val 3.388238 | tok/s 2628392.87 | norm 0.4462 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7601 | train 3.138037 | val 3.388238 | tok/s 2628057.73 | norm 0.4146 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7602 | train 2.988810 | val 3.388238 | tok/s 2628515.02 | norm 0.3698 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7603 | train 3.024761 | val 3.388238 | tok/s 2624256.31 | norm 0.3845 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7604 | train 3.048036 | val 3.388238 | tok/s 2623886.75 | norm 0.4398 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7605 | train 2.960498 | val 3.388238 | tok/s 2627686.52 | norm 0.4179 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7606 | train 3.116796 | val 3.388238 | tok/s 2627284.24 | norm 0.3699 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7607 | train 2.970719 | val 3.388238 | tok/s 2625160.94 | norm 0.3938 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7608 | train 2.912311 | val 3.388238 | tok/s 2626970.01 | norm 0.4302 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7609 | train 3.031138 | val 3.388238 | tok/s 2627069.05 | norm 0.4198 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7610 | train 3.005027 | val 3.388238 | tok/s 2629030.67 | norm 0.4103 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7611 | train 3.125606 | val 3.388238 | tok/s 2626768.45 | norm 0.3960 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7612 | train 3.085169 | val 3.388238 | tok/s 2627666.54 | norm 0.4035 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7613 | train 2.847384 | val 3.388238 | tok/s 2626748.88 | norm 0.3985 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7614 | train 2.969409 | val 3.388238 | tok/s 2629175.57 | norm 0.4135 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7615 | train 3.010672 | val 3.388238 | tok/s 2632063.19 | norm 0.4065 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7616 | train 3.045243 | val 3.388238 | tok/s 2630548.71 | norm 0.4034 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7617 | train 2.998149 | val 3.388238 | tok/s 2631464.86 | norm 0.4267 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7618 | train 2.916301 | val 3.388238 | tok/s 2630228.08 | norm 0.4219 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7619 | train 3.010639 | val 3.388238 | tok/s 2633139.91 | norm 0.4478 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7620 | train 3.048091 | val 3.388238 | tok/s 2630697.11 | norm 0.4124 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7621 | train 2.992265 | val 3.388238 | tok/s 2628787.84 | norm 0.3949 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7622 | train 3.021062 | val 3.388238 | tok/s 2631717.42 | norm 0.3824 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7623 | train 3.081944 | val 3.388238 | tok/s 2624651.72 | norm 0.4263 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7624 | train 2.961095 | val 3.388238 | tok/s 2626360.47 | norm 0.4517 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7625 | train 2.955629 | val 3.388238 | tok/s 2632522.55 | norm 0.4171 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7626 | train 2.992860 | val 3.388238 | tok/s 2584720.95 | norm 0.4105 | dt 202.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7627 | train 3.003254 | val 3.388238 | tok/s 2627978.70 | norm 0.3828 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7628 | train 2.918246 | val 3.388238 | tok/s 2626525.93 | norm 0.3929 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7629 | train 3.165067 | val 3.388238 | tok/s 2627672.54 | norm 0.4168 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7630 | train 3.054279 | val 3.388238 | tok/s 2626262.59 | norm 0.4062 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7631 | train 3.035015 | val 3.388238 | tok/s 2623819.08 | norm 0.3753 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7632 | train 3.015781 | val 3.388238 | tok/s 2606677.80 | norm 0.4372 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7633 | train 3.056507 | val 3.388238 | tok/s 2631777.62 | norm 0.4381 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7634 | train 3.028707 | val 3.388238 | tok/s 2631780.49 | norm 0.4512 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7635 | train 3.095394 | val 3.388238 | tok/s 2630912.67 | norm 0.4135 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7636 | train 3.054548 | val 3.388238 | tok/s 2624504.21 | norm 0.3662 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7637 | train 3.019676 | val 3.388238 | tok/s 2629695.36 | norm 0.3951 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7638 | train 2.923344 | val 3.388238 | tok/s 2623813.83 | norm 0.4169 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7639 | train 3.025486 | val 3.388238 | tok/s 2631076.12 | norm 0.4103 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7640 | train 3.042801 | val 3.388238 | tok/s 2631782.01 | norm 0.3904 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7641 | train 3.044461 | val 3.388238 | tok/s 2623861.88 | norm 0.3993 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7642 | train 3.104126 | val 3.388238 | tok/s 2626510.00 | norm 0.4170 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7643 | train 3.089216 | val 3.388238 | tok/s 2589801.94 | norm 0.6190 | dt 202.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7644 | train 3.068300 | val 3.388238 | tok/s 2629287.45 | norm 0.4733 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7645 | train 2.986613 | val 3.388238 | tok/s 2620486.18 | norm 0.4281 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7646 | train 3.009708 | val 3.388238 | tok/s 2624977.84 | norm 0.4451 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7647 | train 3.142635 | val 3.388238 | tok/s 2630855.96 | norm 0.4048 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7648 | train 3.009775 | val 3.388238 | tok/s 2627001.55 | norm 0.4532 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7649 | train 2.987703 | val 3.388238 | tok/s 2624517.18 | norm 0.4490 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7650 | train 2.953807 | val 3.388238 | tok/s 2615834.57 | norm 0.4069 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7651 | train 3.127370 | val 3.388238 | tok/s 2627621.45 | norm 0.4453 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7652 | train 2.981615 | val 3.388238 | tok/s 2633601.74 | norm 0.4726 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7653 | train 3.039292 | val 3.388238 | tok/s 2628643.85 | norm 0.4182 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7654 | train 3.035599 | val 3.388238 | tok/s 2626462.69 | norm 0.3911 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7655 | train 3.080806 | val 3.388238 | tok/s 2623571.60 | norm 0.4052 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7656 | train 3.029346 | val 3.388238 | tok/s 2626229.74 | norm 0.4235 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7657 | train 3.090338 | val 3.388238 | tok/s 2626750.99 | norm 0.4261 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7658 | train 2.991134 | val 3.388238 | tok/s 2629182.43 | norm 0.4118 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7659 | train 2.977163 | val 3.388238 | tok/s 2626557.30 | norm 0.4039 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7660 | train 3.005015 | val 3.388238 | tok/s 2631368.56 | norm 0.4166 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7661 | train 3.135676 | val 3.388238 | tok/s 2629655.81 | norm 0.4698 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7662 | train 3.050626 | val 3.388238 | tok/s 2633708.75 | norm 0.4580 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7663 | train 2.935681 | val 3.388238 | tok/s 2624303.31 | norm 0.4366 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7664 | train 3.067775 | val 3.388238 | tok/s 2630473.65 | norm 0.4849 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7665 | train 2.973561 | val 3.388238 | tok/s 2628718.69 | norm 0.4248 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7666 | train 3.090956 | val 3.388238 | tok/s 2629402.05 | norm 0.4181 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7667 | train 3.010943 | val 3.388238 | tok/s 2627839.75 | norm 0.4155 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7668 | train 2.931530 | val 3.388238 | tok/s 2624726.79 | norm 0.3897 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7669 | train 2.980598 | val 3.388238 | tok/s 2631979.13 | norm 0.3903 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7670 | train 3.130248 | val 3.388238 | tok/s 2622478.61 | norm 0.5014 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7671 | train 2.906594 | val 3.388238 | tok/s 2625129.70 | norm 0.4292 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7672 | train 3.076600 | val 3.388238 | tok/s 2630359.52 | norm 0.4194 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7673 | train 3.020477 | val 3.388238 | tok/s 2625622.47 | norm 0.3992 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7674 | train 2.955094 | val 3.388238 | tok/s 2628059.49 | norm 0.4170 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7675 | train 2.967142 | val 3.388238 | tok/s 2621395.80 | norm 0.3774 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7676 | train 3.147399 | val 3.388238 | tok/s 2627374.73 | norm 0.4304 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7677 | train 3.223121 | val 3.388238 | tok/s 2631952.02 | norm 0.4712 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7678 | train 3.089545 | val 3.388238 | tok/s 2629504.94 | norm 0.4017 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7679 | train 2.880518 | val 3.388238 | tok/s 2627621.94 | norm 0.3828 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7680 | train 3.219712 | val 3.388238 | tok/s 2625478.48 | norm 0.3980 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7681 | train 2.932544 | val 3.388238 | tok/s 2625159.65 | norm 0.4168 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7682 | train 3.108297 | val 3.388238 | tok/s 2629663.84 | norm 0.4372 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7683 | train 2.948402 | val 3.388238 | tok/s 2524645.09 | norm 0.4482 | dt 207.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7684 | train 2.918251 | val 3.388238 | tok/s 2582662.04 | norm 0.4444 | dt 203.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7685 | train 2.964049 | val 3.388238 | tok/s 2630199.58 | norm 0.4238 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7686 | train 3.120527 | val 3.388238 | tok/s 2629646.56 | norm 0.4585 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7687 | train 3.104767 | val 3.388238 | tok/s 2626247.39 | norm 0.3996 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7688 | train 2.964519 | val 3.388238 | tok/s 2625364.33 | norm 0.4108 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7689 | train 2.894281 | val 3.388238 | tok/s 2628974.90 | norm 0.4580 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7690 | train 2.949819 | val 3.388238 | tok/s 2626687.57 | norm 0.4659 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7691 | train 3.052030 | val 3.388238 | tok/s 2630361.57 | norm 0.4253 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7692 | train 3.076999 | val 3.388238 | tok/s 2626784.55 | norm 0.3988 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7693 | train 3.048974 | val 3.388238 | tok/s 2627555.89 | norm 0.3964 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7694 | train 3.251255 | val 3.388238 | tok/s 2631142.31 | norm 0.4954 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7695 | train 3.104456 | val 3.388238 | tok/s 2623433.93 | norm 0.5210 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7696 | train 3.088166 | val 3.388238 | tok/s 2627234.55 | norm 0.3653 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7697 | train 3.016979 | val 3.388238 | tok/s 2611736.69 | norm 0.3653 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7698 | train 3.034997 | val 3.388238 | tok/s 2625851.60 | norm 0.3979 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7699 | train 2.940285 | val 3.388238 | tok/s 2565756.57 | norm 0.4030 | dt 204.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7700 | train 3.067215 | val 3.388238 | tok/s 2628207.07 | norm 0.5371 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7701 | train 3.048081 | val 3.388238 | tok/s 2629670.67 | norm 0.4772 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7702 | train 3.006098 | val 3.388238 | tok/s 2624733.04 | norm 0.4048 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7703 | train 3.045571 | val 3.388238 | tok/s 2625675.98 | norm 0.4560 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7704 | train 2.977215 | val 3.388238 | tok/s 2629463.87 | norm 0.4613 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7705 | train 3.085078 | val 3.388238 | tok/s 2626257.08 | norm 0.5559 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7706 | train 3.064695 | val 3.388238 | tok/s 2625827.09 | norm 0.4037 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7707 | train 3.010730 | val 3.388238 | tok/s 2623791.86 | norm 0.3673 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7708 | train 3.082955 | val 3.388238 | tok/s 2628538.55 | norm 0.4157 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7709 | train 3.100109 | val 3.388238 | tok/s 2632617.30 | norm 0.4551 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7710 | train 2.985962 | val 3.388238 | tok/s 2631589.61 | norm 0.4703 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7711 | train 3.074612 | val 3.388238 | tok/s 2628919.64 | norm 0.4046 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7712 | train 3.045166 | val 3.388238 | tok/s 2628757.74 | norm 0.3843 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7713 | train 2.967242 | val 3.388238 | tok/s 2626196.38 | norm 0.3794 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7714 | train 3.054039 | val 3.388238 | tok/s 2629438.04 | norm 0.3914 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7715 | train 2.955018 | val 3.388238 | tok/s 2627285.13 | norm 0.3772 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7716 | train 3.137489 | val 3.388238 | tok/s 2616891.15 | norm 0.4646 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7717 | train 3.121360 | val 3.388238 | tok/s 2604104.15 | norm 0.4737 | dt 201.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7718 | train 3.045920 | val 3.388238 | tok/s 2626270.68 | norm 0.4525 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7719 | train 3.170226 | val 3.388238 | tok/s 2627217.55 | norm 0.4441 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7720 | train 2.935187 | val 3.388238 | tok/s 2633013.49 | norm 0.4594 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7721 | train 3.113424 | val 3.388238 | tok/s 2630890.43 | norm 0.4143 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7722 | train 3.051061 | val 3.388238 | tok/s 2621639.05 | norm 0.4162 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7723 | train 3.147952 | val 3.388238 | tok/s 2626448.16 | norm 0.4469 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7724 | train 3.048038 | val 3.388238 | tok/s 2627447.82 | norm 0.4536 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7725 | train 2.921703 | val 3.388238 | tok/s 2629662.81 | norm 0.4200 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7726 | train 3.008204 | val 3.388238 | tok/s 2631919.67 | norm 0.3799 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7727 | train 2.884541 | val 3.388238 | tok/s 2607465.21 | norm 0.3736 | dt 201.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7728 | train 2.972357 | val 3.388238 | tok/s 2629287.04 | norm 0.4026 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7729 | train 2.965803 | val 3.388238 | tok/s 2604844.28 | norm 0.4199 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7730 | train 2.993085 | val 3.388238 | tok/s 2627003.15 | norm 0.4088 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7731 | train 3.039002 | val 3.388238 | tok/s 2628237.04 | norm 0.4108 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7732 | train 3.054941 | val 3.388238 | tok/s 2628525.01 | norm 0.4075 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7733 | train 3.008264 | val 3.388238 | tok/s 2630560.46 | norm 0.3695 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7734 | train 3.023990 | val 3.388238 | tok/s 2625843.95 | norm 0.3593 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7735 | train 3.103091 | val 3.388238 | tok/s 2499217.44 | norm 0.4024 | dt 209.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7736 | train 3.046493 | val 3.388238 | tok/s 2628107.47 | norm 0.4048 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7737 | train 2.935511 | val 3.388238 | tok/s 2627638.14 | norm 0.3916 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7738 | train 2.970884 | val 3.388238 | tok/s 2632730.57 | norm 0.4152 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7739 | train 3.040256 | val 3.388238 | tok/s 2628805.24 | norm 0.4330 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7740 | train 3.039337 | val 3.388238 | tok/s 2621533.41 | norm 0.4223 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7741 | train 3.020025 | val 3.388238 | tok/s 2630469.57 | norm 0.4306 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7742 | train 3.033541 | val 3.388238 | tok/s 2623598.89 | norm 0.4574 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7743 | train 3.070868 | val 3.388238 | tok/s 2629662.48 | norm 0.4167 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7744 | train 2.960519 | val 3.388238 | tok/s 2547462.69 | norm 0.3882 | dt 205.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7745 | train 2.980257 | val 3.388238 | tok/s 2568392.93 | norm 0.4313 | dt 204.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7746 | train 2.993171 | val 3.388238 | tok/s 2628757.61 | norm 0.4161 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7747 | train 3.045640 | val 3.388238 | tok/s 2631331.95 | norm 0.3748 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7748 | train 3.034680 | val 3.388238 | tok/s 2630970.24 | norm 0.3995 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7749 | train 3.076277 | val 3.388238 | tok/s 2630178.57 | norm 0.3803 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 7750 | val_loss 3.389971 step 7750 | train 3.239069 | val 3.389971 | tok/s 2513473.23 | norm 0.4145 | dt 208.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7751 | train 3.050310 | val 3.389971 | tok/s 2575691.62 | norm 0.4198 | dt 203.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7752 | train 2.994945 | val 3.389971 | tok/s 2627090.06 | norm 0.3903 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7753 | train 3.017164 | val 3.389971 | tok/s 2627220.40 | norm 0.4324 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7754 | train 3.005520 | val 3.389971 | tok/s 2630954.21 | norm 0.4206 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7755 | train 3.028022 | val 3.389971 | tok/s 2627947.18 | norm 0.3902 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7756 | train 3.064230 | val 3.389971 | tok/s 2632929.55 | norm 0.4068 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7757 | train 3.025251 | val 3.389971 | tok/s 2627015.71 | norm 0.4138 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7758 | train 2.891383 | val 3.389971 | tok/s 2629125.97 | norm 0.4107 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7759 | train 2.947820 | val 3.389971 | tok/s 2629809.70 | norm 0.4122 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7760 | train 3.000402 | val 3.389971 | tok/s 2626569.86 | norm 0.3905 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7761 | train 3.003300 | val 3.389971 | tok/s 2626618.39 | norm 0.4198 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7762 | train 3.090379 | val 3.389971 | tok/s 2629686.72 | norm 0.3862 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7763 | train 3.024959 | val 3.389971 | tok/s 2628018.90 | norm 0.3692 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7764 | train 3.165377 | val 3.389971 | tok/s 2629088.16 | norm 0.4261 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7765 | train 3.009994 | val 3.389971 | tok/s 2628395.76 | norm 0.4221 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7766 | train 3.071365 | val 3.389971 | tok/s 2626815.13 | norm 0.4229 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7767 | train 3.035122 | val 3.389971 | tok/s 2627537.11 | norm 0.4459 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7768 | train 2.997189 | val 3.389971 | tok/s 2625365.45 | norm 0.4373 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7769 | train 3.018061 | val 3.389971 | tok/s 2627817.48 | norm 0.4483 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7770 | train 3.031341 | val 3.389971 | tok/s 2623065.79 | norm 0.4455 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7771 | train 2.933423 | val 3.389971 | tok/s 2622560.36 | norm 0.4224 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7772 | train 3.126989 | val 3.389971 | tok/s 2624906.44 | norm 0.4647 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7773 | train 2.951337 | val 3.389971 | tok/s 2632849.57 | norm 0.3975 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7774 | train 3.024991 | val 3.389971 | tok/s 2634906.22 | norm 0.3675 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7775 | train 3.033962 | val 3.389971 | tok/s 2628965.55 | norm 0.4270 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7776 | train 3.008193 | val 3.389971 | tok/s 2629397.31 | norm 0.4250 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7777 | train 2.997196 | val 3.389971 | tok/s 2628786.09 | norm 0.4236 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7778 | train 2.996615 | val 3.389971 | tok/s 2630446.65 | norm 0.4363 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7779 | train 3.047039 | val 3.389971 | tok/s 2629559.21 | norm 0.4256 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7780 | train 2.957219 | val 3.389971 | tok/s 2628177.64 | norm 0.4407 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7781 | train 3.089706 | val 3.389971 | tok/s 2631668.55 | norm 0.4598 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7782 | train 3.096377 | val 3.389971 | tok/s 2631868.79 | norm 0.4347 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7783 | train 3.062356 | val 3.389971 | tok/s 2630700.87 | norm 0.3940 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7784 | train 3.049388 | val 3.389971 | tok/s 2631718.10 | norm 0.3750 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7785 | train 3.014387 | val 3.389971 | tok/s 2627480.67 | norm 0.3813 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7786 | train 3.097681 | val 3.389971 | tok/s 2629691.46 | norm 0.3962 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7787 | train 3.005441 | val 3.389971 | tok/s 2629660.99 | norm 0.4098 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7788 | train 3.046077 | val 3.389971 | tok/s 2628846.28 | norm 0.4345 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7789 | train 3.115753 | val 3.389971 | tok/s 2632404.40 | norm 0.4197 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7790 | train 3.070668 | val 3.389971 | tok/s 2626176.15 | norm 0.4318 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7791 | train 3.162433 | val 3.389971 | tok/s 2627168.12 | norm 0.4329 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7792 | train 3.054114 | val 3.389971 | tok/s 2627814.13 | norm 0.4019 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7793 | train 2.941561 | val 3.389971 | tok/s 2626163.97 | norm 0.3983 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7794 | train 2.980573 | val 3.389971 | tok/s 2625079.38 | norm 0.4504 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7795 | train 3.048714 | val 3.389971 | tok/s 2628514.09 | norm 0.4405 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7796 | train 2.950872 | val 3.389971 | tok/s 2630478.22 | norm 0.4004 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7797 | train 2.955342 | val 3.389971 | tok/s 2631730.34 | norm 0.3989 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7798 | train 2.914346 | val 3.389971 | tok/s 2631401.09 | norm 0.3926 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7799 | train 2.949235 | val 3.389971 | tok/s 2630574.59 | norm 0.3816 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7800 | train 2.971562 | val 3.389971 | tok/s 2630442.25 | norm 0.3787 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7801 | train 3.023241 | val 3.389971 | tok/s 2544726.37 | norm 0.4170 | dt 206.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7802 | train 2.893389 | val 3.389971 | tok/s 2446869.72 | norm 0.4725 | dt 214.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7803 | train 2.917797 | val 3.389971 | tok/s 2624370.65 | norm 0.4578 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7804 | train 2.949768 | val 3.389971 | tok/s 2631313.21 | norm 0.4580 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7805 | train 3.112378 | val 3.389971 | tok/s 2626882.57 | norm 0.4151 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7806 | train 3.050640 | val 3.389971 | tok/s 2631996.23 | norm 0.4307 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7807 | train 3.049419 | val 3.389971 | tok/s 2625739.27 | norm 0.4275 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7808 | train 2.989049 | val 3.389971 | tok/s 2629828.35 | norm 0.4317 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7809 | train 2.932525 | val 3.389971 | tok/s 2593960.71 | norm 0.4335 | dt 202.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7810 | train 2.993941 | val 3.389971 | tok/s 2631019.61 | norm 0.4098 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7811 | train 3.130283 | val 3.389971 | tok/s 2604606.14 | norm 0.3509 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7812 | train 3.034571 | val 3.389971 | tok/s 2631569.42 | norm 0.3738 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7813 | train 3.013941 | val 3.389971 | tok/s 2630899.62 | norm 0.3614 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7814 | train 3.237892 | val 3.389971 | tok/s 2629860.30 | norm 0.3827 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7815 | train 3.062306 | val 3.389971 | tok/s 2629110.59 | norm 0.4469 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7816 | train 3.002464 | val 3.389971 | tok/s 2631109.49 | norm 0.4160 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7817 | train 3.127114 | val 3.389971 | tok/s 2630460.49 | norm 0.4310 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7818 | train 3.172172 | val 3.389971 | tok/s 2630886.57 | norm 0.4098 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7819 | train 2.837780 | val 3.389971 | tok/s 2632999.62 | norm 0.4417 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7820 | train 3.012199 | val 3.389971 | tok/s 2628087.02 | norm 0.4285 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7821 | train 2.944791 | val 3.389971 | tok/s 2631124.38 | norm 0.4193 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7822 | train 3.047258 | val 3.389971 | tok/s 2625391.14 | norm 0.4117 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7823 | train 2.948532 | val 3.389971 | tok/s 2628890.89 | norm 0.4264 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7824 | train 3.201697 | val 3.389971 | tok/s 2632004.11 | norm 0.4206 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7825 | train 3.130717 | val 3.389971 | tok/s 2628576.74 | norm 0.4009 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7826 | train 2.867934 | val 3.389971 | tok/s 2627884.01 | norm 0.4367 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7827 | train 2.999286 | val 3.389971 | tok/s 2623543.41 | norm 0.3994 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7828 | train 2.976060 | val 3.389971 | tok/s 2620547.47 | norm 0.3785 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7829 | train 3.006974 | val 3.389971 | tok/s 2627858.20 | norm 0.3961 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7830 | train 3.113809 | val 3.389971 | tok/s 2630388.95 | norm 0.4309 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7831 | train 3.125500 | val 3.389971 | tok/s 2630240.51 | norm 0.4396 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7832 | train 2.972874 | val 3.389971 | tok/s 2630772.44 | norm 0.4260 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7833 | train 3.041904 | val 3.389971 | tok/s 2630441.90 | norm 0.4317 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7834 | train 3.001589 | val 3.389971 | tok/s 2634035.36 | norm 0.4712 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7835 | train 2.957988 | val 3.389971 | tok/s 2626498.48 | norm 0.4385 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7836 | train 3.029766 | val 3.389971 | tok/s 2632403.80 | norm 0.4059 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7837 | train 3.087343 | val 3.389971 | tok/s 2629089.32 | norm 0.4205 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7838 | train 3.025223 | val 3.389971 | tok/s 2629552.79 | norm 0.4233 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7839 | train 3.014011 | val 3.389971 | tok/s 2625152.00 | norm 0.3849 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7840 | train 2.987696 | val 3.389971 | tok/s 2626144.12 | norm 0.3718 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7841 | train 2.966778 | val 3.389971 | tok/s 2630013.74 | norm 0.3810 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7842 | train 3.267383 | val 3.389971 | tok/s 2633389.98 | norm 0.3952 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7843 | train 3.062587 | val 3.389971 | tok/s 2631096.97 | norm 0.4264 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7844 | train 2.881225 | val 3.389971 | tok/s 2630623.44 | norm 0.4288 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7845 | train 2.967404 | val 3.389971 | tok/s 2630107.61 | norm 0.4415 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7846 | train 3.106596 | val 3.389971 | tok/s 2626108.83 | norm 0.4197 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7847 | train 3.022735 | val 3.389971 | tok/s 2576752.27 | norm 0.4337 | dt 203.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7848 | train 3.110432 | val 3.389971 | tok/s 2587623.22 | norm 0.4421 | dt 202.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7849 | train 2.962149 | val 3.389971 | tok/s 2625751.69 | norm 0.4313 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7850 | train 3.010970 | val 3.389971 | tok/s 2622752.69 | norm 0.4000 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7851 | train 3.061437 | val 3.389971 | tok/s 2630136.94 | norm 0.4289 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7852 | train 2.866576 | val 3.389971 | tok/s 2626393.34 | norm 0.4645 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7853 | train 2.980724 | val 3.389971 | tok/s 2627342.69 | norm 0.4646 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7854 | train 2.953026 | val 3.389971 | tok/s 2626863.18 | norm 0.4151 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7855 | train 2.971845 | val 3.389971 | tok/s 2470523.04 | norm 0.4268 | dt 212.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7856 | train 2.959219 | val 3.389971 | tok/s 2629248.68 | norm 0.4350 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7857 | train 3.068581 | val 3.389971 | tok/s 2629474.94 | norm 0.3988 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7858 | train 2.950771 | val 3.389971 | tok/s 2629922.02 | norm 0.4434 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7859 | train 2.964959 | val 3.389971 | tok/s 2627271.54 | norm 0.4608 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7860 | train 2.963301 | val 3.389971 | tok/s 2598504.47 | norm 0.4707 | dt 201.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7861 | train 3.139495 | val 3.389971 | tok/s 2632333.62 | norm 0.4139 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7862 | train 2.980268 | val 3.389971 | tok/s 2628756.24 | norm 0.3648 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7863 | train 2.829150 | val 3.389971 | tok/s 2631098.76 | norm 0.4258 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7864 | train 3.081054 | val 3.389971 | tok/s 2622637.27 | norm 0.4299 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7865 | train 3.039155 | val 3.389971 | tok/s 2602458.27 | norm 0.4007 | dt 201.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7866 | train 2.975025 | val 3.389971 | tok/s 2630296.31 | norm 0.4229 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7867 | train 3.057800 | val 3.389971 | tok/s 2632694.95 | norm 0.4366 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7868 | train 3.073212 | val 3.389971 | tok/s 2632999.21 | norm 0.4596 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7869 | train 3.042536 | val 3.389971 | tok/s 2630774.24 | norm 0.4515 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7870 | train 3.011062 | val 3.389971 | tok/s 2632075.27 | norm 0.4300 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7871 | train 2.958257 | val 3.389971 | tok/s 2621821.84 | norm 0.3803 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7872 | train 2.994935 | val 3.389971 | tok/s 2627122.10 | norm 0.3687 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7873 | train 3.048305 | val 3.389971 | tok/s 2633090.50 | norm 0.4064 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7874 | train 3.114743 | val 3.389971 | tok/s 2622119.72 | norm 0.4374 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7875 | train 2.988362 | val 3.389971 | tok/s 2626645.88 | norm 0.4092 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7876 | train 2.926576 | val 3.389971 | tok/s 2626571.84 | norm 0.4091 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7877 | train 3.091364 | val 3.389971 | tok/s 2624146.10 | norm 0.4369 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7878 | train 2.947432 | val 3.389971 | tok/s 2627331.91 | norm 0.3938 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7879 | train 3.049347 | val 3.389971 | tok/s 2626478.37 | norm 0.4082 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7880 | train 2.982107 | val 3.389971 | tok/s 2625999.79 | norm 0.4447 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7881 | train 3.066843 | val 3.389971 | tok/s 2630263.81 | norm 0.4557 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7882 | train 3.043798 | val 3.389971 | tok/s 2628933.31 | norm 0.4070 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7883 | train 3.007715 | val 3.389971 | tok/s 2628837.76 | norm 0.4041 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7884 | train 3.023454 | val 3.389971 | tok/s 2632115.35 | norm 0.4169 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7885 | train 3.087702 | val 3.389971 | tok/s 2632990.31 | norm 0.4281 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7886 | train 2.969820 | val 3.389971 | tok/s 2630720.39 | norm 0.4436 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7887 | train 3.015665 | val 3.389971 | tok/s 2631518.60 | norm 0.4179 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7888 | train 3.030119 | val 3.389971 | tok/s 2631821.62 | norm 0.3783 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7889 | train 3.120002 | val 3.389971 | tok/s 2625868.28 | norm 0.4125 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7890 | train 3.005134 | val 3.389971 | tok/s 2628804.72 | norm 0.4687 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7891 | train 3.043177 | val 3.389971 | tok/s 2629030.04 | norm 0.4477 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7892 | train 3.039169 | val 3.389971 | tok/s 2621563.74 | norm 0.4439 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7893 | train 3.132168 | val 3.389971 | tok/s 2630363.44 | norm 0.4529 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7894 | train 3.013807 | val 3.389971 | tok/s 2628909.69 | norm 0.4133 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7895 | train 3.066424 | val 3.389971 | tok/s 2629818.31 | norm 0.4022 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7896 | train 2.983797 | val 3.389971 | tok/s 2617910.34 | norm 0.4346 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7897 | train 2.965604 | val 3.389971 | tok/s 2628792.68 | norm 0.4211 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7898 | train 2.972015 | val 3.389971 | tok/s 2631644.39 | norm 0.4061 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7899 | train 3.031679 | val 3.389971 | tok/s 2476047.31 | norm 0.4286 | dt 211.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7900 | train 3.083527 | val 3.389971 | tok/s 2624974.16 | norm 0.4323 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7901 | train 2.975155 | val 3.389971 | tok/s 2620951.01 | norm 0.4421 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7902 | train 3.102484 | val 3.389971 | tok/s 2626504.10 | norm 0.4067 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7903 | train 3.135887 | val 3.389971 | tok/s 2633841.17 | norm 0.4245 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7904 | train 3.019319 | val 3.389971 | tok/s 2627377.94 | norm 0.4325 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7905 | train 2.985264 | val 3.389971 | tok/s 2632699.99 | norm 0.4869 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7906 | train 2.958413 | val 3.389971 | tok/s 2627717.08 | norm 0.3955 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7907 | train 2.992702 | val 3.389971 | tok/s 2628977.77 | norm 0.3950 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7908 | train 3.087846 | val 3.389971 | tok/s 2630838.99 | norm 0.4676 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7909 | train 3.103063 | val 3.389971 | tok/s 2628850.43 | norm 0.4108 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7910 | train 3.053535 | val 3.389971 | tok/s 2630882.12 | norm 0.3692 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7911 | train 2.999870 | val 3.389971 | tok/s 2627871.56 | norm 0.3810 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7912 | train 2.998965 | val 3.389971 | tok/s 2627867.46 | norm 0.3927 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7913 | train 3.129989 | val 3.389971 | tok/s 2632591.21 | norm 0.4389 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7914 | train 2.957433 | val 3.389971 | tok/s 2633785.36 | norm 0.4231 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7915 | train 3.095501 | val 3.389971 | tok/s 2630908.19 | norm 0.4208 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7916 | train 2.970742 | val 3.389971 | tok/s 2625386.10 | norm 0.4314 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7917 | train 2.985357 | val 3.389971 | tok/s 2628623.75 | norm 0.4486 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7918 | train 2.966697 | val 3.389971 | tok/s 2630116.85 | norm 0.4227 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7919 | train 2.980748 | val 3.389971 | tok/s 2607257.42 | norm 0.3952 | dt 201.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7920 | train 2.996035 | val 3.389971 | tok/s 2634440.93 | norm 0.4037 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7921 | train 3.109915 | val 3.389971 | tok/s 2626666.63 | norm 0.4427 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7922 | train 3.036258 | val 3.389971 | tok/s 2625974.60 | norm 0.4196 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7923 | train 3.044117 | val 3.389971 | tok/s 2631523.82 | norm 0.4241 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7924 | train 3.019815 | val 3.389971 | tok/s 2627463.07 | norm 0.4610 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7925 | train 3.054836 | val 3.389971 | tok/s 2625400.23 | norm 0.4575 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7926 | train 3.015895 | val 3.389971 | tok/s 2625498.65 | norm 0.4312 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7927 | train 2.988779 | val 3.389971 | tok/s 2628216.08 | norm 0.4098 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7928 | train 2.939467 | val 3.389971 | tok/s 2630876.31 | norm 0.3855 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7929 | train 3.008622 | val 3.389971 | tok/s 2625314.86 | norm 0.3871 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7930 | train 2.956203 | val 3.389971 | tok/s 2631560.75 | norm 0.4811 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7931 | train 3.079091 | val 3.389971 | tok/s 2629111.58 | norm 0.4383 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7932 | train 3.219438 | val 3.389971 | tok/s 2628903.17 | norm 0.4384 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7933 | train 3.055664 | val 3.389971 | tok/s 2629371.22 | norm 0.4462 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7934 | train 2.938334 | val 3.389971 | tok/s 2626838.99 | norm 0.3939 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7935 | train 2.923376 | val 3.389971 | tok/s 2631941.18 | norm 0.3617 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7936 | train 2.993390 | val 3.389971 | tok/s 2629397.25 | norm 0.3663 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7937 | train 2.997452 | val 3.389971 | tok/s 2628644.48 | norm 0.4189 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7938 | train 3.018857 | val 3.389971 | tok/s 2628565.00 | norm 0.4474 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7939 | train 3.175875 | val 3.389971 | tok/s 2518209.91 | norm 0.4309 | dt 208.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7940 | train 3.059165 | val 3.389971 | tok/s 2628474.14 | norm 0.4962 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7941 | train 3.077269 | val 3.389971 | tok/s 2624478.82 | norm 0.4023 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7942 | train 3.037651 | val 3.389971 | tok/s 2630064.31 | norm 0.3917 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7943 | train 2.911739 | val 3.389971 | tok/s 2622579.72 | norm 0.3977 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7944 | train 3.109330 | val 3.389971 | tok/s 2621807.67 | norm 0.4291 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7945 | train 2.855010 | val 3.389971 | tok/s 2627829.19 | norm 0.4215 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7946 | train 3.010485 | val 3.389971 | tok/s 2626905.61 | norm 0.4080 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7947 | train 3.135190 | val 3.389971 | tok/s 2627822.42 | norm 0.4349 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7948 | train 3.115197 | val 3.389971 | tok/s 2473435.44 | norm 0.3958 | dt 211.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7949 | train 3.050386 | val 3.389971 | tok/s 2581955.31 | norm 0.3727 | dt 203.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7950 | train 3.040509 | val 3.389971 | tok/s 2628864.62 | norm 0.3903 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7951 | train 3.033305 | val 3.389971 | tok/s 2627375.24 | norm 0.4190 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7952 | train 3.163767 | val 3.389971 | tok/s 2630212.68 | norm 0.4508 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7953 | train 2.983094 | val 3.389971 | tok/s 2628515.65 | norm 0.3957 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7954 | train 3.094234 | val 3.389971 | tok/s 2629525.01 | norm 0.4010 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7955 | train 3.094818 | val 3.389971 | tok/s 2631800.12 | norm 0.4144 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7956 | train 3.049109 | val 3.389971 | tok/s 2627353.11 | norm 0.3575 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7957 | train 3.120625 | val 3.389971 | tok/s 2628219.07 | norm 0.3407 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7958 | train 3.057445 | val 3.389971 | tok/s 2628308.40 | norm 0.4074 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7959 | train 3.072826 | val 3.389971 | tok/s 2629514.87 | norm 0.4226 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7960 | train 3.081584 | val 3.389971 | tok/s 2630386.44 | norm 0.5733 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7961 | train 3.012799 | val 3.389971 | tok/s 2628826.29 | norm 0.3954 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7962 | train 3.158137 | val 3.389971 | tok/s 2629933.07 | norm 0.4333 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7963 | train 2.965739 | val 3.389971 | tok/s 2624834.99 | norm 0.4108 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7964 | train 3.182638 | val 3.389971 | tok/s 2627507.51 | norm 0.3883 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7965 | train 2.923654 | val 3.389971 | tok/s 2633322.29 | norm 0.4078 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7966 | train 3.131205 | val 3.389971 | tok/s 2627019.53 | norm 0.4116 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7967 | train 2.994154 | val 3.389971 | tok/s 2628493.22 | norm 0.4148 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7968 | train 3.060801 | val 3.389971 | tok/s 2629848.37 | norm 0.4091 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7969 | train 2.964619 | val 3.389971 | tok/s 2630930.48 | norm 0.4065 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7970 | train 3.078632 | val 3.389971 | tok/s 2630304.60 | norm 0.4144 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7971 | train 3.004955 | val 3.389971 | tok/s 2626700.37 | norm 0.4028 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7972 | train 3.109595 | val 3.389971 | tok/s 2631177.88 | norm 0.5193 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7973 | train 2.979508 | val 3.389971 | tok/s 2628338.05 | norm 0.4250 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7974 | train 3.064654 | val 3.389971 | tok/s 2631300.46 | norm 0.3835 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7975 | train 3.042235 | val 3.389971 | tok/s 2633229.40 | norm 0.4367 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7976 | train 2.916857 | val 3.389971 | tok/s 2631504.53 | norm 0.4048 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7977 | train 3.082914 | val 3.389971 | tok/s 2630106.07 | norm 0.3806 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7978 | train 3.326058 | val 3.389971 | tok/s 2630365.14 | norm 0.4935 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7979 | train 3.068310 | val 3.389971 | tok/s 2632023.34 | norm 0.4558 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7980 | train 3.031346 | val 3.389971 | tok/s 2631859.21 | norm 0.4054 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7981 | train 3.045029 | val 3.389971 | tok/s 2629545.44 | norm 0.4110 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7982 | train 3.102978 | val 3.389971 | tok/s 2629351.59 | norm 0.4462 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7983 | train 3.002562 | val 3.389971 | tok/s 2627583.26 | norm 0.4148 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7984 | train 2.991423 | val 3.389971 | tok/s 2628807.19 | norm 0.3968 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7985 | train 3.017568 | val 3.389971 | tok/s 2630122.63 | norm 0.4173 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7986 | train 3.057141 | val 3.389971 | tok/s 2632075.41 | norm 0.4147 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7987 | train 3.065477 | val 3.389971 | tok/s 2634876.22 | norm 0.3909 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7988 | train 3.046785 | val 3.389971 | tok/s 2629815.91 | norm 0.4134 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7989 | train 2.993149 | val 3.389971 | tok/s 2627539.56 | norm 0.4029 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7990 | train 3.012284 | val 3.389971 | tok/s 2629217.16 | norm 0.3915 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7991 | train 2.957891 | val 3.389971 | tok/s 2630197.40 | norm 0.4338 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7992 | train 2.934417 | val 3.389971 | tok/s 2625420.62 | norm 0.4439 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7993 | train 3.025965 | val 3.389971 | tok/s 2542991.16 | norm 0.4351 | dt 206.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7994 | train 3.002813 | val 3.389971 | tok/s 2624470.91 | norm 0.4191 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7995 | train 2.967956 | val 3.389971 | tok/s 2623447.17 | norm 0.3949 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7996 | train 3.113580 | val 3.389971 | tok/s 2624942.64 | norm 0.4078 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7997 | train 2.962307 | val 3.389971 | tok/s 2629728.13 | norm 0.4240 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7998 | train 3.000877 | val 3.389971 | tok/s 2622935.69 | norm 0.4299 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 7999 | train 3.020015 | val 3.389971 | tok/s 2488971.13 | norm 0.3868 | dt 210.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 8000 | val_loss 3.385586 step 8000 | train 3.026077 | val 3.385586 | tok/s 2581820.39 | norm 0.3999 | dt 203.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8001 | train 2.897603 | val 3.385586 | tok/s 2623176.80 | norm 0.4089 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8002 | train 3.139513 | val 3.385586 | tok/s 2627141.74 | norm 0.4219 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8003 | train 3.027364 | val 3.385586 | tok/s 2624722.33 | norm 0.4316 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8004 | train 3.034166 | val 3.385586 | tok/s 2628243.55 | norm 0.4722 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8005 | train 2.939654 | val 3.385586 | tok/s 2630396.82 | norm 0.4277 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8006 | train 3.042581 | val 3.385586 | tok/s 2462085.16 | norm 0.3950 | dt 212.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8007 | train 3.015574 | val 3.385586 | tok/s 2624340.55 | norm 0.4431 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8008 | train 3.075566 | val 3.385586 | tok/s 2587424.48 | norm 0.4090 | dt 202.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8009 | train 3.052582 | val 3.385586 | tok/s 2625595.06 | norm 0.4082 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8010 | train 3.084272 | val 3.385586 | tok/s 2631935.42 | norm 0.4182 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8011 | train 3.070695 | val 3.385586 | tok/s 2629290.93 | norm 0.4221 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8012 | train 3.074382 | val 3.385586 | tok/s 2628643.65 | norm 0.5532 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8013 | train 2.993190 | val 3.385586 | tok/s 2631466.06 | norm 0.4077 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8014 | train 2.957227 | val 3.385586 | tok/s 2630420.73 | norm 0.3245 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8015 | train 3.057434 | val 3.385586 | tok/s 2628850.90 | norm 0.4067 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8016 | train 3.069047 | val 3.385586 | tok/s 2604635.85 | norm 0.4084 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8017 | train 3.139346 | val 3.385586 | tok/s 2622606.10 | norm 0.3851 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8018 | train 3.079605 | val 3.385586 | tok/s 2627696.11 | norm 0.4064 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8019 | train 3.117514 | val 3.385586 | tok/s 2623600.93 | norm 0.4342 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8020 | train 2.927160 | val 3.385586 | tok/s 2618837.54 | norm 0.4216 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8021 | train 2.955070 | val 3.385586 | tok/s 2491637.27 | norm 0.4563 | dt 210.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8022 | train 3.094383 | val 3.385586 | tok/s 2625790.04 | norm 0.4717 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8023 | train 2.977133 | val 3.385586 | tok/s 2630961.91 | norm 0.3994 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8024 | train 2.966325 | val 3.385586 | tok/s 2620720.82 | norm 0.4117 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8025 | train 2.925565 | val 3.385586 | tok/s 2627323.62 | norm 0.4629 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8026 | train 2.824872 | val 3.385586 | tok/s 2625565.34 | norm 0.4508 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8027 | train 3.109579 | val 3.385586 | tok/s 2628971.06 | norm 0.4085 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8028 | train 3.061743 | val 3.385586 | tok/s 2630383.59 | norm 0.3950 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8029 | train 2.930087 | val 3.385586 | tok/s 2537817.57 | norm 0.4095 | dt 206.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8030 | train 3.008938 | val 3.385586 | tok/s 2626963.22 | norm 0.4812 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8031 | train 3.096036 | val 3.385586 | tok/s 2630713.55 | norm 0.4334 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8032 | train 3.027299 | val 3.385586 | tok/s 2627943.70 | norm 0.4154 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8033 | train 3.029753 | val 3.385586 | tok/s 2630249.89 | norm 0.4223 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8034 | train 3.001127 | val 3.385586 | tok/s 2630385.85 | norm 0.4020 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8035 | train 2.960315 | val 3.385586 | tok/s 2626080.93 | norm 0.4317 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8036 | train 3.027971 | val 3.385586 | tok/s 2617653.26 | norm 0.4571 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8037 | train 3.063239 | val 3.385586 | tok/s 2627511.38 | norm 0.4418 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8038 | train 2.928595 | val 3.385586 | tok/s 2628924.19 | norm 0.4203 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8039 | train 2.964978 | val 3.385586 | tok/s 2628647.65 | norm 0.4210 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8040 | train 2.939810 | val 3.385586 | tok/s 2629782.18 | norm 0.3999 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8041 | train 2.893667 | val 3.385586 | tok/s 2625620.93 | norm 0.4109 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8042 | train 2.967767 | val 3.385586 | tok/s 2626105.95 | norm 0.4155 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8043 | train 3.011915 | val 3.385586 | tok/s 2625153.88 | norm 0.3932 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8044 | train 3.095588 | val 3.385586 | tok/s 2628168.61 | norm 0.3939 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8045 | train 3.037406 | val 3.385586 | tok/s 2620476.71 | norm 0.3899 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8046 | train 3.056034 | val 3.385586 | tok/s 2626921.52 | norm 0.3731 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8047 | train 3.010806 | val 3.385586 | tok/s 2630019.74 | norm 0.4238 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8048 | train 3.107430 | val 3.385586 | tok/s 2627849.92 | norm 0.4810 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8049 | train 3.037026 | val 3.385586 | tok/s 2629795.37 | norm 0.4132 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8050 | train 3.021891 | val 3.385586 | tok/s 2633342.00 | norm 0.3808 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8051 | train 2.919049 | val 3.385586 | tok/s 2633796.03 | norm 0.3927 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8052 | train 2.985610 | val 3.385586 | tok/s 2630710.66 | norm 0.4136 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8053 | train 2.993503 | val 3.385586 | tok/s 2627494.42 | norm 0.4490 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8054 | train 2.976373 | val 3.385586 | tok/s 2630166.42 | norm 0.3898 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8055 | train 3.058192 | val 3.385586 | tok/s 2628679.26 | norm 0.4010 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8056 | train 2.960595 | val 3.385586 | tok/s 2628691.28 | norm 0.4327 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8057 | train 3.085554 | val 3.385586 | tok/s 2629745.02 | norm 0.4242 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8058 | train 2.965511 | val 3.385586 | tok/s 2628906.44 | norm 0.4378 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8059 | train 2.996279 | val 3.385586 | tok/s 2625009.65 | norm 0.4268 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8060 | train 2.974366 | val 3.385586 | tok/s 2625576.86 | norm 0.4522 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8061 | train 3.051657 | val 3.385586 | tok/s 2630186.44 | norm 0.4133 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8062 | train 3.001302 | val 3.385586 | tok/s 2628637.01 | norm 0.3982 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8063 | train 2.985174 | val 3.385586 | tok/s 2633135.82 | norm 0.4115 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8064 | train 3.112470 | val 3.385586 | tok/s 2633827.72 | norm 0.3845 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8065 | train 3.079441 | val 3.385586 | tok/s 2630157.92 | norm 0.3994 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8066 | train 3.056666 | val 3.385586 | tok/s 2630294.41 | norm 0.4399 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8067 | train 2.984425 | val 3.385586 | tok/s 2628208.23 | norm 0.4499 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8068 | train 3.156192 | val 3.385586 | tok/s 2624105.48 | norm 0.4102 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8069 | train 3.043597 | val 3.385586 | tok/s 2628787.00 | norm 0.4104 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8070 | train 2.967536 | val 3.385586 | tok/s 2625765.16 | norm 0.4384 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8071 | train 2.957029 | val 3.385586 | tok/s 2627504.69 | norm 0.4371 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8072 | train 3.246974 | val 3.385586 | tok/s 2623889.34 | norm 0.4306 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8073 | train 2.965957 | val 3.385586 | tok/s 2627412.27 | norm 0.4150 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8074 | train 3.082286 | val 3.385586 | tok/s 2625713.23 | norm 0.4065 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8075 | train 3.037318 | val 3.385586 | tok/s 2629284.50 | norm 0.4489 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8076 | train 3.129559 | val 3.385586 | tok/s 2630254.90 | norm 0.4291 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8077 | train 3.053387 | val 3.385586 | tok/s 2624338.56 | norm 0.4081 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8078 | train 2.906991 | val 3.385586 | tok/s 2624322.53 | norm 0.4331 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8079 | train 2.925830 | val 3.385586 | tok/s 2629555.40 | norm 0.4033 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8080 | train 3.289414 | val 3.385586 | tok/s 2630556.13 | norm 0.4455 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8081 | train 2.964160 | val 3.385586 | tok/s 2628454.21 | norm 0.4175 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8082 | train 2.978203 | val 3.385586 | tok/s 2620117.93 | norm 0.4094 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8083 | train 2.944144 | val 3.385586 | tok/s 2626872.90 | norm 0.4096 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8084 | train 3.052501 | val 3.385586 | tok/s 2603445.70 | norm 0.4275 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8085 | train 2.830279 | val 3.385586 | tok/s 2626272.26 | norm 0.4236 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8086 | train 3.086334 | val 3.385586 | tok/s 2623839.15 | norm 0.3921 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8087 | train 3.042021 | val 3.385586 | tok/s 2625043.91 | norm 0.3810 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8088 | train 2.946489 | val 3.385586 | tok/s 2629496.74 | norm 0.4006 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8089 | train 2.942835 | val 3.385586 | tok/s 2627039.05 | norm 0.3716 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8090 | train 2.926324 | val 3.385586 | tok/s 2603618.40 | norm 0.3829 | dt 201.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8091 | train 2.977525 | val 3.385586 | tok/s 2633395.72 | norm 0.4075 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8092 | train 2.972251 | val 3.385586 | tok/s 2631753.91 | norm 0.4150 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8093 | train 3.200393 | val 3.385586 | tok/s 2627990.15 | norm 0.4769 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8094 | train 2.977525 | val 3.385586 | tok/s 2629341.92 | norm 0.5053 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8095 | train 2.883225 | val 3.385586 | tok/s 2616520.89 | norm 0.4166 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8096 | train 2.916668 | val 3.385586 | tok/s 2626622.80 | norm 0.4297 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8097 | train 3.140794 | val 3.385586 | tok/s 2625239.93 | norm 0.4526 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8098 | train 2.937095 | val 3.385586 | tok/s 2600151.30 | norm 0.4312 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8099 | train 3.171056 | val 3.385586 | tok/s 2627540.32 | norm 0.4030 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8100 | train 2.887149 | val 3.385586 | tok/s 2632043.93 | norm 0.3698 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8101 | train 2.866705 | val 3.385586 | tok/s 2626917.39 | norm 0.3963 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8102 | train 2.892757 | val 3.385586 | tok/s 2624466.44 | norm 0.4115 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8103 | train 3.001995 | val 3.385586 | tok/s 2627627.87 | norm 0.3917 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8104 | train 2.971512 | val 3.385586 | tok/s 2634271.28 | norm 0.4040 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8105 | train 3.076915 | val 3.385586 | tok/s 2627876.18 | norm 0.4388 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8106 | train 3.039146 | val 3.385586 | tok/s 2631101.91 | norm 0.4360 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8107 | train 3.061089 | val 3.385586 | tok/s 2634063.80 | norm 0.4215 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8108 | train 3.037539 | val 3.385586 | tok/s 2629856.32 | norm 0.4603 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8109 | train 2.938162 | val 3.385586 | tok/s 2632624.96 | norm 0.3878 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8110 | train 2.968264 | val 3.385586 | tok/s 2629695.55 | norm 0.3874 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8111 | train 2.915421 | val 3.385586 | tok/s 2629650.08 | norm 0.4011 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8112 | train 3.052667 | val 3.385586 | tok/s 2632072.68 | norm 0.4084 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8113 | train 3.091498 | val 3.385586 | tok/s 2634165.03 | norm 0.3970 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8114 | train 3.018377 | val 3.385586 | tok/s 2633046.18 | norm 0.4095 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8115 | train 2.958892 | val 3.385586 | tok/s 2632758.64 | norm 0.4161 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8116 | train 3.049718 | val 3.385586 | tok/s 2628360.66 | norm 0.3902 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8117 | train 2.980494 | val 3.385586 | tok/s 2630304.88 | norm 0.3914 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8118 | train 3.053031 | val 3.385586 | tok/s 2626838.61 | norm 0.3867 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8119 | train 3.071234 | val 3.385586 | tok/s 2627345.83 | norm 0.3759 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8120 | train 3.023800 | val 3.385586 | tok/s 2624148.07 | norm 0.4085 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8121 | train 3.136756 | val 3.385586 | tok/s 2630590.38 | norm 0.4451 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8122 | train 3.095410 | val 3.385586 | tok/s 2627714.63 | norm 0.4226 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8123 | train 2.977076 | val 3.385586 | tok/s 2628507.91 | norm 0.3901 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8124 | train 3.035760 | val 3.385586 | tok/s 2626028.47 | norm 0.4080 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8125 | train 3.034874 | val 3.385586 | tok/s 2627059.71 | norm 0.3992 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8126 | train 3.000548 | val 3.385586 | tok/s 2629442.06 | norm 0.3777 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8127 | train 3.061458 | val 3.385586 | tok/s 2625704.10 | norm 0.4042 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8128 | train 3.109886 | val 3.385586 | tok/s 2628226.04 | norm 0.4859 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8129 | train 3.077801 | val 3.385586 | tok/s 2627551.75 | norm 0.4125 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8130 | train 2.979809 | val 3.385586 | tok/s 2630137.50 | norm 0.3817 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8131 | train 3.020066 | val 3.385586 | tok/s 2626753.54 | norm 0.3992 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8132 | train 3.004380 | val 3.385586 | tok/s 2625525.50 | norm 0.4163 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8133 | train 3.015222 | val 3.385586 | tok/s 2628111.61 | norm 0.4320 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8134 | train 2.951787 | val 3.385586 | tok/s 2441348.44 | norm 0.4190 | dt 214.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8135 | train 2.967056 | val 3.385586 | tok/s 2501857.87 | norm 0.3909 | dt 209.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8136 | train 3.035994 | val 3.385586 | tok/s 2625285.27 | norm 0.3951 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8137 | train 2.864424 | val 3.385586 | tok/s 2626652.78 | norm 0.4068 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8138 | train 2.944778 | val 3.385586 | tok/s 2628269.62 | norm 0.4313 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8139 | train 3.049043 | val 3.385586 | tok/s 2624413.43 | norm 0.4219 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8140 | train 2.967066 | val 3.385586 | tok/s 2620572.58 | norm 0.4051 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8141 | train 3.049761 | val 3.385586 | tok/s 2624684.15 | norm 0.4520 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8142 | train 3.059370 | val 3.385586 | tok/s 2620948.77 | norm 0.4527 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8143 | train 3.057354 | val 3.385586 | tok/s 2624990.29 | norm 0.4424 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8144 | train 3.094856 | val 3.385586 | tok/s 2632633.19 | norm 0.4494 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8145 | train 2.953906 | val 3.385586 | tok/s 2630127.34 | norm 0.4056 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8146 | train 3.025875 | val 3.385586 | tok/s 2630782.36 | norm 0.3919 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8147 | train 2.984034 | val 3.385586 | tok/s 2626519.39 | norm 0.4177 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8148 | train 2.925583 | val 3.385586 | tok/s 2630644.24 | norm 0.4504 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8149 | train 2.916938 | val 3.385586 | tok/s 2632400.91 | norm 0.4282 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8150 | train 3.000176 | val 3.385586 | tok/s 2632475.98 | norm 0.3879 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8151 | train 2.995350 | val 3.385586 | tok/s 2632056.37 | norm 0.3998 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8152 | train 3.047677 | val 3.385586 | tok/s 2626473.02 | norm 0.4148 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8153 | train 2.951462 | val 3.385586 | tok/s 2632645.06 | norm 0.4185 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8154 | train 3.086390 | val 3.385586 | tok/s 2627583.33 | norm 0.4168 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8155 | train 2.974838 | val 3.385586 | tok/s 2627275.11 | norm 0.4063 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8156 | train 2.913606 | val 3.385586 | tok/s 2628543.78 | norm 0.4036 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8157 | train 3.136838 | val 3.385586 | tok/s 2631988.50 | norm 0.4466 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8158 | train 2.934766 | val 3.385586 | tok/s 2631435.34 | norm 0.4280 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8159 | train 3.004594 | val 3.385586 | tok/s 2623591.54 | norm 0.3977 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8160 | train 3.000066 | val 3.385586 | tok/s 2628979.56 | norm 0.3813 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8161 | train 2.992376 | val 3.385586 | tok/s 2624714.82 | norm 0.3974 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8162 | train 3.311823 | val 3.385586 | tok/s 2625599.82 | norm 0.4375 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8163 | train 3.156678 | val 3.385586 | tok/s 2631645.64 | norm 0.4384 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8164 | train 3.061461 | val 3.385586 | tok/s 2627279.63 | norm 0.3929 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8165 | train 3.124587 | val 3.385586 | tok/s 2618490.28 | norm 0.3988 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8166 | train 3.149048 | val 3.385586 | tok/s 2626143.92 | norm 0.3832 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8167 | train 3.001633 | val 3.385586 | tok/s 2619717.39 | norm 0.3860 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8168 | train 3.087851 | val 3.385586 | tok/s 2628375.02 | norm 0.3915 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8169 | train 3.249725 | val 3.385586 | tok/s 2628428.81 | norm 0.3893 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8170 | train 3.055140 | val 3.385586 | tok/s 2627972.90 | norm 0.4148 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8171 | train 3.062456 | val 3.385586 | tok/s 2628008.22 | norm 0.3838 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8172 | train 2.958239 | val 3.385586 | tok/s 2624449.39 | norm 0.3822 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8173 | train 3.009181 | val 3.385586 | tok/s 2627722.12 | norm 0.4218 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8174 | train 3.121347 | val 3.385586 | tok/s 2622626.00 | norm 0.3969 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8175 | train 3.093041 | val 3.385586 | tok/s 2627053.78 | norm 0.4471 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8176 | train 3.198175 | val 3.385586 | tok/s 2500901.57 | norm 0.4171 | dt 209.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8177 | train 3.040318 | val 3.385586 | tok/s 2604660.62 | norm 0.4077 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8178 | train 2.986615 | val 3.385586 | tok/s 2626186.42 | norm 0.4344 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8179 | train 3.035181 | val 3.385586 | tok/s 2626429.03 | norm 0.4121 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8180 | train 2.921163 | val 3.385586 | tok/s 2624673.58 | norm 0.4023 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8181 | train 3.005516 | val 3.385586 | tok/s 2592365.07 | norm 0.4141 | dt 202.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8182 | train 2.962332 | val 3.385586 | tok/s 2624286.09 | norm 0.4019 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8183 | train 3.004544 | val 3.385586 | tok/s 2631971.72 | norm 0.4104 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8184 | train 3.144609 | val 3.385586 | tok/s 2536117.33 | norm 0.4907 | dt 206.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8185 | train 3.003637 | val 3.385586 | tok/s 2622482.46 | norm 0.3967 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8186 | train 3.009350 | val 3.385586 | tok/s 2618746.25 | norm 0.4091 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8187 | train 2.986629 | val 3.385586 | tok/s 2589934.98 | norm 0.3754 | dt 202.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8188 | train 3.086699 | val 3.385586 | tok/s 2628563.09 | norm 0.4309 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8189 | train 3.068441 | val 3.385586 | tok/s 2624796.14 | norm 0.4494 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8190 | train 3.024311 | val 3.385586 | tok/s 2628025.23 | norm 0.3881 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8191 | train 3.062170 | val 3.385586 | tok/s 2620424.23 | norm 0.3946 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8192 | train 2.990857 | val 3.385586 | tok/s 2622509.47 | norm 0.4027 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8193 | train 2.894981 | val 3.385586 | tok/s 2628653.04 | norm 0.4388 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8194 | train 3.086198 | val 3.385586 | tok/s 2627224.98 | norm 0.4278 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8195 | train 2.935101 | val 3.385586 | tok/s 2625722.11 | norm 0.4439 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8196 | train 3.050487 | val 3.385586 | tok/s 2624337.23 | norm 0.4257 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8197 | train 3.043390 | val 3.385586 | tok/s 2629301.43 | norm 0.4217 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8198 | train 3.013731 | val 3.385586 | tok/s 2628148.65 | norm 0.4300 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8199 | train 2.977089 | val 3.385586 | tok/s 2625196.00 | norm 0.4397 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8200 | train 3.064659 | val 3.385586 | tok/s 2630262.85 | norm 0.4224 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8201 | train 3.031012 | val 3.385586 | tok/s 2626510.11 | norm 0.4043 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8202 | train 3.087754 | val 3.385586 | tok/s 2626437.26 | norm 0.4193 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8203 | train 3.050904 | val 3.385586 | tok/s 2627942.54 | norm 0.4121 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8204 | train 3.164755 | val 3.385586 | tok/s 2622106.21 | norm 0.3780 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8205 | train 2.961304 | val 3.385586 | tok/s 2629079.75 | norm 0.4040 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8206 | train 2.986895 | val 3.385586 | tok/s 2627453.43 | norm 0.4178 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8207 | train 3.078057 | val 3.385586 | tok/s 2631828.59 | norm 0.4212 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8208 | train 3.059438 | val 3.385586 | tok/s 2626771.18 | norm 0.3813 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8209 | train 3.268024 | val 3.385586 | tok/s 2628249.16 | norm 0.4340 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8210 | train 2.995407 | val 3.385586 | tok/s 2626083.12 | norm 0.4450 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8211 | train 2.933058 | val 3.385586 | tok/s 2628068.25 | norm 0.4528 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8212 | train 3.160243 | val 3.385586 | tok/s 2632868.82 | norm 0.4776 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8213 | train 3.054231 | val 3.385586 | tok/s 2625121.26 | norm 0.4486 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8214 | train 2.952610 | val 3.385586 | tok/s 2623368.70 | norm 0.4213 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8215 | train 3.028901 | val 3.385586 | tok/s 2623603.96 | norm 0.4172 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8216 | train 3.036722 | val 3.385586 | tok/s 2631203.24 | norm 0.4382 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8217 | train 3.053851 | val 3.385586 | tok/s 2628452.13 | norm 0.3938 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8218 | train 2.935408 | val 3.385586 | tok/s 2627950.22 | norm 0.4219 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8219 | train 3.022570 | val 3.385586 | tok/s 2627160.17 | norm 0.4285 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8220 | train 3.057183 | val 3.385586 | tok/s 2628767.79 | norm 0.4144 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8221 | train 3.006680 | val 3.385586 | tok/s 2622458.97 | norm 0.4186 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8222 | train 3.081440 | val 3.385586 | tok/s 2626168.02 | norm 0.4136 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8223 | train 3.019204 | val 3.385586 | tok/s 2579924.74 | norm 0.4071 | dt 203.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8224 | train 3.073396 | val 3.385586 | tok/s 2627877.60 | norm 0.3906 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8225 | train 3.123767 | val 3.385586 | tok/s 2629620.25 | norm 0.4043 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8226 | train 2.951044 | val 3.385586 | tok/s 2625887.39 | norm 0.4676 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8227 | train 3.098055 | val 3.385586 | tok/s 2627520.31 | norm 0.4067 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8228 | train 3.072406 | val 3.385586 | tok/s 2632937.45 | norm 0.3908 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8229 | train 2.985442 | val 3.385586 | tok/s 2627419.61 | norm 0.3827 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8230 | train 2.931129 | val 3.385586 | tok/s 2626707.40 | norm 0.3907 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8231 | train 2.997934 | val 3.385586 | tok/s 2500942.05 | norm 0.3940 | dt 209.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8232 | train 3.031406 | val 3.385586 | tok/s 2625222.45 | norm 0.3998 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8233 | train 2.972275 | val 3.385586 | tok/s 2630104.93 | norm 0.4277 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8234 | train 3.159105 | val 3.385586 | tok/s 2629890.85 | norm 0.4527 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8235 | train 3.033297 | val 3.385586 | tok/s 2630896.41 | norm 0.4179 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8236 | train 3.039410 | val 3.385586 | tok/s 2627049.14 | norm 0.3968 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8237 | train 2.845730 | val 3.385586 | tok/s 2625712.78 | norm 0.4497 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8238 | train 3.062717 | val 3.385586 | tok/s 2619935.99 | norm 0.4613 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8239 | train 3.055399 | val 3.385586 | tok/s 2564957.34 | norm 0.4411 | dt 204.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8240 | train 2.952512 | val 3.385586 | tok/s 2630207.01 | norm 0.4221 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8241 | train 2.965981 | val 3.385586 | tok/s 2624823.52 | norm 0.4496 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8242 | train 3.076102 | val 3.385586 | tok/s 2627277.15 | norm 0.4644 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8243 | train 3.216375 | val 3.385586 | tok/s 2622668.09 | norm 0.4347 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8244 | train 3.030966 | val 3.385586 | tok/s 2625116.28 | norm 0.4040 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8245 | train 2.898358 | val 3.385586 | tok/s 2628848.57 | norm 0.3791 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8246 | train 3.053656 | val 3.385586 | tok/s 2625435.48 | norm 0.3886 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8247 | train 2.999064 | val 3.385586 | tok/s 2630668.32 | norm 0.4116 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8248 | train 3.085602 | val 3.385586 | tok/s 2621140.91 | norm 0.3735 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8249 | train 2.910559 | val 3.385586 | tok/s 2622041.39 | norm 0.3842 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 8250 | val_loss 3.381412 step 8250 | train 2.971795 | val 3.381412 | tok/s 2617165.20 | norm 0.4061 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8251 | train 3.053933 | val 3.381412 | tok/s 2631286.39 | norm 0.4645 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8252 | train 2.936120 | val 3.381412 | tok/s 2631849.49 | norm 0.4473 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8253 | train 2.876565 | val 3.381412 | tok/s 2631133.79 | norm 0.4159 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8254 | train 3.003197 | val 3.381412 | tok/s 2628326.10 | norm 0.4174 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8255 | train 3.002805 | val 3.381412 | tok/s 2628678.32 | norm 0.4202 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8256 | train 2.981118 | val 3.381412 | tok/s 2628635.14 | norm 0.4082 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8257 | train 2.957075 | val 3.381412 | tok/s 2628384.32 | norm 0.4413 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8258 | train 3.048210 | val 3.381412 | tok/s 2628133.07 | norm 0.4358 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8259 | train 2.986840 | val 3.381412 | tok/s 2624921.89 | norm 0.3936 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8260 | train 2.942983 | val 3.381412 | tok/s 2619002.68 | norm 0.3576 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8261 | train 3.002961 | val 3.381412 | tok/s 2626427.13 | norm 0.4120 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8262 | train 2.941491 | val 3.381412 | tok/s 2633103.01 | norm 0.4277 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8263 | train 3.040643 | val 3.381412 | tok/s 2632034.07 | norm 0.4119 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8264 | train 2.909825 | val 3.381412 | tok/s 2624792.20 | norm 0.3772 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8265 | train 3.068038 | val 3.381412 | tok/s 2629689.31 | norm 0.3534 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8266 | train 3.070531 | val 3.381412 | tok/s 2626935.34 | norm 0.4020 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8267 | train 3.054491 | val 3.381412 | tok/s 2622026.55 | norm 0.4022 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8268 | train 3.018278 | val 3.381412 | tok/s 2629444.71 | norm 0.4407 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8269 | train 2.978947 | val 3.381412 | tok/s 2622532.25 | norm 0.4261 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8270 | train 2.965294 | val 3.381412 | tok/s 2625560.62 | norm 0.4086 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8271 | train 2.904848 | val 3.381412 | tok/s 2625811.96 | norm 0.4116 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8272 | train 3.056818 | val 3.381412 | tok/s 2529578.14 | norm 0.4340 | dt 207.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8273 | train 3.082950 | val 3.381412 | tok/s 2626863.32 | norm 0.4356 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8274 | train 2.975199 | val 3.381412 | tok/s 2630471.42 | norm 0.4161 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8275 | train 2.966912 | val 3.381412 | tok/s 2628735.69 | norm 0.4101 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8276 | train 2.885687 | val 3.381412 | tok/s 2629213.34 | norm 0.4203 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8277 | train 3.197314 | val 3.381412 | tok/s 2629693.08 | norm 0.4402 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8278 | train 3.068285 | val 3.381412 | tok/s 2595769.99 | norm 0.4056 | dt 201.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8279 | train 3.054107 | val 3.381412 | tok/s 2627293.64 | norm 0.3759 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8280 | train 3.056882 | val 3.381412 | tok/s 2630254.42 | norm 0.4391 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8281 | train 3.020899 | val 3.381412 | tok/s 2633251.83 | norm 0.4393 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8282 | train 3.024989 | val 3.381412 | tok/s 2630128.38 | norm 0.3936 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8283 | train 3.041924 | val 3.381412 | tok/s 2631567.57 | norm 0.4158 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8284 | train 2.890435 | val 3.381412 | tok/s 2631531.20 | norm 0.3841 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8285 | train 2.895725 | val 3.381412 | tok/s 2630754.53 | norm 0.3851 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8286 | train 2.990314 | val 3.381412 | tok/s 2629233.34 | norm 0.4015 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8287 | train 3.118782 | val 3.381412 | tok/s 2628297.20 | norm 0.4614 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8288 | train 3.068346 | val 3.381412 | tok/s 2618406.82 | norm 0.4182 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8289 | train 3.053081 | val 3.381412 | tok/s 2624738.89 | norm 0.3816 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8290 | train 2.953858 | val 3.381412 | tok/s 2627928.82 | norm 0.3978 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8291 | train 3.020064 | val 3.381412 | tok/s 2622357.71 | norm 0.3851 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8292 | train 2.920579 | val 3.381412 | tok/s 2626790.11 | norm 0.3960 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8293 | train 3.054269 | val 3.381412 | tok/s 2626231.14 | norm 0.4169 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8294 | train 3.143152 | val 3.381412 | tok/s 2628745.22 | norm 0.4463 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8295 | train 3.066331 | val 3.381412 | tok/s 2630450.38 | norm 0.4130 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8296 | train 2.990764 | val 3.381412 | tok/s 2619309.51 | norm 0.4153 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8297 | train 2.962885 | val 3.381412 | tok/s 2627615.61 | norm 0.4470 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8298 | train 2.977978 | val 3.381412 | tok/s 2626350.63 | norm 0.4414 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8299 | train 3.061784 | val 3.381412 | tok/s 2629411.28 | norm 0.4577 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8300 | train 3.026781 | val 3.381412 | tok/s 2634772.41 | norm 0.4050 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8301 | train 2.957709 | val 3.381412 | tok/s 2629149.15 | norm 0.4072 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8302 | train 3.200379 | val 3.381412 | tok/s 2632463.49 | norm 0.4443 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8303 | train 2.988499 | val 3.381412 | tok/s 2630627.18 | norm 0.4373 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8304 | train 3.060397 | val 3.381412 | tok/s 2623570.67 | norm 0.4166 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8305 | train 2.987552 | val 3.381412 | tok/s 2625872.30 | norm 0.4252 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8306 | train 3.125245 | val 3.381412 | tok/s 2627214.50 | norm 0.4219 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8307 | train 2.972929 | val 3.381412 | tok/s 2629825.51 | norm 0.4180 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8308 | train 2.977908 | val 3.381412 | tok/s 2631053.14 | norm 0.4196 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8309 | train 3.021289 | val 3.381412 | tok/s 2629762.20 | norm 0.4285 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8310 | train 3.039091 | val 3.381412 | tok/s 2628903.01 | norm 0.4332 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8311 | train 3.168538 | val 3.381412 | tok/s 2625242.82 | norm 0.4486 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8312 | train 3.005794 | val 3.381412 | tok/s 2627487.44 | norm 0.3952 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8313 | train 3.021370 | val 3.381412 | tok/s 2615883.72 | norm 0.3656 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8314 | train 3.014995 | val 3.381412 | tok/s 2628244.24 | norm 0.4009 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8315 | train 3.025163 | val 3.381412 | tok/s 2629134.37 | norm 0.3894 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8316 | train 3.034283 | val 3.381412 | tok/s 2627293.12 | norm 0.4058 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8317 | train 3.089241 | val 3.381412 | tok/s 2628150.52 | norm 0.3797 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8318 | train 3.058552 | val 3.381412 | tok/s 2632143.74 | norm 0.3860 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8319 | train 3.057726 | val 3.381412 | tok/s 2631003.20 | norm 0.4046 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8320 | train 3.093406 | val 3.381412 | tok/s 2631632.27 | norm 0.3984 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8321 | train 2.876367 | val 3.381412 | tok/s 2624584.97 | norm 0.3703 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8322 | train 3.043814 | val 3.381412 | tok/s 2625399.44 | norm 0.3842 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8323 | train 2.944231 | val 3.381412 | tok/s 2628817.18 | norm 0.3648 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8324 | train 3.029721 | val 3.381412 | tok/s 2627905.70 | norm 0.3954 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8325 | train 2.940800 | val 3.381412 | tok/s 2626970.57 | norm 0.4215 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8326 | train 3.005182 | val 3.381412 | tok/s 2628988.36 | norm 0.3952 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8327 | train 2.882973 | val 3.381412 | tok/s 2630793.84 | norm 0.3994 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8328 | train 3.003403 | val 3.381412 | tok/s 2628706.73 | norm 0.4239 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8329 | train 3.024521 | val 3.381412 | tok/s 2625803.78 | norm 0.4179 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8330 | train 3.037055 | val 3.381412 | tok/s 2628988.93 | norm 0.4210 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8331 | train 3.049516 | val 3.381412 | tok/s 2624228.69 | norm 0.4310 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8332 | train 3.036856 | val 3.381412 | tok/s 2621975.38 | norm 0.4567 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8333 | train 3.083703 | val 3.381412 | tok/s 2627448.76 | norm 0.3933 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8334 | train 3.136322 | val 3.381412 | tok/s 2624687.72 | norm 0.3710 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8335 | train 2.994015 | val 3.381412 | tok/s 2628609.91 | norm 0.4225 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8336 | train 2.974607 | val 3.381412 | tok/s 2621347.07 | norm 0.4273 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8337 | train 2.986644 | val 3.381412 | tok/s 2627082.53 | norm 0.4249 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8338 | train 3.006135 | val 3.381412 | tok/s 2628695.87 | norm 0.4375 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8339 | train 2.987893 | val 3.381412 | tok/s 2625882.84 | norm 0.4112 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8340 | train 2.910848 | val 3.381412 | tok/s 2628204.21 | norm 0.4061 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8341 | train 2.943877 | val 3.381412 | tok/s 2628336.60 | norm 0.4025 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8342 | train 3.039237 | val 3.381412 | tok/s 2624950.92 | norm 0.4248 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8343 | train 2.966389 | val 3.381412 | tok/s 2625052.61 | norm 0.4291 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8344 | train 2.994647 | val 3.381412 | tok/s 2622942.05 | norm 0.4182 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8345 | train 3.077380 | val 3.381412 | tok/s 2628840.98 | norm 0.4295 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8346 | train 3.140783 | val 3.381412 | tok/s 2628564.58 | norm 0.4354 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8347 | train 2.913910 | val 3.381412 | tok/s 2630389.49 | norm 0.4394 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8348 | train 2.983419 | val 3.381412 | tok/s 2623843.99 | norm 0.3896 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8349 | train 3.017587 | val 3.381412 | tok/s 2626009.31 | norm 0.3931 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8350 | train 3.127494 | val 3.381412 | tok/s 2628295.50 | norm 0.3979 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8351 | train 2.976293 | val 3.381412 | tok/s 2631874.47 | norm 0.4185 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8352 | train 3.003878 | val 3.381412 | tok/s 2630249.39 | norm 0.4201 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8353 | train 3.049299 | val 3.381412 | tok/s 2626010.16 | norm 0.4381 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8354 | train 3.096602 | val 3.381412 | tok/s 2627499.16 | norm 0.4371 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8355 | train 3.077692 | val 3.381412 | tok/s 2628819.20 | norm 0.4132 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8356 | train 3.082754 | val 3.381412 | tok/s 2628182.60 | norm 0.4174 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8357 | train 2.956553 | val 3.381412 | tok/s 2628871.65 | norm 0.4244 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8358 | train 3.006683 | val 3.381412 | tok/s 2624804.66 | norm 0.3755 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8359 | train 2.868835 | val 3.381412 | tok/s 2626053.73 | norm 0.3991 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8360 | train 3.065500 | val 3.381412 | tok/s 2631469.11 | norm 0.3790 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8361 | train 2.955704 | val 3.381412 | tok/s 2633945.61 | norm 0.3517 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8362 | train 3.091768 | val 3.381412 | tok/s 2634004.70 | norm 0.3858 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8363 | train 3.039697 | val 3.381412 | tok/s 2620170.23 | norm 0.4107 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8364 | train 2.983029 | val 3.381412 | tok/s 2627268.74 | norm 0.4204 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8365 | train 2.901010 | val 3.381412 | tok/s 2627746.68 | norm 0.3994 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8366 | train 2.978632 | val 3.381412 | tok/s 2625973.45 | norm 0.3970 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8367 | train 2.978757 | val 3.381412 | tok/s 2626981.14 | norm 0.3773 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8368 | train 3.136909 | val 3.381412 | tok/s 2628585.53 | norm 0.4340 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8369 | train 2.972072 | val 3.381412 | tok/s 2597373.13 | norm 0.5303 | dt 201.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8370 | train 3.203135 | val 3.381412 | tok/s 2631396.12 | norm 0.3708 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8371 | train 2.913996 | val 3.381412 | tok/s 2627256.65 | norm 0.3670 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8372 | train 2.933800 | val 3.381412 | tok/s 2627111.96 | norm 0.3963 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8373 | train 3.117730 | val 3.381412 | tok/s 2630560.31 | norm 0.4050 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8374 | train 3.234063 | val 3.381412 | tok/s 2629736.87 | norm 0.5154 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8375 | train 3.139105 | val 3.381412 | tok/s 2629008.65 | norm 0.3906 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8376 | train 3.051990 | val 3.381412 | tok/s 2628613.04 | norm 0.4083 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8377 | train 3.073581 | val 3.381412 | tok/s 2632254.83 | norm 0.4025 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8378 | train 3.114554 | val 3.381412 | tok/s 2618088.92 | norm 0.4134 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8379 | train 2.950054 | val 3.381412 | tok/s 2621180.01 | norm 0.4296 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8380 | train 2.899902 | val 3.381412 | tok/s 2628617.72 | norm 0.4181 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8381 | train 2.900474 | val 3.381412 | tok/s 2629260.31 | norm 0.4282 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8382 | train 3.304242 | val 3.381412 | tok/s 2630090.55 | norm 0.6164 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8383 | train 3.095213 | val 3.381412 | tok/s 2630639.28 | norm 0.4755 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8384 | train 3.046247 | val 3.381412 | tok/s 2627530.89 | norm 0.3624 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8385 | train 3.091840 | val 3.381412 | tok/s 2632106.68 | norm 0.3643 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8386 | train 2.933257 | val 3.381412 | tok/s 2626142.73 | norm 0.3851 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8387 | train 2.894005 | val 3.381412 | tok/s 2628547.22 | norm 0.3864 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8388 | train 3.108089 | val 3.381412 | tok/s 2630455.15 | norm 0.4356 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8389 | train 2.955907 | val 3.381412 | tok/s 2630490.49 | norm 0.4026 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8390 | train 2.978728 | val 3.381412 | tok/s 2628833.91 | norm 0.4383 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8391 | train 2.948032 | val 3.381412 | tok/s 2631695.94 | norm 0.4313 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8392 | train 3.012238 | val 3.381412 | tok/s 2629933.70 | norm 0.4328 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8393 | train 2.951562 | val 3.381412 | tok/s 2626851.34 | norm 0.4120 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8394 | train 3.068684 | val 3.381412 | tok/s 2628313.68 | norm 0.3982 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8395 | train 3.001206 | val 3.381412 | tok/s 2625798.50 | norm 0.4101 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8396 | train 2.925813 | val 3.381412 | tok/s 2626167.03 | norm 0.3874 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8397 | train 3.089420 | val 3.381412 | tok/s 2632001.83 | norm 0.3844 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8398 | train 3.003413 | val 3.381412 | tok/s 2630676.63 | norm 0.4015 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8399 | train 2.998214 | val 3.381412 | tok/s 2628376.76 | norm 0.3840 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8400 | train 2.967938 | val 3.381412 | tok/s 2629153.30 | norm 0.3971 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8401 | train 3.031291 | val 3.381412 | tok/s 2622758.29 | norm 0.4096 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8402 | train 3.043034 | val 3.381412 | tok/s 2627343.05 | norm 0.4458 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8403 | train 3.042727 | val 3.381412 | tok/s 2626326.48 | norm 0.4084 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8404 | train 2.895803 | val 3.381412 | tok/s 2628534.62 | norm 0.4071 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8405 | train 2.973356 | val 3.381412 | tok/s 2626203.13 | norm 0.4441 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8406 | train 3.076578 | val 3.381412 | tok/s 2572976.94 | norm 0.4198 | dt 203.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8407 | train 3.061681 | val 3.381412 | tok/s 2629242.86 | norm 0.4010 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8408 | train 3.136333 | val 3.381412 | tok/s 2628261.50 | norm 0.4081 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8409 | train 2.933644 | val 3.381412 | tok/s 2630259.14 | norm 0.4398 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8410 | train 3.117882 | val 3.381412 | tok/s 2627540.86 | norm 0.4410 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8411 | train 2.950110 | val 3.381412 | tok/s 2630857.88 | norm 0.4471 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8412 | train 2.990858 | val 3.381412 | tok/s 2629733.51 | norm 0.4355 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8413 | train 3.189127 | val 3.381412 | tok/s 2628099.56 | norm 0.4515 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8414 | train 3.020645 | val 3.381412 | tok/s 2628652.13 | norm 0.4326 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8415 | train 2.924283 | val 3.381412 | tok/s 2626194.55 | norm 0.3675 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8416 | train 3.093994 | val 3.381412 | tok/s 2627011.85 | norm 0.4001 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8417 | train 3.047211 | val 3.381412 | tok/s 2629848.80 | norm 0.4040 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8418 | train 3.269424 | val 3.381412 | tok/s 2630717.75 | norm 0.4100 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8419 | train 3.114572 | val 3.381412 | tok/s 2629629.35 | norm 0.4379 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8420 | train 2.981857 | val 3.381412 | tok/s 2623106.26 | norm 0.4156 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8421 | train 2.871887 | val 3.381412 | tok/s 2622565.50 | norm 0.4270 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8422 | train 2.991957 | val 3.381412 | tok/s 2629112.06 | norm 0.4195 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8423 | train 3.001268 | val 3.381412 | tok/s 2628102.38 | norm 0.4282 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8424 | train 3.157577 | val 3.381412 | tok/s 2630337.00 | norm 0.4740 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8425 | train 3.106909 | val 3.381412 | tok/s 2633495.55 | norm 0.5226 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8426 | train 3.122099 | val 3.381412 | tok/s 2629894.08 | norm 0.4068 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8427 | train 3.014314 | val 3.381412 | tok/s 2629540.87 | norm 0.4312 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8428 | train 2.878686 | val 3.381412 | tok/s 2623973.78 | norm 0.4194 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8429 | train 2.981655 | val 3.381412 | tok/s 2629446.31 | norm 0.4356 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8430 | train 3.000436 | val 3.381412 | tok/s 2629280.16 | norm 0.4500 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8431 | train 3.050357 | val 3.381412 | tok/s 2624644.95 | norm 0.4437 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8432 | train 3.078606 | val 3.381412 | tok/s 2626627.12 | norm 0.3951 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8433 | train 2.933714 | val 3.381412 | tok/s 2625971.21 | norm 0.3743 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8434 | train 3.052802 | val 3.381412 | tok/s 2628421.27 | norm 0.3799 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8435 | train 2.999558 | val 3.381412 | tok/s 2625647.67 | norm 0.3676 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8436 | train 3.002527 | val 3.381412 | tok/s 2630448.68 | norm 0.3590 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8437 | train 3.024082 | val 3.381412 | tok/s 2630666.00 | norm 0.4239 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8438 | train 3.135949 | val 3.381412 | tok/s 2630567.29 | norm 0.4430 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8439 | train 2.915890 | val 3.381412 | tok/s 2629960.89 | norm 0.4486 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8440 | train 2.868477 | val 3.381412 | tok/s 2630911.61 | norm 0.3812 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8441 | train 2.996271 | val 3.381412 | tok/s 2628214.12 | norm 0.4123 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8442 | train 2.996953 | val 3.381412 | tok/s 2633316.08 | norm 0.4382 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8443 | train 3.119885 | val 3.381412 | tok/s 2634506.59 | norm 0.4228 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8444 | train 3.023405 | val 3.381412 | tok/s 2634924.50 | norm 0.4156 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8445 | train 2.921749 | val 3.381412 | tok/s 2629748.76 | norm 0.4129 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8446 | train 2.946700 | val 3.381412 | tok/s 2629939.29 | norm 0.4521 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8447 | train 3.029252 | val 3.381412 | tok/s 2630818.18 | norm 0.4581 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8448 | train 3.087034 | val 3.381412 | tok/s 2628444.03 | norm 0.4537 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8449 | train 3.047798 | val 3.381412 | tok/s 2632964.71 | norm 0.4269 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8450 | train 3.124211 | val 3.381412 | tok/s 2629486.90 | norm 0.4373 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8451 | train 3.001438 | val 3.381412 | tok/s 2628388.55 | norm 0.4466 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8452 | train 3.033650 | val 3.381412 | tok/s 2627624.38 | norm 0.4560 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8453 | train 3.076771 | val 3.381412 | tok/s 2627833.28 | norm 0.4397 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8454 | train 2.911501 | val 3.381412 | tok/s 2629585.50 | norm 0.4273 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8455 | train 3.232134 | val 3.381412 | tok/s 2626946.88 | norm 0.4052 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8456 | train 2.961438 | val 3.381412 | tok/s 2628107.42 | norm 0.3955 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8457 | train 2.938971 | val 3.381412 | tok/s 2629580.35 | norm 0.3993 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8458 | train 3.236074 | val 3.381412 | tok/s 2631049.09 | norm 0.4538 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8459 | train 3.066144 | val 3.381412 | tok/s 2628668.81 | norm 0.4598 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8460 | train 3.132045 | val 3.381412 | tok/s 2625851.04 | norm 0.4262 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8461 | train 3.068058 | val 3.381412 | tok/s 2628900.53 | norm 0.3946 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8462 | train 3.039513 | val 3.381412 | tok/s 2624246.68 | norm 0.3934 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8463 | train 3.084772 | val 3.381412 | tok/s 2627382.23 | norm 0.4325 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8464 | train 3.064550 | val 3.381412 | tok/s 2626059.59 | norm 0.4206 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8465 | train 3.173438 | val 3.381412 | tok/s 2625492.36 | norm 0.3949 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8466 | train 2.942854 | val 3.381412 | tok/s 2630440.09 | norm 0.3962 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8467 | train 3.120558 | val 3.381412 | tok/s 2627614.20 | norm 0.4155 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8468 | train 3.013273 | val 3.381412 | tok/s 2629028.73 | norm 0.3967 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8469 | train 3.007281 | val 3.381412 | tok/s 2627872.52 | norm 0.3743 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8470 | train 2.954132 | val 3.381412 | tok/s 2622089.66 | norm 0.3887 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8471 | train 2.935390 | val 3.381412 | tok/s 2626930.38 | norm 0.3965 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8472 | train 2.999388 | val 3.381412 | tok/s 2620996.96 | norm 0.4074 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8473 | train 2.905083 | val 3.381412 | tok/s 2470089.63 | norm 0.3972 | dt 212.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8474 | train 3.057551 | val 3.381412 | tok/s 2599962.94 | norm 0.4000 | dt 201.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8475 | train 2.998335 | val 3.381412 | tok/s 2625836.78 | norm 0.3886 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8476 | train 3.016969 | val 3.381412 | tok/s 2628759.91 | norm 0.3879 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8477 | train 2.984938 | val 3.381412 | tok/s 2631102.72 | norm 0.4065 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8478 | train 3.073101 | val 3.381412 | tok/s 2629841.58 | norm 0.4535 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8479 | train 3.106743 | val 3.381412 | tok/s 2629926.34 | norm 0.3658 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8480 | train 2.994725 | val 3.381412 | tok/s 2629623.26 | norm 0.3681 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8481 | train 3.098424 | val 3.381412 | tok/s 2630749.83 | norm 0.4055 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8482 | train 3.013968 | val 3.381412 | tok/s 2517066.26 | norm 0.4221 | dt 208.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8483 | train 3.094665 | val 3.381412 | tok/s 2633385.88 | norm 0.4012 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8484 | train 3.053585 | val 3.381412 | tok/s 2631150.08 | norm 0.3708 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8485 | train 2.904167 | val 3.381412 | tok/s 2628054.91 | norm 0.3567 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8486 | train 2.997619 | val 3.381412 | tok/s 2635068.50 | norm 0.3929 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8487 | train 2.825873 | val 3.381412 | tok/s 2632595.65 | norm 0.3939 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8488 | train 2.983653 | val 3.381412 | tok/s 2627460.31 | norm 0.4149 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8489 | train 2.983308 | val 3.381412 | tok/s 2634343.53 | norm 0.3835 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8490 | train 3.008744 | val 3.381412 | tok/s 2628773.13 | norm 0.3953 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8491 | train 2.998793 | val 3.381412 | tok/s 2633244.06 | norm 0.4143 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8492 | train 3.117163 | val 3.381412 | tok/s 2626006.06 | norm 0.4421 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8493 | train 3.015115 | val 3.381412 | tok/s 2633488.37 | norm 0.4719 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8494 | train 3.042993 | val 3.381412 | tok/s 2630737.62 | norm 0.4275 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8495 | train 2.988245 | val 3.381412 | tok/s 2630053.39 | norm 0.4191 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8496 | train 3.008133 | val 3.381412 | tok/s 2625957.05 | norm 0.4018 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8497 | train 3.094727 | val 3.381412 | tok/s 2627550.36 | norm 0.3731 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8498 | train 3.073896 | val 3.381412 | tok/s 2628187.73 | norm 0.4044 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8499 | train 3.005308 | val 3.381412 | tok/s 2629331.48 | norm 0.4504 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 8500 | val_loss 3.382579 step 8500 | train 2.971710 | val 3.382579 | tok/s 2613969.61 | norm 0.4261 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8501 | train 2.951336 | val 3.382579 | tok/s 2631530.38 | norm 0.4061 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8502 | train 2.919663 | val 3.382579 | tok/s 2625919.94 | norm 0.4095 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8503 | train 3.009782 | val 3.382579 | tok/s 2632647.79 | norm 0.3981 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8504 | train 2.944617 | val 3.382579 | tok/s 2632545.74 | norm 0.3871 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8505 | train 3.028214 | val 3.382579 | tok/s 2632602.95 | norm 0.4123 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8506 | train 3.008139 | val 3.382579 | tok/s 2627885.92 | norm 0.4212 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8507 | train 3.090704 | val 3.382579 | tok/s 2631409.85 | norm 0.4691 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8508 | train 3.073552 | val 3.382579 | tok/s 2632832.16 | norm 0.4021 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8509 | train 3.102876 | val 3.382579 | tok/s 2627086.07 | norm 0.4062 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8510 | train 2.986145 | val 3.382579 | tok/s 2629142.84 | norm 0.3843 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8511 | train 3.075789 | val 3.382579 | tok/s 2631132.83 | norm 0.3925 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8512 | train 3.152452 | val 3.382579 | tok/s 2632287.19 | norm 0.4081 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8513 | train 3.058855 | val 3.382579 | tok/s 2631250.00 | norm 0.4008 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8514 | train 3.036967 | val 3.382579 | tok/s 2629944.12 | norm 0.3928 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8515 | train 3.022059 | val 3.382579 | tok/s 2627627.72 | norm 0.3989 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8516 | train 3.037502 | val 3.382579 | tok/s 2622839.91 | norm 0.4319 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8517 | train 2.988279 | val 3.382579 | tok/s 2632379.19 | norm 0.4599 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8518 | train 2.948199 | val 3.382579 | tok/s 2629063.85 | norm 0.3982 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8519 | train 3.134806 | val 3.382579 | tok/s 2628503.25 | norm 0.4260 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8520 | train 3.161656 | val 3.382579 | tok/s 2628380.49 | norm 0.4645 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8521 | train 2.989124 | val 3.382579 | tok/s 2565903.11 | norm 0.3815 | dt 204.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8522 | train 3.069665 | val 3.382579 | tok/s 2626734.99 | norm 0.3523 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8523 | train 3.047437 | val 3.382579 | tok/s 2627911.26 | norm 0.4029 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8524 | train 3.007792 | val 3.382579 | tok/s 2625511.62 | norm 0.4035 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8525 | train 2.976884 | val 3.382579 | tok/s 2621276.56 | norm 0.3953 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8526 | train 3.044515 | val 3.382579 | tok/s 2626729.77 | norm 0.3898 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8527 | train 2.961041 | val 3.382579 | tok/s 2633526.65 | norm 0.4152 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8528 | train 2.909148 | val 3.382579 | tok/s 2631555.71 | norm 0.3961 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8529 | train 2.976951 | val 3.382579 | tok/s 2629060.01 | norm 0.4016 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8530 | train 3.059515 | val 3.382579 | tok/s 2627384.43 | norm 0.3973 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8531 | train 3.039036 | val 3.382579 | tok/s 2626648.95 | norm 0.3983 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8532 | train 2.904603 | val 3.382579 | tok/s 2628722.22 | norm 0.4215 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8533 | train 2.996101 | val 3.382579 | tok/s 2629270.43 | norm 0.4498 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8534 | train 2.958927 | val 3.382579 | tok/s 2631117.94 | norm 0.3813 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8535 | train 2.955186 | val 3.382579 | tok/s 2481465.07 | norm 0.3562 | dt 211.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8536 | train 3.079699 | val 3.382579 | tok/s 2626343.76 | norm 0.4566 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8537 | train 2.935724 | val 3.382579 | tok/s 2628537.52 | norm 0.4080 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8538 | train 3.061295 | val 3.382579 | tok/s 2629071.60 | norm 0.4050 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8539 | train 3.073748 | val 3.382579 | tok/s 2632036.06 | norm 0.4131 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8540 | train 3.112855 | val 3.382579 | tok/s 2629178.51 | norm 0.3622 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8541 | train 2.953897 | val 3.382579 | tok/s 2627483.20 | norm 0.3478 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8542 | train 3.004571 | val 3.382579 | tok/s 2634271.76 | norm 0.3641 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8543 | train 3.027922 | val 3.382579 | tok/s 2633247.38 | norm 0.3940 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8544 | train 2.979682 | val 3.382579 | tok/s 2628698.24 | norm 0.4223 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8545 | train 2.928843 | val 3.382579 | tok/s 2628493.56 | norm 0.4505 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8546 | train 2.994885 | val 3.382579 | tok/s 2628207.46 | norm 0.4364 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8547 | train 3.000555 | val 3.382579 | tok/s 2628471.86 | norm 0.4328 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8548 | train 2.973656 | val 3.382579 | tok/s 2625434.32 | norm 0.4686 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8549 | train 3.000153 | val 3.382579 | tok/s 2619128.97 | norm 0.4430 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8550 | train 3.028897 | val 3.382579 | tok/s 2628408.88 | norm 0.4289 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8551 | train 3.059398 | val 3.382579 | tok/s 2627707.35 | norm 0.4198 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8552 | train 3.027308 | val 3.382579 | tok/s 2624389.39 | norm 0.4184 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8553 | train 3.142537 | val 3.382579 | tok/s 2623766.81 | norm 0.3730 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8554 | train 3.111736 | val 3.382579 | tok/s 2631398.30 | norm 0.3540 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8555 | train 3.075343 | val 3.382579 | tok/s 2625819.78 | norm 0.4113 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8556 | train 2.892382 | val 3.382579 | tok/s 2626569.57 | norm 0.3987 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8557 | train 3.030416 | val 3.382579 | tok/s 2625979.95 | norm 0.4031 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8558 | train 2.986424 | val 3.382579 | tok/s 2628094.16 | norm 0.4132 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8559 | train 2.951336 | val 3.382579 | tok/s 2627482.33 | norm 0.3956 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8560 | train 3.014473 | val 3.382579 | tok/s 2627114.01 | norm 0.4088 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8561 | train 3.010826 | val 3.382579 | tok/s 2626222.88 | norm 0.4172 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8562 | train 3.005590 | val 3.382579 | tok/s 2625080.75 | norm 0.4316 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8563 | train 2.985035 | val 3.382579 | tok/s 2628052.59 | norm 0.4163 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8564 | train 2.985652 | val 3.382579 | tok/s 2627283.10 | norm 0.3782 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8565 | train 2.970258 | val 3.382579 | tok/s 2627233.50 | norm 0.3882 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8566 | train 3.067225 | val 3.382579 | tok/s 2629226.57 | norm 0.4488 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8567 | train 2.951961 | val 3.382579 | tok/s 2626797.01 | norm 0.4139 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8568 | train 3.042291 | val 3.382579 | tok/s 2622110.05 | norm 0.4152 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8569 | train 2.948064 | val 3.382579 | tok/s 2631375.27 | norm 0.4485 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8570 | train 2.931608 | val 3.382579 | tok/s 2624438.59 | norm 0.4416 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8571 | train 2.964891 | val 3.382579 | tok/s 2624248.09 | norm 0.4202 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8572 | train 3.061907 | val 3.382579 | tok/s 2626335.49 | norm 0.4368 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8573 | train 3.045102 | val 3.382579 | tok/s 2619608.23 | norm 0.4364 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8574 | train 3.144268 | val 3.382579 | tok/s 2628147.56 | norm 0.5124 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8575 | train 3.215487 | val 3.382579 | tok/s 2626820.03 | norm 0.4327 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8576 | train 3.033376 | val 3.382579 | tok/s 2626178.97 | norm 0.3678 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8577 | train 3.032802 | val 3.382579 | tok/s 2627941.70 | norm 0.4091 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8578 | train 2.963924 | val 3.382579 | tok/s 2629919.73 | norm 0.4144 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8579 | train 3.048045 | val 3.382579 | tok/s 2628255.30 | norm 0.3847 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8580 | train 2.958988 | val 3.382579 | tok/s 2628455.77 | norm 0.3850 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8581 | train 3.060609 | val 3.382579 | tok/s 2626345.42 | norm 0.4067 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8582 | train 3.050876 | val 3.382579 | tok/s 2625614.11 | norm 0.3951 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8583 | train 3.056310 | val 3.382579 | tok/s 2629012.29 | norm 0.3915 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8584 | train 3.092531 | val 3.382579 | tok/s 2559044.43 | norm 0.3908 | dt 204.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8585 | train 3.027398 | val 3.382579 | tok/s 2595996.29 | norm 0.4024 | dt 201.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8586 | train 2.888502 | val 3.382579 | tok/s 2579719.31 | norm 0.3506 | dt 203.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8587 | train 3.015901 | val 3.382579 | tok/s 2622965.10 | norm 0.3646 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8588 | train 3.039082 | val 3.382579 | tok/s 2624812.58 | norm 0.4034 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8589 | train 3.083620 | val 3.382579 | tok/s 2629479.90 | norm 0.3887 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8590 | train 2.953606 | val 3.382579 | tok/s 2624712.07 | norm 0.3691 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8591 | train 3.058184 | val 3.382579 | tok/s 2628730.92 | norm 0.4171 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8592 | train 3.020858 | val 3.382579 | tok/s 2623360.08 | norm 0.4009 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8593 | train 3.102964 | val 3.382579 | tok/s 2628054.89 | norm 0.4006 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8594 | train 3.128237 | val 3.382579 | tok/s 2622598.39 | norm 0.4237 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8595 | train 3.107861 | val 3.382579 | tok/s 2628630.68 | norm 0.4284 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8596 | train 3.128499 | val 3.382579 | tok/s 2629665.92 | norm 0.3825 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8597 | train 2.945106 | val 3.382579 | tok/s 2633391.89 | norm 0.3860 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8598 | train 3.121815 | val 3.382579 | tok/s 2632020.99 | norm 0.3825 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8599 | train 2.949337 | val 3.382579 | tok/s 2564701.14 | norm 0.3736 | dt 204.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8600 | train 2.946834 | val 3.382579 | tok/s 2624387.29 | norm 0.3828 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8601 | train 2.930202 | val 3.382579 | tok/s 2628233.02 | norm 0.4130 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8602 | train 3.111483 | val 3.382579 | tok/s 2629108.16 | norm 0.4666 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8603 | train 3.021099 | val 3.382579 | tok/s 2621564.45 | norm 0.3909 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8604 | train 3.035053 | val 3.382579 | tok/s 2630439.38 | norm 0.3540 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8605 | train 3.152328 | val 3.382579 | tok/s 2627821.69 | norm 0.4912 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8606 | train 3.184325 | val 3.382579 | tok/s 2629234.46 | norm 0.4363 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8607 | train 3.020655 | val 3.382579 | tok/s 2631006.87 | norm 0.3830 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8608 | train 3.120290 | val 3.382579 | tok/s 2629385.79 | norm 0.3904 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8609 | train 3.085282 | val 3.382579 | tok/s 2630305.79 | norm 0.4106 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8610 | train 3.000099 | val 3.382579 | tok/s 2629266.14 | norm 0.4138 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8611 | train 3.002586 | val 3.382579 | tok/s 2633373.16 | norm 0.3834 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8612 | train 3.073296 | val 3.382579 | tok/s 2631600.87 | norm 0.3877 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8613 | train 3.075992 | val 3.382579 | tok/s 2625076.24 | norm 0.4222 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8614 | train 3.003402 | val 3.382579 | tok/s 2623598.21 | norm 0.4218 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8615 | train 3.107165 | val 3.382579 | tok/s 2622341.43 | norm 0.4098 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8616 | train 3.014565 | val 3.382579 | tok/s 2628728.11 | norm 0.4028 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8617 | train 3.022382 | val 3.382579 | tok/s 2621970.07 | norm 0.4192 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8618 | train 3.002748 | val 3.382579 | tok/s 2624262.13 | norm 0.4470 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8619 | train 2.987694 | val 3.382579 | tok/s 2628037.44 | norm 0.4589 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8620 | train 3.099818 | val 3.382579 | tok/s 2629459.98 | norm 0.4388 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8621 | train 3.029819 | val 3.382579 | tok/s 2628258.35 | norm 0.4186 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8622 | train 2.958138 | val 3.382579 | tok/s 2626474.04 | norm 0.4224 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8623 | train 3.141024 | val 3.382579 | tok/s 2629572.56 | norm 0.4558 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8624 | train 2.905302 | val 3.382579 | tok/s 2625920.29 | norm 0.4410 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8625 | train 2.958695 | val 3.382579 | tok/s 2623531.52 | norm 0.4508 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8626 | train 2.912870 | val 3.382579 | tok/s 2621314.10 | norm 0.4368 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8627 | train 2.988872 | val 3.382579 | tok/s 2622779.90 | norm 0.4290 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8628 | train 2.856627 | val 3.382579 | tok/s 2622448.00 | norm 0.4250 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8629 | train 2.932786 | val 3.382579 | tok/s 2617355.56 | norm 0.4372 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8630 | train 2.968112 | val 3.382579 | tok/s 2622688.53 | norm 0.4883 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8631 | train 2.925649 | val 3.382579 | tok/s 2626906.88 | norm 0.4729 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8632 | train 3.199399 | val 3.382579 | tok/s 2624448.42 | norm 0.4169 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8633 | train 3.072683 | val 3.382579 | tok/s 2625557.42 | norm 0.3939 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8634 | train 3.050840 | val 3.382579 | tok/s 2623842.64 | norm 0.3944 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8635 | train 2.977182 | val 3.382579 | tok/s 2627607.22 | norm 0.4055 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8636 | train 3.025002 | val 3.382579 | tok/s 2626845.46 | norm 0.4163 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8637 | train 3.106538 | val 3.382579 | tok/s 2626383.12 | norm 0.4106 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8638 | train 3.088852 | val 3.382579 | tok/s 2627121.45 | norm 0.4117 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8639 | train 3.010027 | val 3.382579 | tok/s 2617799.96 | norm 0.4407 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8640 | train 2.954189 | val 3.382579 | tok/s 2447565.14 | norm 0.3983 | dt 214.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8641 | train 3.026878 | val 3.382579 | tok/s 2601881.91 | norm 0.3963 | dt 201.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8642 | train 3.093519 | val 3.382579 | tok/s 2628502.54 | norm 0.4320 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8643 | train 3.161551 | val 3.382579 | tok/s 2628968.21 | norm 0.4769 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8644 | train 3.013740 | val 3.382579 | tok/s 2629754.67 | norm 0.5181 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8645 | train 3.037052 | val 3.382579 | tok/s 2626685.03 | norm 0.4552 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8646 | train 3.096097 | val 3.382579 | tok/s 2629738.60 | norm 0.4504 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8647 | train 2.973825 | val 3.382579 | tok/s 2627743.25 | norm 0.4361 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8648 | train 3.018144 | val 3.382579 | tok/s 2627679.29 | norm 0.4048 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8649 | train 3.020585 | val 3.382579 | tok/s 2627734.66 | norm 0.3792 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8650 | train 3.137064 | val 3.382579 | tok/s 2623589.39 | norm 0.3970 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8651 | train 3.131243 | val 3.382579 | tok/s 2632422.45 | norm 0.5078 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8652 | train 3.036594 | val 3.382579 | tok/s 2628003.08 | norm 0.4355 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8653 | train 3.025387 | val 3.382579 | tok/s 2628077.33 | norm 0.3945 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8654 | train 2.845586 | val 3.382579 | tok/s 2622850.10 | norm 0.4054 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8655 | train 3.236754 | val 3.382579 | tok/s 2601370.89 | norm 0.4046 | dt 201.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8656 | train 3.106020 | val 3.382579 | tok/s 2629343.27 | norm 0.4555 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8657 | train 3.061633 | val 3.382579 | tok/s 2601910.91 | norm 0.3884 | dt 201.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8658 | train 3.020327 | val 3.382579 | tok/s 2629904.24 | norm 0.4032 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8659 | train 2.920280 | val 3.382579 | tok/s 2627916.91 | norm 0.4488 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8660 | train 2.993876 | val 3.382579 | tok/s 2625712.35 | norm 0.4434 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8661 | train 3.043178 | val 3.382579 | tok/s 2622626.90 | norm 0.4168 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8662 | train 3.121856 | val 3.382579 | tok/s 2628415.50 | norm 0.3743 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8663 | train 2.937298 | val 3.382579 | tok/s 2633565.96 | norm 0.3736 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8664 | train 3.108207 | val 3.382579 | tok/s 2633751.37 | norm 0.4270 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8665 | train 2.970485 | val 3.382579 | tok/s 2631992.18 | norm 0.4689 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8666 | train 2.983572 | val 3.382579 | tok/s 2629389.38 | norm 0.4066 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8667 | train 3.183362 | val 3.382579 | tok/s 2621830.43 | norm 0.4162 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8668 | train 3.023304 | val 3.382579 | tok/s 2623147.20 | norm 0.4285 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8669 | train 2.991087 | val 3.382579 | tok/s 2627600.98 | norm 0.3742 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8670 | train 2.979238 | val 3.382579 | tok/s 2628919.42 | norm 0.3401 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8671 | train 2.938910 | val 3.382579 | tok/s 2629961.97 | norm 0.3735 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8672 | train 2.925124 | val 3.382579 | tok/s 2629436.50 | norm 0.3943 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8673 | train 2.954473 | val 3.382579 | tok/s 2627621.47 | norm 0.4107 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8674 | train 3.002149 | val 3.382579 | tok/s 2630114.05 | norm 0.4258 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8675 | train 2.971177 | val 3.382579 | tok/s 2631051.80 | norm 0.4341 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8676 | train 2.979671 | val 3.382579 | tok/s 2628510.84 | norm 0.3995 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8677 | train 3.026102 | val 3.382579 | tok/s 2628463.86 | norm 0.4607 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8678 | train 2.978045 | val 3.382579 | tok/s 2627260.00 | norm 0.4571 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8679 | train 2.900358 | val 3.382579 | tok/s 2626000.77 | norm 0.4388 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8680 | train 3.325935 | val 3.382579 | tok/s 2629472.24 | norm 0.4543 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8681 | train 2.986413 | val 3.382579 | tok/s 2630742.81 | norm 0.4198 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8682 | train 3.042451 | val 3.382579 | tok/s 2633193.06 | norm 0.3917 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8683 | train 2.901061 | val 3.382579 | tok/s 2631328.32 | norm 0.3985 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8684 | train 2.959334 | val 3.382579 | tok/s 2628592.90 | norm 0.4222 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8685 | train 2.916222 | val 3.382579 | tok/s 2613974.91 | norm 0.4222 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8686 | train 3.021744 | val 3.382579 | tok/s 2625496.89 | norm 0.4022 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8687 | train 2.961315 | val 3.382579 | tok/s 2627297.03 | norm 0.3971 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8688 | train 2.912910 | val 3.382579 | tok/s 2626958.86 | norm 0.4248 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8689 | train 3.052114 | val 3.382579 | tok/s 2628206.61 | norm 0.4042 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8690 | train 3.145457 | val 3.382579 | tok/s 2627997.99 | norm 0.4384 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8691 | train 3.002939 | val 3.382579 | tok/s 2626605.72 | norm 0.4167 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8692 | train 3.054329 | val 3.382579 | tok/s 2629353.47 | norm 0.3718 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8693 | train 2.946113 | val 3.382579 | tok/s 2628862.42 | norm 0.4095 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8694 | train 3.005370 | val 3.382579 | tok/s 2627930.40 | norm 0.4384 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8695 | train 2.965792 | val 3.382579 | tok/s 2631273.00 | norm 0.4289 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8696 | train 3.042532 | val 3.382579 | tok/s 2623870.22 | norm 0.4220 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8697 | train 3.046461 | val 3.382579 | tok/s 2626758.61 | norm 0.4137 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8698 | train 3.005080 | val 3.382579 | tok/s 2624229.79 | norm 0.4322 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8699 | train 2.935609 | val 3.382579 | tok/s 2626176.68 | norm 0.4109 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8700 | train 2.855361 | val 3.382579 | tok/s 2632134.73 | norm 0.3908 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8701 | train 3.035016 | val 3.382579 | tok/s 2628593.40 | norm 0.4407 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8702 | train 2.987888 | val 3.382579 | tok/s 2628990.57 | norm 0.4499 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8703 | train 3.013111 | val 3.382579 | tok/s 2627339.32 | norm 0.3986 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8704 | train 2.972945 | val 3.382579 | tok/s 2627247.77 | norm 0.3703 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8705 | train 3.046482 | val 3.382579 | tok/s 2627257.96 | norm 0.3995 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8706 | train 2.915983 | val 3.382579 | tok/s 2622712.75 | norm 0.3884 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8707 | train 3.113016 | val 3.382579 | tok/s 2623658.36 | norm 0.3965 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8708 | train 3.020026 | val 3.382579 | tok/s 2628782.41 | norm 0.3788 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8709 | train 3.076920 | val 3.382579 | tok/s 2471397.21 | norm 0.3965 | dt 212.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8710 | train 2.966043 | val 3.382579 | tok/s 2627627.72 | norm 0.4170 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8711 | train 3.020912 | val 3.382579 | tok/s 2621218.23 | norm 0.3954 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8712 | train 2.843099 | val 3.382579 | tok/s 2626111.66 | norm 0.4089 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8713 | train 3.075589 | val 3.382579 | tok/s 2630698.57 | norm 0.3661 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8714 | train 2.925550 | val 3.382579 | tok/s 2627245.37 | norm 0.3793 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8715 | train 2.975154 | val 3.382579 | tok/s 2628768.65 | norm 0.3854 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8716 | train 2.951614 | val 3.382579 | tok/s 2629139.70 | norm 0.4334 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8717 | train 3.062678 | val 3.382579 | tok/s 2627208.64 | norm 0.4854 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8718 | train 3.008295 | val 3.382579 | tok/s 2629304.35 | norm 0.4357 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8719 | train 3.061163 | val 3.382579 | tok/s 2628159.22 | norm 0.3951 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8720 | train 2.906128 | val 3.382579 | tok/s 2624843.87 | norm 0.4034 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8721 | train 2.935688 | val 3.382579 | tok/s 2630844.45 | norm 0.4119 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8722 | train 3.008971 | val 3.382579 | tok/s 2630949.74 | norm 0.4328 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8723 | train 2.894110 | val 3.382579 | tok/s 2627539.50 | norm 0.4249 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8724 | train 2.897091 | val 3.382579 | tok/s 2624936.49 | norm 0.4227 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8725 | train 2.963155 | val 3.382579 | tok/s 2628651.06 | norm 0.4265 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8726 | train 2.970559 | val 3.382579 | tok/s 2632141.58 | norm 0.4189 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8727 | train 2.995962 | val 3.382579 | tok/s 2628577.68 | norm 0.4096 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8728 | train 3.044397 | val 3.382579 | tok/s 2629123.23 | norm 0.4337 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8729 | train 3.035105 | val 3.382579 | tok/s 2625089.28 | norm 0.3849 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8730 | train 2.949326 | val 3.382579 | tok/s 2627567.63 | norm 0.3571 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8731 | train 2.957775 | val 3.382579 | tok/s 2624351.01 | norm 0.3971 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8732 | train 3.145216 | val 3.382579 | tok/s 2624859.40 | norm 0.4125 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8733 | train 3.138714 | val 3.382579 | tok/s 2623811.35 | norm 0.4166 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8734 | train 3.049854 | val 3.382579 | tok/s 2618632.71 | norm 0.4007 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8735 | train 3.060798 | val 3.382579 | tok/s 2625489.08 | norm 0.4246 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8736 | train 2.930103 | val 3.382579 | tok/s 2622835.02 | norm 0.4178 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8737 | train 2.895584 | val 3.382579 | tok/s 2623424.88 | norm 0.4027 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8738 | train 2.975125 | val 3.382579 | tok/s 2613594.08 | norm 0.4544 | dt 200.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8739 | train 2.967101 | val 3.382579 | tok/s 2624631.21 | norm 0.3837 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8740 | train 3.032018 | val 3.382579 | tok/s 2630597.74 | norm 0.3605 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8741 | train 2.914606 | val 3.382579 | tok/s 2626419.36 | norm 0.4367 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8742 | train 3.096207 | val 3.382579 | tok/s 2627641.35 | norm 0.4646 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8743 | train 2.938039 | val 3.382579 | tok/s 2620740.39 | norm 0.4371 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8744 | train 2.919128 | val 3.382579 | tok/s 2622956.61 | norm 0.3785 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8745 | train 3.032614 | val 3.382579 | tok/s 2627601.27 | norm 0.3856 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8746 | train 3.078677 | val 3.382579 | tok/s 2628588.53 | norm 0.3394 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8747 | train 2.940518 | val 3.382579 | tok/s 2628782.29 | norm 0.3676 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8748 | train 3.126053 | val 3.382579 | tok/s 2626590.69 | norm 0.3819 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8749 | train 3.014389 | val 3.382579 | tok/s 2632191.03 | norm 0.3914 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 8750 | val_loss 3.381831 step 8750 | train 2.955061 | val 3.381831 | tok/s 2605472.34 | norm 0.3987 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8751 | train 2.887490 | val 3.381831 | tok/s 2623645.32 | norm 0.4085 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8752 | train 2.902069 | val 3.381831 | tok/s 2627592.89 | norm 0.4376 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8753 | train 3.149993 | val 3.381831 | tok/s 2623160.48 | norm 0.4177 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8754 | train 2.990215 | val 3.381831 | tok/s 2624773.07 | norm 0.3819 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8755 | train 2.996773 | val 3.381831 | tok/s 2603307.11 | norm 0.3842 | dt 201.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8756 | train 2.947701 | val 3.381831 | tok/s 2629536.08 | norm 0.4237 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8757 | train 2.975732 | val 3.381831 | tok/s 2623384.02 | norm 0.4170 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8758 | train 3.023248 | val 3.381831 | tok/s 2629151.62 | norm 0.3948 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8759 | train 3.041813 | val 3.381831 | tok/s 2630409.01 | norm 0.3780 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8760 | train 2.874272 | val 3.381831 | tok/s 2627939.55 | norm 0.4077 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8761 | train 3.016694 | val 3.381831 | tok/s 2499132.48 | norm 0.4236 | dt 209.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8762 | train 3.064031 | val 3.381831 | tok/s 2626082.02 | norm 0.3883 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8763 | train 3.115224 | val 3.381831 | tok/s 2630291.27 | norm 0.4383 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8764 | train 3.143023 | val 3.381831 | tok/s 2630275.48 | norm 0.4621 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8765 | train 2.949781 | val 3.381831 | tok/s 2628798.24 | norm 0.4302 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8766 | train 3.190114 | val 3.381831 | tok/s 2631736.50 | norm 0.4150 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8767 | train 3.184127 | val 3.381831 | tok/s 2632906.56 | norm 0.3857 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8768 | train 2.964565 | val 3.381831 | tok/s 2629192.34 | norm 0.4204 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8769 | train 3.004867 | val 3.381831 | tok/s 2595671.70 | norm 0.4186 | dt 201.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8770 | train 2.979032 | val 3.381831 | tok/s 2620527.64 | norm 0.3819 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8771 | train 2.953662 | val 3.381831 | tok/s 2625351.70 | norm 0.4298 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8772 | train 2.922797 | val 3.381831 | tok/s 2626837.76 | norm 0.3964 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8773 | train 2.991513 | val 3.381831 | tok/s 2626507.02 | norm 0.4034 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8774 | train 2.984846 | val 3.381831 | tok/s 2627796.07 | norm 0.4041 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8775 | train 3.001997 | val 3.381831 | tok/s 2567979.82 | norm 0.4113 | dt 204.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8776 | train 3.089875 | val 3.381831 | tok/s 2591256.64 | norm 0.4249 | dt 202.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8777 | train 2.999751 | val 3.381831 | tok/s 2630818.91 | norm 0.3967 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8778 | train 3.041964 | val 3.381831 | tok/s 2632008.26 | norm 0.3876 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8779 | train 3.051676 | val 3.381831 | tok/s 2632185.58 | norm 0.3954 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8780 | train 3.006010 | val 3.381831 | tok/s 2628719.36 | norm 0.3972 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8781 | train 2.964370 | val 3.381831 | tok/s 2625145.77 | norm 0.4268 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8782 | train 2.957482 | val 3.381831 | tok/s 2625154.50 | norm 0.4058 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8783 | train 2.940457 | val 3.381831 | tok/s 2629554.67 | norm 0.3879 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8784 | train 2.989439 | val 3.381831 | tok/s 2623640.87 | norm 0.4153 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8785 | train 2.941658 | val 3.381831 | tok/s 2628112.76 | norm 0.4228 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8786 | train 3.156383 | val 3.381831 | tok/s 2628314.69 | norm 0.3876 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8787 | train 2.934191 | val 3.381831 | tok/s 2628642.52 | norm 0.4143 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8788 | train 3.015671 | val 3.381831 | tok/s 2629715.40 | norm 0.4102 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8789 | train 3.122127 | val 3.381831 | tok/s 2630844.03 | norm 0.4346 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8790 | train 3.122433 | val 3.381831 | tok/s 2628470.63 | norm 0.4419 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8791 | train 3.037669 | val 3.381831 | tok/s 2628525.44 | norm 0.3751 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8792 | train 3.019467 | val 3.381831 | tok/s 2626387.88 | norm 0.3753 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8793 | train 3.090722 | val 3.381831 | tok/s 2631101.50 | norm 0.3803 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8794 | train 2.935195 | val 3.381831 | tok/s 2623686.28 | norm 0.3628 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8795 | train 2.964787 | val 3.381831 | tok/s 2628601.07 | norm 0.3817 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8796 | train 2.940845 | val 3.381831 | tok/s 2629101.83 | norm 0.4023 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8797 | train 3.062228 | val 3.381831 | tok/s 2626475.22 | norm 0.4282 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8798 | train 3.012804 | val 3.381831 | tok/s 2628625.89 | norm 0.4389 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8799 | train 3.174151 | val 3.381831 | tok/s 2626825.64 | norm 0.4269 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8800 | train 3.082687 | val 3.381831 | tok/s 2629264.53 | norm 0.3862 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8801 | train 3.036617 | val 3.381831 | tok/s 2626573.23 | norm 0.3864 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8802 | train 3.032853 | val 3.381831 | tok/s 2632147.25 | norm 0.3705 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8803 | train 3.076594 | val 3.381831 | tok/s 2627493.55 | norm 0.3779 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8804 | train 2.959981 | val 3.381831 | tok/s 2624162.78 | norm 0.4289 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8805 | train 3.068973 | val 3.381831 | tok/s 2626481.41 | norm 0.4700 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8806 | train 3.075143 | val 3.381831 | tok/s 2627894.65 | norm 0.4119 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8807 | train 2.956975 | val 3.381831 | tok/s 2628913.41 | norm 0.4161 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8808 | train 2.999228 | val 3.381831 | tok/s 2623028.95 | norm 0.4281 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8809 | train 2.963231 | val 3.381831 | tok/s 2621993.94 | norm 0.3956 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8810 | train 3.112856 | val 3.381831 | tok/s 2628746.34 | norm 0.4205 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8811 | train 3.069734 | val 3.381831 | tok/s 2627419.35 | norm 0.4227 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8812 | train 2.995481 | val 3.381831 | tok/s 2631590.21 | norm 0.4152 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8813 | train 3.048430 | val 3.381831 | tok/s 2626656.83 | norm 0.4131 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8814 | train 3.097776 | val 3.381831 | tok/s 2626322.02 | norm 0.4050 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8815 | train 2.911104 | val 3.381831 | tok/s 2622920.26 | norm 0.3813 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8816 | train 3.118954 | val 3.381831 | tok/s 2630821.40 | norm 0.4124 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8817 | train 3.058738 | val 3.381831 | tok/s 2630901.30 | norm 0.3850 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8818 | train 3.006751 | val 3.381831 | tok/s 2630400.29 | norm 0.3994 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8819 | train 3.091503 | val 3.381831 | tok/s 2630562.40 | norm 0.4048 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8820 | train 3.004290 | val 3.381831 | tok/s 2630460.24 | norm 0.3992 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8821 | train 2.947820 | val 3.381831 | tok/s 2624841.01 | norm 0.4091 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8822 | train 3.108026 | val 3.381831 | tok/s 2612699.71 | norm 0.4287 | dt 200.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8823 | train 3.003607 | val 3.381831 | tok/s 2628070.21 | norm 0.3835 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8824 | train 3.007734 | val 3.381831 | tok/s 2630000.10 | norm 0.3745 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8825 | train 3.091945 | val 3.381831 | tok/s 2630043.85 | norm 0.4108 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8826 | train 2.989021 | val 3.381831 | tok/s 2626927.60 | norm 0.4328 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8827 | train 2.931601 | val 3.381831 | tok/s 2628259.97 | norm 0.4347 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8828 | train 2.997080 | val 3.381831 | tok/s 2631144.27 | norm 0.4518 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8829 | train 3.112962 | val 3.381831 | tok/s 2633701.20 | norm 0.4429 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8830 | train 3.190052 | val 3.381831 | tok/s 2632036.63 | norm 0.4366 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8831 | train 3.035860 | val 3.381831 | tok/s 2625136.77 | norm 0.3859 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8832 | train 2.986276 | val 3.381831 | tok/s 2624791.65 | norm 0.4114 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8833 | train 2.921496 | val 3.381831 | tok/s 2624069.34 | norm 0.4375 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8834 | train 3.044257 | val 3.381831 | tok/s 2627530.79 | norm 0.3973 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8835 | train 3.089872 | val 3.381831 | tok/s 2631227.90 | norm 0.4005 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8836 | train 2.970921 | val 3.381831 | tok/s 2624294.17 | norm 0.3986 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8837 | train 3.030803 | val 3.381831 | tok/s 2623696.80 | norm 0.3921 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8838 | train 2.998390 | val 3.381831 | tok/s 2631274.98 | norm 0.4118 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8839 | train 3.102282 | val 3.381831 | tok/s 2624524.04 | norm 0.4193 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8840 | train 3.092233 | val 3.381831 | tok/s 2629065.63 | norm 0.3858 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8841 | train 2.975352 | val 3.381831 | tok/s 2633751.89 | norm 0.3887 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8842 | train 3.079164 | val 3.381831 | tok/s 2633857.88 | norm 0.3773 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8843 | train 3.067783 | val 3.381831 | tok/s 2633606.79 | norm 0.3887 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8844 | train 2.976391 | val 3.381831 | tok/s 2627473.77 | norm 0.4176 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8845 | train 2.944621 | val 3.381831 | tok/s 2633661.90 | norm 0.5284 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8846 | train 3.019972 | val 3.381831 | tok/s 2632280.64 | norm 0.4868 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8847 | train 3.071134 | val 3.381831 | tok/s 2633414.50 | norm 0.4635 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8848 | train 2.933073 | val 3.381831 | tok/s 2630016.97 | norm 0.5040 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8849 | train 2.953577 | val 3.381831 | tok/s 2632784.31 | norm 0.3828 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8850 | train 3.011122 | val 3.381831 | tok/s 2628685.41 | norm 0.4179 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8851 | train 2.912927 | val 3.381831 | tok/s 2623093.18 | norm 0.4375 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8852 | train 3.048953 | val 3.381831 | tok/s 2628052.52 | norm 0.4422 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8853 | train 3.097981 | val 3.381831 | tok/s 2633086.95 | norm 0.4540 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8854 | train 3.101620 | val 3.381831 | tok/s 2631756.63 | norm 0.4245 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8855 | train 3.068948 | val 3.381831 | tok/s 2628370.96 | norm 0.4154 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8856 | train 3.115850 | val 3.381831 | tok/s 2627907.19 | norm 0.3755 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8857 | train 3.042324 | val 3.381831 | tok/s 2627739.35 | norm 0.3703 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8858 | train 3.238952 | val 3.381831 | tok/s 2628616.65 | norm 0.4053 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8859 | train 2.870450 | val 3.381831 | tok/s 2629735.78 | norm 0.4311 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8860 | train 3.073785 | val 3.381831 | tok/s 2634684.35 | norm 0.4469 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8861 | train 3.011513 | val 3.381831 | tok/s 2633072.24 | norm 0.4114 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8862 | train 2.996202 | val 3.381831 | tok/s 2634148.61 | norm 0.3775 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8863 | train 3.056147 | val 3.381831 | tok/s 2632925.07 | norm 0.4074 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8864 | train 2.867122 | val 3.381831 | tok/s 2633879.44 | norm 0.3874 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8865 | train 3.017321 | val 3.381831 | tok/s 2628014.74 | norm 0.3796 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8866 | train 2.994994 | val 3.381831 | tok/s 2624113.87 | norm 0.3636 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8867 | train 3.002578 | val 3.381831 | tok/s 2626899.21 | norm 0.3585 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8868 | train 3.066278 | val 3.381831 | tok/s 2630965.79 | norm 0.3663 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8869 | train 3.133248 | val 3.381831 | tok/s 2629012.75 | norm 0.4069 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8870 | train 2.956667 | val 3.381831 | tok/s 2627366.77 | norm 0.3880 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8871 | train 3.078958 | val 3.381831 | tok/s 2629875.14 | norm 0.3798 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8872 | train 2.999048 | val 3.381831 | tok/s 2627918.75 | norm 0.3456 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8873 | train 3.081708 | val 3.381831 | tok/s 2629337.94 | norm 0.3646 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8874 | train 3.019664 | val 3.381831 | tok/s 2629058.31 | norm 0.4020 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8875 | train 3.082101 | val 3.381831 | tok/s 2630152.47 | norm 0.4160 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8876 | train 2.984443 | val 3.381831 | tok/s 2630089.51 | norm 0.4114 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8877 | train 2.999656 | val 3.381831 | tok/s 2627469.94 | norm 0.4388 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8878 | train 2.896100 | val 3.381831 | tok/s 2621686.88 | norm 0.4080 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8879 | train 2.945711 | val 3.381831 | tok/s 2628129.35 | norm 0.4077 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8880 | train 2.984017 | val 3.381831 | tok/s 2628075.56 | norm 0.3934 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8881 | train 3.102206 | val 3.381831 | tok/s 2631516.99 | norm 0.4613 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8882 | train 3.018567 | val 3.381831 | tok/s 2627435.87 | norm 0.4351 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8883 | train 3.249909 | val 3.381831 | tok/s 2630793.59 | norm 0.4194 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8884 | train 3.091804 | val 3.381831 | tok/s 2634108.69 | norm 0.3971 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8885 | train 2.966578 | val 3.381831 | tok/s 2627837.02 | norm 0.4010 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8886 | train 3.059832 | val 3.381831 | tok/s 2631000.79 | norm 0.5136 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8887 | train 2.902153 | val 3.381831 | tok/s 2633766.25 | norm 0.3590 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8888 | train 2.951574 | val 3.381831 | tok/s 2630335.39 | norm 0.4515 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8889 | train 2.932443 | val 3.381831 | tok/s 2630079.46 | norm 0.4306 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8890 | train 2.970975 | val 3.381831 | tok/s 2631357.45 | norm 0.3819 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8891 | train 2.998103 | val 3.381831 | tok/s 2629355.25 | norm 0.4036 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8892 | train 2.833036 | val 3.381831 | tok/s 2626561.69 | norm 0.4098 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8893 | train 3.036442 | val 3.381831 | tok/s 2627203.97 | norm 0.4214 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8894 | train 3.045353 | val 3.381831 | tok/s 2627710.47 | norm 0.4185 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8895 | train 2.951742 | val 3.381831 | tok/s 2626308.45 | norm 0.4333 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8896 | train 2.939052 | val 3.381831 | tok/s 2622787.51 | norm 0.4226 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8897 | train 2.949260 | val 3.381831 | tok/s 2632247.60 | norm 0.4018 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8898 | train 3.007259 | val 3.381831 | tok/s 2633925.00 | norm 0.4043 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8899 | train 3.004931 | val 3.381831 | tok/s 2627810.19 | norm 0.3889 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8900 | train 3.050176 | val 3.381831 | tok/s 2630136.77 | norm 0.3693 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8901 | train 2.983348 | val 3.381831 | tok/s 2629169.71 | norm 0.4189 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8902 | train 3.130534 | val 3.381831 | tok/s 2625271.56 | norm 0.3942 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8903 | train 2.952335 | val 3.381831 | tok/s 2624401.66 | norm 0.3589 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8904 | train 3.117234 | val 3.381831 | tok/s 2626972.67 | norm 0.3764 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8905 | train 3.041451 | val 3.381831 | tok/s 2629884.28 | norm 0.3600 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8906 | train 2.942582 | val 3.381831 | tok/s 2627046.72 | norm 0.3838 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8907 | train 3.045749 | val 3.381831 | tok/s 2627800.04 | norm 0.3914 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8908 | train 2.921717 | val 3.381831 | tok/s 2623617.98 | norm 0.4727 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8909 | train 3.038572 | val 3.381831 | tok/s 2626606.90 | norm 0.4258 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8910 | train 3.034141 | val 3.381831 | tok/s 2633592.52 | norm 0.3832 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8911 | train 3.014640 | val 3.381831 | tok/s 2626399.51 | norm 0.4476 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8912 | train 3.085961 | val 3.381831 | tok/s 2556865.22 | norm 0.4483 | dt 205.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8913 | train 2.958687 | val 3.381831 | tok/s 2627905.34 | norm 0.3756 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8914 | train 3.048762 | val 3.381831 | tok/s 2627741.50 | norm 0.3869 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8915 | train 3.106338 | val 3.381831 | tok/s 2626858.17 | norm 0.4008 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8916 | train 3.141327 | val 3.381831 | tok/s 2626923.03 | norm 0.3838 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8917 | train 3.073797 | val 3.381831 | tok/s 2630186.91 | norm 0.4327 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8918 | train 3.205622 | val 3.381831 | tok/s 2626692.95 | norm 0.4440 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8919 | train 3.030167 | val 3.381831 | tok/s 2624296.15 | norm 0.4261 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8920 | train 2.975374 | val 3.381831 | tok/s 2590739.12 | norm 0.3746 | dt 202.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8921 | train 3.064719 | val 3.381831 | tok/s 2629287.73 | norm 0.4511 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8922 | train 3.102059 | val 3.381831 | tok/s 2629796.19 | norm 0.4196 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8923 | train 3.011847 | val 3.381831 | tok/s 2627100.61 | norm 0.3939 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8924 | train 3.029724 | val 3.381831 | tok/s 2625660.17 | norm 0.4023 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8925 | train 3.000738 | val 3.381831 | tok/s 2620933.11 | norm 0.4395 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8926 | train 3.026041 | val 3.381831 | tok/s 2620992.83 | norm 0.4078 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8927 | train 3.167202 | val 3.381831 | tok/s 2628713.56 | norm 0.4025 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8928 | train 2.971725 | val 3.381831 | tok/s 2628488.74 | norm 0.3871 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8929 | train 2.974422 | val 3.381831 | tok/s 2623751.60 | norm 0.3603 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8930 | train 3.018424 | val 3.381831 | tok/s 2465102.46 | norm 0.3864 | dt 212.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8931 | train 2.894851 | val 3.381831 | tok/s 2601473.07 | norm 0.4279 | dt 201.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8932 | train 2.945080 | val 3.381831 | tok/s 2623202.56 | norm 0.4111 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8933 | train 2.994308 | val 3.381831 | tok/s 2625353.21 | norm 0.3706 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8934 | train 3.046339 | val 3.381831 | tok/s 2625762.94 | norm 0.3940 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8935 | train 3.042060 | val 3.381831 | tok/s 2629792.00 | norm 0.3787 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8936 | train 3.116445 | val 3.381831 | tok/s 2627163.00 | norm 0.4101 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8937 | train 2.930793 | val 3.381831 | tok/s 2627532.85 | norm 0.4081 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8938 | train 2.904466 | val 3.381831 | tok/s 2628690.99 | norm 0.4175 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8939 | train 3.178020 | val 3.381831 | tok/s 2630680.94 | norm 0.4857 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8940 | train 2.979133 | val 3.381831 | tok/s 2625198.71 | norm 0.3987 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8941 | train 2.978868 | val 3.381831 | tok/s 2554687.55 | norm 0.3729 | dt 205.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8942 | train 3.113366 | val 3.381831 | tok/s 2623780.25 | norm 0.4502 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8943 | train 2.965528 | val 3.381831 | tok/s 2480022.86 | norm 0.4150 | dt 211.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8944 | train 3.065968 | val 3.381831 | tok/s 2628973.60 | norm 0.3931 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8945 | train 3.074356 | val 3.381831 | tok/s 2632017.91 | norm 0.4259 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8946 | train 2.982042 | val 3.381831 | tok/s 2461153.38 | norm 0.4375 | dt 213.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8947 | train 3.020315 | val 3.381831 | tok/s 2619820.75 | norm 0.4535 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8948 | train 2.856969 | val 3.381831 | tok/s 2603499.48 | norm 0.4491 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8949 | train 2.792842 | val 3.381831 | tok/s 2626292.72 | norm 0.4240 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8950 | train 2.893354 | val 3.381831 | tok/s 2625361.56 | norm 0.4383 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8951 | train 2.881618 | val 3.381831 | tok/s 2628185.05 | norm 0.4366 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8952 | train 2.960690 | val 3.381831 | tok/s 2624760.80 | norm 0.4166 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8953 | train 3.087423 | val 3.381831 | tok/s 2489278.67 | norm 0.3896 | dt 210.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8954 | train 3.144021 | val 3.381831 | tok/s 2599964.02 | norm 0.4090 | dt 201.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8955 | train 2.964630 | val 3.381831 | tok/s 2626164.97 | norm 0.4308 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8956 | train 3.003057 | val 3.381831 | tok/s 2629090.80 | norm 0.4183 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8957 | train 3.043699 | val 3.381831 | tok/s 2629501.87 | norm 0.4141 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8958 | train 3.126125 | val 3.381831 | tok/s 2482299.05 | norm 0.4372 | dt 211.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8959 | train 2.946136 | val 3.381831 | tok/s 2598902.82 | norm 0.4099 | dt 201.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8960 | train 3.241773 | val 3.381831 | tok/s 2459430.72 | norm 0.5105 | dt 213.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8961 | train 2.953424 | val 3.381831 | tok/s 2611440.27 | norm 0.4272 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8962 | train 3.129866 | val 3.381831 | tok/s 2628007.10 | norm 0.3816 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8963 | train 2.959432 | val 3.381831 | tok/s 2628975.14 | norm 0.4518 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8964 | train 2.945002 | val 3.381831 | tok/s 2473091.92 | norm 0.4282 | dt 212.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8965 | train 2.972233 | val 3.381831 | tok/s 2602198.32 | norm 0.3956 | dt 201.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8966 | train 2.919975 | val 3.381831 | tok/s 2628482.73 | norm 0.3845 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8967 | train 2.925250 | val 3.381831 | tok/s 2626583.72 | norm 0.3953 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8968 | train 3.048713 | val 3.381831 | tok/s 2628403.03 | norm 0.4238 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8969 | train 3.076325 | val 3.381831 | tok/s 2627962.35 | norm 0.4143 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8970 | train 3.006423 | val 3.381831 | tok/s 2625078.25 | norm 0.4484 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8971 | train 3.037054 | val 3.381831 | tok/s 2624546.98 | norm 0.4171 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8972 | train 3.270472 | val 3.381831 | tok/s 2611029.01 | norm 0.4435 | dt 200.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8973 | train 3.053082 | val 3.381831 | tok/s 2628004.89 | norm 0.4154 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8974 | train 2.957045 | val 3.381831 | tok/s 2616227.65 | norm 0.3931 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8975 | train 3.038340 | val 3.381831 | tok/s 2624445.15 | norm 0.4494 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8976 | train 3.019267 | val 3.381831 | tok/s 2631356.15 | norm 0.4432 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8977 | train 3.053823 | val 3.381831 | tok/s 2626945.85 | norm 0.4132 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8978 | train 3.039622 | val 3.381831 | tok/s 2629636.51 | norm 0.3827 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8979 | train 2.951240 | val 3.381831 | tok/s 2626548.65 | norm 0.3674 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8980 | train 3.031242 | val 3.381831 | tok/s 2624942.97 | norm 0.4080 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8981 | train 3.052508 | val 3.381831 | tok/s 2626739.47 | norm 0.4418 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8982 | train 3.023599 | val 3.381831 | tok/s 2629213.98 | norm 0.4249 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8983 | train 3.086569 | val 3.381831 | tok/s 2628085.50 | norm 0.4308 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8984 | train 2.986759 | val 3.381831 | tok/s 2631766.33 | norm 0.4236 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8985 | train 2.938080 | val 3.381831 | tok/s 2629453.31 | norm 0.3878 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8986 | train 2.992208 | val 3.381831 | tok/s 2629308.55 | norm 0.3845 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8987 | train 3.024765 | val 3.381831 | tok/s 2629257.23 | norm 0.4185 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8988 | train 2.993332 | val 3.381831 | tok/s 2630407.24 | norm 0.3997 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8989 | train 3.026074 | val 3.381831 | tok/s 2619481.96 | norm 0.3853 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8990 | train 2.979409 | val 3.381831 | tok/s 2626326.03 | norm 0.3945 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8991 | train 2.955597 | val 3.381831 | tok/s 2627086.34 | norm 0.4353 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8992 | train 2.971838 | val 3.381831 | tok/s 2626740.67 | norm 0.4513 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8993 | train 2.943946 | val 3.381831 | tok/s 2624448.06 | norm 0.4528 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8994 | train 2.963835 | val 3.381831 | tok/s 2630017.62 | norm 0.4113 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8995 | train 3.084579 | val 3.381831 | tok/s 2630365.12 | norm 0.3632 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8996 | train 2.932572 | val 3.381831 | tok/s 2628301.60 | norm 0.3798 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8997 | train 3.050194 | val 3.381831 | tok/s 2631379.71 | norm 0.4261 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8998 | train 2.935272 | val 3.381831 | tok/s 2624428.37 | norm 0.4467 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 8999 | train 2.991681 | val 3.381831 | tok/s 2619453.55 | norm 0.4203 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 9000 | val_loss 3.369910 step 9000 | train 3.029495 | val 3.369910 | tok/s 2613146.18 | norm 0.3963 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9001 | train 2.963220 | val 3.369910 | tok/s 2627471.39 | norm 0.4075 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9002 | train 3.084222 | val 3.369910 | tok/s 2627104.85 | norm 0.4141 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9003 | train 2.918472 | val 3.369910 | tok/s 2632353.53 | norm 0.3852 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9004 | train 2.978806 | val 3.369910 | tok/s 2633950.66 | norm 0.3975 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9005 | train 3.024521 | val 3.369910 | tok/s 2626162.69 | norm 0.4053 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9006 | train 2.939745 | val 3.369910 | tok/s 2628109.41 | norm 0.4378 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9007 | train 3.014885 | val 3.369910 | tok/s 2623620.43 | norm 0.4774 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9008 | train 2.972189 | val 3.369910 | tok/s 2627423.75 | norm 0.4784 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9009 | train 3.056301 | val 3.369910 | tok/s 2628831.29 | norm 0.4037 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9010 | train 2.985701 | val 3.369910 | tok/s 2629297.99 | norm 0.4095 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9011 | train 3.006038 | val 3.369910 | tok/s 2629754.34 | norm 0.4295 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9012 | train 3.072757 | val 3.369910 | tok/s 2629261.88 | norm 0.4184 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9013 | train 3.088090 | val 3.369910 | tok/s 2629161.33 | norm 0.4326 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9014 | train 3.085933 | val 3.369910 | tok/s 2624192.27 | norm 0.4367 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9015 | train 3.018295 | val 3.369910 | tok/s 2626465.10 | norm 0.4210 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9016 | train 3.053258 | val 3.369910 | tok/s 2626739.64 | norm 0.3756 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9017 | train 3.008461 | val 3.369910 | tok/s 2627993.24 | norm 0.3758 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9018 | train 3.065161 | val 3.369910 | tok/s 2629013.58 | norm 0.4312 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9019 | train 3.020337 | val 3.369910 | tok/s 2628788.23 | norm 0.4316 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9020 | train 2.986415 | val 3.369910 | tok/s 2628488.56 | norm 0.3851 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9021 | train 2.991314 | val 3.369910 | tok/s 2624204.69 | norm 0.4084 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9022 | train 2.927642 | val 3.369910 | tok/s 2628849.97 | norm 0.4250 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9023 | train 2.973555 | val 3.369910 | tok/s 2507311.15 | norm 0.4086 | dt 209.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9024 | train 3.210212 | val 3.369910 | tok/s 2474090.91 | norm 0.5287 | dt 211.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9025 | train 2.925850 | val 3.369910 | tok/s 2630677.77 | norm 0.4437 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9026 | train 2.955603 | val 3.369910 | tok/s 2628492.26 | norm 0.4027 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9027 | train 3.238547 | val 3.369910 | tok/s 2631769.66 | norm 0.6783 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9028 | train 2.928360 | val 3.369910 | tok/s 2625459.11 | norm 0.4050 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9029 | train 3.016192 | val 3.369910 | tok/s 2630292.67 | norm 0.3874 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9030 | train 3.004072 | val 3.369910 | tok/s 2627766.10 | norm 0.4197 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9031 | train 2.977603 | val 3.369910 | tok/s 2626906.74 | norm 0.4948 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9032 | train 2.962241 | val 3.369910 | tok/s 2630384.58 | norm 0.3960 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9033 | train 3.021352 | val 3.369910 | tok/s 2626467.82 | norm 0.3714 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9034 | train 3.151386 | val 3.369910 | tok/s 2631430.09 | norm 0.4171 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9035 | train 3.034525 | val 3.369910 | tok/s 2625667.78 | norm 0.3799 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9036 | train 2.957726 | val 3.369910 | tok/s 2627470.44 | norm 0.3514 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9037 | train 3.026364 | val 3.369910 | tok/s 2619841.23 | norm 0.4028 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9038 | train 3.160463 | val 3.369910 | tok/s 2619540.92 | norm 0.4093 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9039 | train 2.993824 | val 3.369910 | tok/s 2626349.78 | norm 0.3935 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9040 | train 2.976443 | val 3.369910 | tok/s 2624396.21 | norm 0.3665 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9041 | train 2.987478 | val 3.369910 | tok/s 2627932.58 | norm 0.3721 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9042 | train 2.974475 | val 3.369910 | tok/s 2630508.44 | norm 0.3920 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9043 | train 3.040322 | val 3.369910 | tok/s 2628576.86 | norm 0.4453 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9044 | train 2.970378 | val 3.369910 | tok/s 2628822.44 | norm 0.4516 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9045 | train 3.002614 | val 3.369910 | tok/s 2633148.58 | norm 0.4303 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9046 | train 3.177720 | val 3.369910 | tok/s 2626845.38 | norm 0.4117 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9047 | train 2.943156 | val 3.369910 | tok/s 2624248.63 | norm 0.3877 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9048 | train 3.007646 | val 3.369910 | tok/s 2629358.81 | norm 0.3802 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9049 | train 3.021123 | val 3.369910 | tok/s 2629516.90 | norm 0.4115 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9050 | train 2.943364 | val 3.369910 | tok/s 2630965.42 | norm 0.4077 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9051 | train 3.025949 | val 3.369910 | tok/s 2629292.67 | norm 0.3727 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9052 | train 2.891427 | val 3.369910 | tok/s 2626841.31 | norm 0.3896 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9053 | train 3.092717 | val 3.369910 | tok/s 2629281.24 | norm 0.4291 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9054 | train 2.995759 | val 3.369910 | tok/s 2632278.86 | norm 0.4092 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9055 | train 2.979955 | val 3.369910 | tok/s 2627777.20 | norm 0.3934 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9056 | train 3.370085 | val 3.369910 | tok/s 2627717.60 | norm 0.5096 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9057 | train 3.063902 | val 3.369910 | tok/s 2633165.63 | norm 0.4155 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9058 | train 2.869704 | val 3.369910 | tok/s 2633420.26 | norm 0.4094 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9059 | train 3.010411 | val 3.369910 | tok/s 2632340.88 | norm 0.4306 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9060 | train 3.011887 | val 3.369910 | tok/s 2631184.16 | norm 0.4403 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9061 | train 3.129025 | val 3.369910 | tok/s 2624599.12 | norm 0.4351 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9062 | train 2.978924 | val 3.369910 | tok/s 2625625.06 | norm 0.4333 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9063 | train 3.053103 | val 3.369910 | tok/s 2630191.74 | norm 0.4366 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9064 | train 2.939775 | val 3.369910 | tok/s 2632759.40 | norm 0.3911 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9065 | train 2.895294 | val 3.369910 | tok/s 2622893.91 | norm 0.4110 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9066 | train 2.970591 | val 3.369910 | tok/s 2626400.51 | norm 0.3931 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9067 | train 2.980271 | val 3.369910 | tok/s 2632003.45 | norm 0.3930 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9068 | train 2.940108 | val 3.369910 | tok/s 2627178.23 | norm 0.4158 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9069 | train 3.086012 | val 3.369910 | tok/s 2623909.95 | norm 0.3897 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9070 | train 3.056484 | val 3.369910 | tok/s 2625227.92 | norm 0.4244 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9071 | train 2.995982 | val 3.369910 | tok/s 2630466.51 | norm 0.4030 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9072 | train 3.057239 | val 3.369910 | tok/s 2628155.12 | norm 0.3956 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9073 | train 3.061243 | val 3.369910 | tok/s 2632409.53 | norm 0.4308 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9074 | train 3.037501 | val 3.369910 | tok/s 2627267.73 | norm 0.3729 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9075 | train 2.947954 | val 3.369910 | tok/s 2631986.48 | norm 0.3444 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9076 | train 2.947274 | val 3.369910 | tok/s 2630840.12 | norm 0.3513 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9077 | train 3.108754 | val 3.369910 | tok/s 2605837.48 | norm 0.3782 | dt 201.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9078 | train 3.118126 | val 3.369910 | tok/s 2633489.33 | norm 0.4097 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9079 | train 3.021363 | val 3.369910 | tok/s 2628022.76 | norm 0.4445 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9080 | train 3.204747 | val 3.369910 | tok/s 2626177.87 | norm 0.4717 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9081 | train 2.939905 | val 3.369910 | tok/s 2630535.18 | norm 0.4083 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9082 | train 3.070145 | val 3.369910 | tok/s 2628668.15 | norm 0.3932 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9083 | train 2.962540 | val 3.369910 | tok/s 2628640.95 | norm 0.4154 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9084 | train 2.912364 | val 3.369910 | tok/s 2630476.46 | norm 0.4246 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9085 | train 3.032149 | val 3.369910 | tok/s 2626744.28 | norm 0.4202 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9086 | train 3.050081 | val 3.369910 | tok/s 2627682.51 | norm 0.3813 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9087 | train 3.028577 | val 3.369910 | tok/s 2489062.91 | norm 0.4098 | dt 210.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9088 | train 3.043352 | val 3.369910 | tok/s 2629035.32 | norm 0.4320 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9089 | train 3.031099 | val 3.369910 | tok/s 2628915.06 | norm 0.4798 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9090 | train 3.004129 | val 3.369910 | tok/s 2629259.36 | norm 0.4458 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9091 | train 3.058148 | val 3.369910 | tok/s 2632178.92 | norm 0.4342 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9092 | train 3.061980 | val 3.369910 | tok/s 2629057.14 | norm 0.4244 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9093 | train 3.049267 | val 3.369910 | tok/s 2630999.17 | norm 0.3826 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9094 | train 3.027216 | val 3.369910 | tok/s 2630200.17 | norm 0.3691 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9095 | train 3.059604 | val 3.369910 | tok/s 2620670.01 | norm 0.3970 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9096 | train 3.069983 | val 3.369910 | tok/s 2632042.94 | norm 0.3898 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9097 | train 2.974761 | val 3.369910 | tok/s 2627142.71 | norm 0.3954 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9098 | train 3.085706 | val 3.369910 | tok/s 2634406.48 | norm 0.4127 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9099 | train 3.034348 | val 3.369910 | tok/s 2631155.36 | norm 0.4057 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9100 | train 3.054002 | val 3.369910 | tok/s 2631583.16 | norm 0.4091 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9101 | train 3.051050 | val 3.369910 | tok/s 2632900.44 | norm 0.4123 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9102 | train 2.965446 | val 3.369910 | tok/s 2630919.35 | norm 0.3555 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9103 | train 3.012625 | val 3.369910 | tok/s 2626341.55 | norm 0.3511 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9104 | train 3.017263 | val 3.369910 | tok/s 2622184.54 | norm 0.3863 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9105 | train 3.049310 | val 3.369910 | tok/s 2628612.97 | norm 0.4089 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9106 | train 2.894541 | val 3.369910 | tok/s 2624189.93 | norm 0.4123 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9107 | train 3.202517 | val 3.369910 | tok/s 2630952.91 | norm 0.4184 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9108 | train 3.027030 | val 3.369910 | tok/s 2626385.29 | norm 0.4452 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9109 | train 2.942075 | val 3.369910 | tok/s 2629269.31 | norm 0.4111 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9110 | train 3.051742 | val 3.369910 | tok/s 2628231.86 | norm 0.4024 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9111 | train 2.982212 | val 3.369910 | tok/s 2629496.29 | norm 0.4046 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9112 | train 2.953554 | val 3.369910 | tok/s 2627764.31 | norm 0.3949 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9113 | train 2.997713 | val 3.369910 | tok/s 2630521.88 | norm 0.4014 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9114 | train 2.967647 | val 3.369910 | tok/s 2632786.52 | norm 0.4086 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9115 | train 3.022400 | val 3.369910 | tok/s 2632307.56 | norm 0.4065 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9116 | train 3.020857 | val 3.369910 | tok/s 2632192.07 | norm 0.4179 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9117 | train 3.118074 | val 3.369910 | tok/s 2632550.27 | norm 0.4630 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9118 | train 3.054517 | val 3.369910 | tok/s 2627624.13 | norm 0.4289 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9119 | train 2.963595 | val 3.369910 | tok/s 2627547.89 | norm 0.3953 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9120 | train 3.024845 | val 3.369910 | tok/s 2626508.45 | norm 0.4131 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9121 | train 2.988200 | val 3.369910 | tok/s 2618759.63 | norm 0.4500 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9122 | train 3.017416 | val 3.369910 | tok/s 2621575.23 | norm 0.4676 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9123 | train 3.133950 | val 3.369910 | tok/s 2628431.91 | norm 0.4008 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9124 | train 2.962158 | val 3.369910 | tok/s 2631950.59 | norm 0.3977 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9125 | train 3.134977 | val 3.369910 | tok/s 2632777.32 | norm 0.4170 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9126 | train 3.163726 | val 3.369910 | tok/s 2633548.71 | norm 0.4230 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9127 | train 3.108963 | val 3.369910 | tok/s 2628562.25 | norm 0.4039 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9128 | train 3.105548 | val 3.369910 | tok/s 2630959.73 | norm 0.4415 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9129 | train 2.918226 | val 3.369910 | tok/s 2628585.44 | norm 0.4086 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9130 | train 3.009063 | val 3.369910 | tok/s 2631582.20 | norm 0.3951 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9131 | train 2.956517 | val 3.369910 | tok/s 2633527.99 | norm 0.3840 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9132 | train 2.991244 | val 3.369910 | tok/s 2629953.17 | norm 0.4316 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9133 | train 3.089396 | val 3.369910 | tok/s 2630930.70 | norm 0.3839 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9134 | train 2.982243 | val 3.369910 | tok/s 2626063.61 | norm 0.4096 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9135 | train 3.098992 | val 3.369910 | tok/s 2624624.50 | norm 0.4531 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9136 | train 2.870245 | val 3.369910 | tok/s 2629234.23 | norm 0.3982 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9137 | train 3.004801 | val 3.369910 | tok/s 2630147.38 | norm 0.3736 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9138 | train 3.044857 | val 3.369910 | tok/s 2544140.69 | norm 0.4252 | dt 206.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9139 | train 2.943483 | val 3.369910 | tok/s 2516355.55 | norm 0.4133 | dt 208.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9140 | train 3.146817 | val 3.369910 | tok/s 2620394.00 | norm 0.5284 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9141 | train 2.969305 | val 3.369910 | tok/s 2630531.80 | norm 0.4254 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9142 | train 2.996797 | val 3.369910 | tok/s 2629783.16 | norm 0.3689 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9143 | train 3.013414 | val 3.369910 | tok/s 2630183.31 | norm 0.3760 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9144 | train 3.061022 | val 3.369910 | tok/s 2631496.58 | norm 0.4965 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9145 | train 3.154830 | val 3.369910 | tok/s 2631483.73 | norm 0.4078 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9146 | train 2.982430 | val 3.369910 | tok/s 2635039.23 | norm 0.3647 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9147 | train 3.107350 | val 3.369910 | tok/s 2513179.91 | norm 0.4148 | dt 208.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9148 | train 2.952779 | val 3.369910 | tok/s 2593946.07 | norm 0.4451 | dt 202.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9149 | train 3.116813 | val 3.369910 | tok/s 2627584.51 | norm 0.4114 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9150 | train 3.044584 | val 3.369910 | tok/s 2627324.51 | norm 0.3446 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9151 | train 2.959319 | val 3.369910 | tok/s 2631141.19 | norm 0.3517 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9152 | train 3.051281 | val 3.369910 | tok/s 2628535.52 | norm 0.4353 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9153 | train 3.033676 | val 3.369910 | tok/s 2628512.01 | norm 0.4350 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9154 | train 3.141861 | val 3.369910 | tok/s 2630274.96 | norm 0.4110 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9155 | train 2.934892 | val 3.369910 | tok/s 2619478.89 | norm 0.4266 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9156 | train 2.956381 | val 3.369910 | tok/s 2628881.30 | norm 0.3785 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9157 | train 2.904418 | val 3.369910 | tok/s 2624344.32 | norm 0.3698 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9158 | train 2.985899 | val 3.369910 | tok/s 2629620.33 | norm 0.3889 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9159 | train 3.061989 | val 3.369910 | tok/s 2628524.20 | norm 0.4059 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9160 | train 3.019467 | val 3.369910 | tok/s 2628059.09 | norm 0.4140 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9161 | train 2.972815 | val 3.369910 | tok/s 2627385.04 | norm 0.4001 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9162 | train 3.070834 | val 3.369910 | tok/s 2626615.26 | norm 0.4671 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9163 | train 3.016750 | val 3.369910 | tok/s 2627595.48 | norm 0.3601 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9164 | train 3.009374 | val 3.369910 | tok/s 2628902.11 | norm 0.3557 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9165 | train 2.975235 | val 3.369910 | tok/s 2623802.79 | norm 0.4018 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9166 | train 3.093446 | val 3.369910 | tok/s 2621647.49 | norm 0.4110 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9167 | train 2.995764 | val 3.369910 | tok/s 2622638.66 | norm 0.4069 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9168 | train 3.035328 | val 3.369910 | tok/s 2627783.63 | norm 0.3973 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9169 | train 2.964150 | val 3.369910 | tok/s 2627635.83 | norm 0.4157 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9170 | train 2.990138 | val 3.369910 | tok/s 2630353.41 | norm 0.5487 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9171 | train 2.945088 | val 3.369910 | tok/s 2633251.36 | norm 0.3503 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9172 | train 3.025876 | val 3.369910 | tok/s 2634266.80 | norm 0.4047 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9173 | train 3.151877 | val 3.369910 | tok/s 2629354.16 | norm 0.4625 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9174 | train 3.305316 | val 3.369910 | tok/s 2627224.44 | norm 0.4435 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9175 | train 2.998428 | val 3.369910 | tok/s 2630770.55 | norm 0.4399 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9176 | train 3.111971 | val 3.369910 | tok/s 2627051.39 | norm 0.3853 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9177 | train 3.154299 | val 3.369910 | tok/s 2626091.39 | norm 0.3944 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9178 | train 3.026927 | val 3.369910 | tok/s 2629021.47 | norm 0.3983 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9179 | train 2.955161 | val 3.369910 | tok/s 2629737.72 | norm 0.3982 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9180 | train 3.074247 | val 3.369910 | tok/s 2625660.17 | norm 0.4104 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9181 | train 2.976777 | val 3.369910 | tok/s 2628693.45 | norm 0.4112 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9182 | train 2.944157 | val 3.369910 | tok/s 2627269.31 | norm 0.4061 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9183 | train 3.001319 | val 3.369910 | tok/s 2630628.31 | norm 0.4479 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9184 | train 3.061368 | val 3.369910 | tok/s 2631623.62 | norm 0.4020 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9185 | train 2.964314 | val 3.369910 | tok/s 2631804.15 | norm 0.3691 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9186 | train 2.957171 | val 3.369910 | tok/s 2620581.25 | norm 0.3956 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9187 | train 3.010866 | val 3.369910 | tok/s 2627221.40 | norm 0.4146 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9188 | train 2.981889 | val 3.369910 | tok/s 2628369.21 | norm 0.3562 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9189 | train 2.987574 | val 3.369910 | tok/s 2625710.30 | norm 0.3992 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9190 | train 2.998032 | val 3.369910 | tok/s 2627948.77 | norm 0.4704 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9191 | train 3.056664 | val 3.369910 | tok/s 2631604.26 | norm 0.4371 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9192 | train 2.931743 | val 3.369910 | tok/s 2630296.52 | norm 0.4140 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9193 | train 3.028858 | val 3.369910 | tok/s 2632459.81 | norm 0.4038 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9194 | train 3.141984 | val 3.369910 | tok/s 2627735.38 | norm 0.4014 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9195 | train 3.064673 | val 3.369910 | tok/s 2547479.12 | norm 0.4135 | dt 205.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9196 | train 2.931379 | val 3.369910 | tok/s 2625035.47 | norm 0.4029 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9197 | train 2.927849 | val 3.369910 | tok/s 2565170.26 | norm 0.4095 | dt 204.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9198 | train 2.958811 | val 3.369910 | tok/s 2451985.60 | norm 0.4046 | dt 213.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9199 | train 2.938387 | val 3.369910 | tok/s 2618658.96 | norm 0.3675 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9200 | train 3.029698 | val 3.369910 | tok/s 2627815.74 | norm 0.3934 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9201 | train 3.074003 | val 3.369910 | tok/s 2628315.92 | norm 0.4139 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9202 | train 3.011008 | val 3.369910 | tok/s 2628917.05 | norm 0.4222 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9203 | train 3.012822 | val 3.369910 | tok/s 2623669.41 | norm 0.4356 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9204 | train 3.062758 | val 3.369910 | tok/s 2626284.63 | norm 0.3980 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9205 | train 2.967704 | val 3.369910 | tok/s 2582459.43 | norm 0.3867 | dt 203.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9206 | train 2.862175 | val 3.369910 | tok/s 2633165.67 | norm 0.4304 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9207 | train 3.023976 | val 3.369910 | tok/s 2615380.61 | norm 0.4498 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9208 | train 3.163596 | val 3.369910 | tok/s 2624197.91 | norm 0.4751 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9209 | train 2.997810 | val 3.369910 | tok/s 2628206.92 | norm 0.4019 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9210 | train 2.967792 | val 3.369910 | tok/s 2631461.98 | norm 0.3837 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9211 | train 3.012317 | val 3.369910 | tok/s 2631835.98 | norm 0.3833 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9212 | train 2.969682 | val 3.369910 | tok/s 2626066.66 | norm 0.4087 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9213 | train 2.973976 | val 3.369910 | tok/s 2632377.59 | norm 0.4201 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9214 | train 3.001090 | val 3.369910 | tok/s 2630229.62 | norm 0.3638 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9215 | train 3.017617 | val 3.369910 | tok/s 2632324.53 | norm 0.3690 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9216 | train 3.045227 | val 3.369910 | tok/s 2632929.49 | norm 0.3890 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9217 | train 2.977357 | val 3.369910 | tok/s 2634519.04 | norm 0.3860 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9218 | train 3.153906 | val 3.369910 | tok/s 2634014.50 | norm 0.4441 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9219 | train 3.044356 | val 3.369910 | tok/s 2630181.71 | norm 0.4287 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9220 | train 3.111323 | val 3.369910 | tok/s 2629727.45 | norm 0.3945 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9221 | train 2.980630 | val 3.369910 | tok/s 2629861.36 | norm 0.4066 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9222 | train 2.903478 | val 3.369910 | tok/s 2628181.42 | norm 0.4039 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9223 | train 2.875413 | val 3.369910 | tok/s 2623809.83 | norm 0.4077 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9224 | train 2.910325 | val 3.369910 | tok/s 2622038.40 | norm 0.4428 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9225 | train 3.054574 | val 3.369910 | tok/s 2626609.76 | norm 0.4146 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9226 | train 3.008111 | val 3.369910 | tok/s 2629926.31 | norm 0.4047 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9227 | train 3.054435 | val 3.369910 | tok/s 2629242.87 | norm 0.4024 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9228 | train 3.066100 | val 3.369910 | tok/s 2624963.65 | norm 0.4269 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9229 | train 2.900447 | val 3.369910 | tok/s 2627406.06 | norm 0.4599 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9230 | train 2.903823 | val 3.369910 | tok/s 2614938.88 | norm 0.4090 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9231 | train 3.147885 | val 3.369910 | tok/s 2630796.44 | norm 0.4245 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9232 | train 2.913990 | val 3.369910 | tok/s 2627258.25 | norm 0.4391 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9233 | train 2.919667 | val 3.369910 | tok/s 2627281.95 | norm 0.4012 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9234 | train 2.971318 | val 3.369910 | tok/s 2629346.83 | norm 0.4117 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9235 | train 3.046978 | val 3.369910 | tok/s 2630288.79 | norm 0.4022 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9236 | train 2.873655 | val 3.369910 | tok/s 2629953.95 | norm 0.4029 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9237 | train 2.966456 | val 3.369910 | tok/s 2628918.33 | norm 0.4160 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9238 | train 3.104331 | val 3.369910 | tok/s 2628842.56 | norm 0.4406 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9239 | train 3.031613 | val 3.369910 | tok/s 2627711.02 | norm 0.3990 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9240 | train 3.015273 | val 3.369910 | tok/s 2630846.31 | norm 0.3556 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9241 | train 3.023932 | val 3.369910 | tok/s 2631234.15 | norm 0.3764 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9242 | train 2.976343 | val 3.369910 | tok/s 2629477.66 | norm 0.3682 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9243 | train 3.030160 | val 3.369910 | tok/s 2627034.84 | norm 0.3701 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9244 | train 2.964620 | val 3.369910 | tok/s 2620468.64 | norm 0.3817 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9245 | train 2.962676 | val 3.369910 | tok/s 2626613.76 | norm 0.3783 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9246 | train 2.994100 | val 3.369910 | tok/s 2625108.18 | norm 0.3542 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9247 | train 3.023879 | val 3.369910 | tok/s 2630205.32 | norm 0.3763 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9248 | train 3.102746 | val 3.369910 | tok/s 2627614.68 | norm 0.4616 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9249 | train 3.024388 | val 3.369910 | tok/s 2539047.01 | norm 0.4048 | dt 206.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 9250 | val_loss 3.367732 step 9250 | train 2.938308 | val 3.367732 | tok/s 2616282.99 | norm 0.3402 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9251 | train 2.932321 | val 3.367732 | tok/s 2629363.19 | norm 0.4040 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9252 | train 2.976027 | val 3.367732 | tok/s 2630228.46 | norm 0.4054 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9253 | train 2.975805 | val 3.367732 | tok/s 2630452.98 | norm 0.3503 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9254 | train 3.111135 | val 3.367732 | tok/s 2623991.11 | norm 0.4301 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9255 | train 3.102277 | val 3.367732 | tok/s 2626938.77 | norm 0.4107 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9256 | train 3.132605 | val 3.367732 | tok/s 2625033.04 | norm 0.3692 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9257 | train 2.982002 | val 3.367732 | tok/s 2620535.15 | norm 0.3802 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9258 | train 2.922852 | val 3.367732 | tok/s 2626393.81 | norm 0.3842 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9259 | train 2.972123 | val 3.367732 | tok/s 2482152.75 | norm 0.3976 | dt 211.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9260 | train 2.987246 | val 3.367732 | tok/s 2626922.06 | norm 0.4129 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9261 | train 2.947776 | val 3.367732 | tok/s 2632965.87 | norm 0.3705 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9262 | train 3.046700 | val 3.367732 | tok/s 2625020.54 | norm 0.3678 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9263 | train 3.065163 | val 3.367732 | tok/s 2630262.61 | norm 0.3855 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9264 | train 2.964204 | val 3.367732 | tok/s 2628946.89 | norm 0.3633 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9265 | train 3.010583 | val 3.367732 | tok/s 2628607.75 | norm 0.4185 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9266 | train 3.147371 | val 3.367732 | tok/s 2631490.52 | norm 0.4386 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9267 | train 2.998638 | val 3.367732 | tok/s 2631935.57 | norm 0.4326 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9268 | train 3.090463 | val 3.367732 | tok/s 2630798.91 | norm 0.4634 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9269 | train 2.862876 | val 3.367732 | tok/s 2624874.58 | norm 0.4102 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9270 | train 3.038170 | val 3.367732 | tok/s 2626846.09 | norm 0.4007 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9271 | train 2.809201 | val 3.367732 | tok/s 2626667.08 | norm 0.4032 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9272 | train 2.964160 | val 3.367732 | tok/s 2627235.14 | norm 0.4092 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9273 | train 2.970601 | val 3.367732 | tok/s 2626303.07 | norm 0.3707 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9274 | train 3.034045 | val 3.367732 | tok/s 2631550.11 | norm 0.3808 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9275 | train 2.900122 | val 3.367732 | tok/s 2629388.93 | norm 0.4405 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9276 | train 2.839350 | val 3.367732 | tok/s 2629693.25 | norm 0.5056 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9277 | train 2.939453 | val 3.367732 | tok/s 2628913.87 | norm 0.4111 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9278 | train 2.981009 | val 3.367732 | tok/s 2627032.56 | norm 0.3986 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9279 | train 3.047423 | val 3.367732 | tok/s 2628167.77 | norm 0.4406 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9280 | train 2.864703 | val 3.367732 | tok/s 2623529.01 | norm 0.3938 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9281 | train 2.935630 | val 3.367732 | tok/s 2627665.74 | norm 0.3956 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9282 | train 3.134309 | val 3.367732 | tok/s 2628773.90 | norm 0.4361 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9283 | train 3.012019 | val 3.367732 | tok/s 2625413.21 | norm 0.4061 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9284 | train 3.037880 | val 3.367732 | tok/s 2627663.12 | norm 0.3883 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9285 | train 3.241079 | val 3.367732 | tok/s 2628573.83 | norm 0.4689 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9286 | train 3.015057 | val 3.367732 | tok/s 2627959.22 | norm 0.4521 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9287 | train 2.982451 | val 3.367732 | tok/s 2629724.58 | norm 0.4176 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9288 | train 2.965255 | val 3.367732 | tok/s 2629628.20 | norm 0.4004 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9289 | train 3.014441 | val 3.367732 | tok/s 2631330.50 | norm 0.3828 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9290 | train 2.934469 | val 3.367732 | tok/s 2628913.61 | norm 0.4245 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9291 | train 3.033572 | val 3.367732 | tok/s 2630915.36 | norm 0.4221 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9292 | train 3.057559 | val 3.367732 | tok/s 2629661.81 | norm 0.3842 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9293 | train 3.063798 | val 3.367732 | tok/s 2631093.35 | norm 0.3988 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9294 | train 3.060647 | val 3.367732 | tok/s 2628942.98 | norm 0.3935 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9295 | train 3.116235 | val 3.367732 | tok/s 2628546.14 | norm 0.3429 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9296 | train 2.957773 | val 3.367732 | tok/s 2627444.05 | norm 0.3461 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9297 | train 3.009238 | val 3.367732 | tok/s 2628983.53 | norm 0.3897 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9298 | train 2.946940 | val 3.367732 | tok/s 2630119.26 | norm 0.3909 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9299 | train 3.006097 | val 3.367732 | tok/s 2628735.95 | norm 0.4011 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9300 | train 3.080964 | val 3.367732 | tok/s 2631011.80 | norm 0.4302 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9301 | train 2.965269 | val 3.367732 | tok/s 2631445.66 | norm 0.4033 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9302 | train 2.914609 | val 3.367732 | tok/s 2625457.55 | norm 0.3896 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9303 | train 2.987174 | val 3.367732 | tok/s 2627701.88 | norm 0.4430 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9304 | train 2.908906 | val 3.367732 | tok/s 2630323.25 | norm 0.4520 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9305 | train 3.022113 | val 3.367732 | tok/s 2631689.51 | norm 0.4086 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9306 | train 3.046059 | val 3.367732 | tok/s 2629209.75 | norm 0.4042 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9307 | train 3.057251 | val 3.367732 | tok/s 2628625.39 | norm 0.4201 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9308 | train 3.023385 | val 3.367732 | tok/s 2628788.25 | norm 0.4001 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9309 | train 3.066288 | val 3.367732 | tok/s 2629304.72 | norm 0.4174 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9310 | train 2.945619 | val 3.367732 | tok/s 2628869.45 | norm 0.4159 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9311 | train 3.185400 | val 3.367732 | tok/s 2628789.31 | norm 0.4089 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9312 | train 3.048097 | val 3.367732 | tok/s 2536492.06 | norm 0.4167 | dt 206.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9313 | train 2.963400 | val 3.367732 | tok/s 2629246.05 | norm 0.3991 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9314 | train 2.966046 | val 3.367732 | tok/s 2620442.21 | norm 0.4057 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9315 | train 2.993039 | val 3.367732 | tok/s 2632373.63 | norm 0.4095 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9316 | train 2.971046 | val 3.367732 | tok/s 2632604.24 | norm 0.4100 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9317 | train 3.002558 | val 3.367732 | tok/s 2630190.97 | norm 0.4383 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9318 | train 2.999623 | val 3.367732 | tok/s 2627783.06 | norm 0.4299 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9319 | train 2.939613 | val 3.367732 | tok/s 2632169.17 | norm 0.3881 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9320 | train 2.877211 | val 3.367732 | tok/s 2631644.60 | norm 0.3764 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9321 | train 2.803290 | val 3.367732 | tok/s 2633316.65 | norm 0.3916 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9322 | train 3.060065 | val 3.367732 | tok/s 2627344.14 | norm 0.4082 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9323 | train 3.039529 | val 3.367732 | tok/s 2631639.10 | norm 0.3999 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9324 | train 3.074013 | val 3.367732 | tok/s 2631386.54 | norm 0.4178 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9325 | train 2.998481 | val 3.367732 | tok/s 2628339.50 | norm 0.4266 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9326 | train 3.107569 | val 3.367732 | tok/s 2632369.00 | norm 0.3945 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9327 | train 2.915127 | val 3.367732 | tok/s 2630399.26 | norm 0.4119 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9328 | train 3.092331 | val 3.367732 | tok/s 2625959.84 | norm 0.4551 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9329 | train 2.995983 | val 3.367732 | tok/s 2628620.81 | norm 0.4218 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9330 | train 3.012592 | val 3.367732 | tok/s 2629228.33 | norm 0.3990 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9331 | train 3.083943 | val 3.367732 | tok/s 2630484.01 | norm 0.4051 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9332 | train 3.057238 | val 3.367732 | tok/s 2630787.45 | norm 0.4269 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9333 | train 3.077552 | val 3.367732 | tok/s 2631546.23 | norm 0.4027 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9334 | train 2.934124 | val 3.367732 | tok/s 2628576.95 | norm 0.3969 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9335 | train 3.011418 | val 3.367732 | tok/s 2625245.12 | norm 0.4285 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9336 | train 3.013847 | val 3.367732 | tok/s 2632817.89 | norm 0.4151 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9337 | train 3.271822 | val 3.367732 | tok/s 2633658.77 | norm 0.5730 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9338 | train 2.907629 | val 3.367732 | tok/s 2627408.92 | norm 0.4161 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9339 | train 2.998096 | val 3.367732 | tok/s 2627038.73 | norm 0.3469 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9340 | train 3.002282 | val 3.367732 | tok/s 2629256.12 | norm 0.3914 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9341 | train 2.955119 | val 3.367732 | tok/s 2629979.68 | norm 0.3922 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9342 | train 3.173954 | val 3.367732 | tok/s 2623185.22 | norm 0.4372 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9343 | train 3.058134 | val 3.367732 | tok/s 2628319.65 | norm 0.4220 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9344 | train 3.010510 | val 3.367732 | tok/s 2628718.59 | norm 0.3968 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9345 | train 3.018421 | val 3.367732 | tok/s 2632653.46 | norm 0.4036 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9346 | train 2.953703 | val 3.367732 | tok/s 2634244.64 | norm 0.5001 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9347 | train 2.926789 | val 3.367732 | tok/s 2630325.74 | norm 0.4450 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9348 | train 2.998993 | val 3.367732 | tok/s 2631524.71 | norm 0.4112 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9349 | train 3.123053 | val 3.367732 | tok/s 2631992.43 | norm 0.3958 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9350 | train 3.070673 | val 3.367732 | tok/s 2628328.51 | norm 0.4254 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9351 | train 2.927542 | val 3.367732 | tok/s 2624017.72 | norm 0.4717 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9352 | train 3.050147 | val 3.367732 | tok/s 2609409.59 | norm 0.4489 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9353 | train 3.046713 | val 3.367732 | tok/s 2624221.03 | norm 0.4390 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9354 | train 2.902875 | val 3.367732 | tok/s 2628820.64 | norm 0.4630 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9355 | train 3.082871 | val 3.367732 | tok/s 2627251.81 | norm 0.4534 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9356 | train 3.050333 | val 3.367732 | tok/s 2524093.21 | norm 0.4541 | dt 207.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9357 | train 3.007490 | val 3.367732 | tok/s 2464753.62 | norm 0.4593 | dt 212.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9358 | train 2.978263 | val 3.367732 | tok/s 2625497.55 | norm 0.4423 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9359 | train 2.984461 | val 3.367732 | tok/s 2607129.84 | norm 0.4278 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9360 | train 3.023198 | val 3.367732 | tok/s 2620270.95 | norm 0.4130 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9361 | train 3.046516 | val 3.367732 | tok/s 2626514.75 | norm 0.3884 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9362 | train 2.988024 | val 3.367732 | tok/s 2620228.87 | norm 0.4212 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9363 | train 3.026408 | val 3.367732 | tok/s 2625825.75 | norm 0.3806 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9364 | train 3.020766 | val 3.367732 | tok/s 2629253.01 | norm 0.4023 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9365 | train 2.932351 | val 3.367732 | tok/s 2202443.59 | norm 0.3915 | dt 238.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9366 | train 2.990251 | val 3.367732 | tok/s 2624627.87 | norm 0.3675 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9367 | train 2.984439 | val 3.367732 | tok/s 2629381.39 | norm 0.3978 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9368 | train 2.949896 | val 3.367732 | tok/s 2621483.39 | norm 0.3724 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9369 | train 2.974877 | val 3.367732 | tok/s 2626070.77 | norm 0.3999 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9370 | train 3.055268 | val 3.367732 | tok/s 2629520.67 | norm 0.4300 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9371 | train 2.970554 | val 3.367732 | tok/s 2625317.58 | norm 0.4072 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9372 | train 3.035065 | val 3.367732 | tok/s 2625497.59 | norm 0.4237 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9373 | train 2.975901 | val 3.367732 | tok/s 2457791.53 | norm 0.4017 | dt 213.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9374 | train 3.074726 | val 3.367732 | tok/s 2620947.08 | norm 0.4073 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9375 | train 2.983905 | val 3.367732 | tok/s 2623024.81 | norm 0.4104 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9376 | train 2.927272 | val 3.367732 | tok/s 2628935.16 | norm 0.3918 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9377 | train 3.119622 | val 3.367732 | tok/s 2623109.48 | norm 0.3781 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9378 | train 2.975110 | val 3.367732 | tok/s 2626019.45 | norm 0.3913 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9379 | train 3.098729 | val 3.367732 | tok/s 2628715.86 | norm 0.4370 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9380 | train 3.000891 | val 3.367732 | tok/s 2630653.23 | norm 0.4111 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9381 | train 3.055314 | val 3.367732 | tok/s 2627911.19 | norm 0.4115 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9382 | train 3.072196 | val 3.367732 | tok/s 2627587.42 | norm 0.3930 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9383 | train 3.113879 | val 3.367732 | tok/s 2627245.24 | norm 0.4303 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9384 | train 2.960572 | val 3.367732 | tok/s 2621657.85 | norm 0.4031 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9385 | train 3.003263 | val 3.367732 | tok/s 2621425.25 | norm 0.3513 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9386 | train 3.057092 | val 3.367732 | tok/s 2587845.70 | norm 0.3844 | dt 202.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9387 | train 3.018916 | val 3.367732 | tok/s 2626442.55 | norm 0.4111 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9388 | train 3.112789 | val 3.367732 | tok/s 2629303.62 | norm 0.4060 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9389 | train 3.052622 | val 3.367732 | tok/s 2627137.82 | norm 0.4111 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9390 | train 3.009414 | val 3.367732 | tok/s 2630015.47 | norm 0.4120 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9391 | train 3.007792 | val 3.367732 | tok/s 2629011.39 | norm 0.4016 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9392 | train 2.985230 | val 3.367732 | tok/s 2625974.85 | norm 0.3814 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9393 | train 3.066305 | val 3.367732 | tok/s 2624077.67 | norm 0.3759 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9394 | train 3.114900 | val 3.367732 | tok/s 2628494.14 | norm 0.4222 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9395 | train 3.092060 | val 3.367732 | tok/s 2629415.82 | norm 0.4565 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9396 | train 2.968959 | val 3.367732 | tok/s 2627515.67 | norm 0.4103 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9397 | train 3.074173 | val 3.367732 | tok/s 2629918.52 | norm 0.4254 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9398 | train 3.128677 | val 3.367732 | tok/s 2627700.96 | norm 0.4206 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9399 | train 2.930256 | val 3.367732 | tok/s 2625203.33 | norm 0.3828 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9400 | train 3.051911 | val 3.367732 | tok/s 2627383.40 | norm 0.4010 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9401 | train 2.894293 | val 3.367732 | tok/s 2629689.51 | norm 0.3780 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9402 | train 3.011709 | val 3.367732 | tok/s 2629204.69 | norm 0.3778 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9403 | train 2.937364 | val 3.367732 | tok/s 2630658.38 | norm 0.3432 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9404 | train 3.021065 | val 3.367732 | tok/s 2626377.96 | norm 0.3795 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9405 | train 2.985822 | val 3.367732 | tok/s 2630003.15 | norm 0.3896 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9406 | train 2.903701 | val 3.367732 | tok/s 2630136.48 | norm 0.4253 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9407 | train 2.977774 | val 3.367732 | tok/s 2628861.72 | norm 0.4210 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9408 | train 2.932287 | val 3.367732 | tok/s 2631529.22 | norm 0.3970 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9409 | train 2.978192 | val 3.367732 | tok/s 2628675.24 | norm 0.3980 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9410 | train 3.024749 | val 3.367732 | tok/s 2631400.50 | norm 0.3886 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9411 | train 3.039466 | val 3.367732 | tok/s 2630281.47 | norm 0.3467 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9412 | train 2.916481 | val 3.367732 | tok/s 2629816.54 | norm 0.3878 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9413 | train 2.946963 | val 3.367732 | tok/s 2626676.33 | norm 0.3849 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9414 | train 3.059856 | val 3.367732 | tok/s 2624758.81 | norm 0.3938 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9415 | train 3.048032 | val 3.367732 | tok/s 2624097.00 | norm 0.4290 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9416 | train 3.015298 | val 3.367732 | tok/s 2627934.20 | norm 0.4298 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9417 | train 3.045181 | val 3.367732 | tok/s 2629447.38 | norm 0.4403 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9418 | train 3.072939 | val 3.367732 | tok/s 2631588.14 | norm 0.4302 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9419 | train 3.031423 | val 3.367732 | tok/s 2622000.31 | norm 0.4024 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9420 | train 2.948863 | val 3.367732 | tok/s 2623279.96 | norm 0.3821 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9421 | train 3.014804 | val 3.367732 | tok/s 2625318.24 | norm 0.4061 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9422 | train 3.067470 | val 3.367732 | tok/s 2623548.59 | norm 0.4094 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9423 | train 2.841885 | val 3.367732 | tok/s 2628127.22 | norm 0.3770 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9424 | train 3.004738 | val 3.367732 | tok/s 2630367.37 | norm 0.4002 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9425 | train 3.031297 | val 3.367732 | tok/s 2627562.72 | norm 0.4120 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9426 | train 3.014242 | val 3.367732 | tok/s 2627687.97 | norm 0.4652 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9427 | train 2.819958 | val 3.367732 | tok/s 2629649.34 | norm 0.3534 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9428 | train 3.051878 | val 3.367732 | tok/s 2631193.23 | norm 0.3744 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9429 | train 2.804825 | val 3.367732 | tok/s 2625862.28 | norm 0.4001 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9430 | train 3.185523 | val 3.367732 | tok/s 2627812.08 | norm 0.3868 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9431 | train 3.133748 | val 3.367732 | tok/s 2628511.71 | norm 0.4380 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9432 | train 3.025800 | val 3.367732 | tok/s 2631237.64 | norm 0.4428 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9433 | train 3.111697 | val 3.367732 | tok/s 2629960.68 | norm 0.4406 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9434 | train 3.103902 | val 3.367732 | tok/s 2594579.04 | norm 0.3981 | dt 202.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9435 | train 3.139925 | val 3.367732 | tok/s 2449731.05 | norm 0.4394 | dt 214.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9436 | train 3.421674 | val 3.367732 | tok/s 2620748.07 | norm 0.5388 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9437 | train 2.941037 | val 3.367732 | tok/s 2625337.17 | norm 0.4220 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9438 | train 3.019806 | val 3.367732 | tok/s 2628982.10 | norm 0.4012 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9439 | train 3.007959 | val 3.367732 | tok/s 2630066.99 | norm 0.3883 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9440 | train 3.133280 | val 3.367732 | tok/s 2629234.11 | norm 0.3937 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9441 | train 2.916594 | val 3.367732 | tok/s 2628040.56 | norm 0.3971 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9442 | train 2.934449 | val 3.367732 | tok/s 2630603.69 | norm 0.4148 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9443 | train 2.944851 | val 3.367732 | tok/s 2625342.05 | norm 0.4083 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9444 | train 2.976448 | val 3.367732 | tok/s 2631526.42 | norm 0.3905 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9445 | train 3.006508 | val 3.367732 | tok/s 2627630.29 | norm 0.3803 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9446 | train 2.957624 | val 3.367732 | tok/s 2628511.27 | norm 0.3609 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9447 | train 3.090812 | val 3.367732 | tok/s 2630093.65 | norm 0.3835 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9448 | train 3.035344 | val 3.367732 | tok/s 2625888.66 | norm 0.4075 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9449 | train 3.123634 | val 3.367732 | tok/s 2630848.40 | norm 0.3796 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9450 | train 3.036984 | val 3.367732 | tok/s 2623725.16 | norm 0.3698 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9451 | train 3.041994 | val 3.367732 | tok/s 2627684.26 | norm 0.3836 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9452 | train 2.995772 | val 3.367732 | tok/s 2626945.18 | norm 0.4029 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9453 | train 3.033934 | val 3.367732 | tok/s 2629595.51 | norm 0.3810 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9454 | train 2.934926 | val 3.367732 | tok/s 2631164.98 | norm 0.3503 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9455 | train 2.993392 | val 3.367732 | tok/s 2629874.26 | norm 0.3718 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9456 | train 3.044592 | val 3.367732 | tok/s 2629628.68 | norm 0.3926 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9457 | train 2.900771 | val 3.367732 | tok/s 2629542.59 | norm 0.3948 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9458 | train 2.992749 | val 3.367732 | tok/s 2630565.29 | norm 0.4178 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9459 | train 3.053441 | val 3.367732 | tok/s 2629273.80 | norm 0.3933 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9460 | train 3.033117 | val 3.367732 | tok/s 2630926.01 | norm 0.3708 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9461 | train 2.957614 | val 3.367732 | tok/s 2629488.80 | norm 0.3877 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9462 | train 2.957058 | val 3.367732 | tok/s 2629834.84 | norm 0.3925 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9463 | train 3.089051 | val 3.367732 | tok/s 2629520.82 | norm 0.4449 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9464 | train 2.984570 | val 3.367732 | tok/s 2629403.35 | norm 0.4154 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9465 | train 3.003154 | val 3.367732 | tok/s 2630024.68 | norm 0.3794 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9466 | train 2.998544 | val 3.367732 | tok/s 2623008.06 | norm 0.3687 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9467 | train 3.021132 | val 3.367732 | tok/s 2623999.11 | norm 0.3793 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9468 | train 2.969646 | val 3.367732 | tok/s 2625856.57 | norm 0.3990 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9469 | train 2.998451 | val 3.367732 | tok/s 2629250.53 | norm 0.4398 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9470 | train 3.011992 | val 3.367732 | tok/s 2630590.46 | norm 0.4579 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9471 | train 3.074334 | val 3.367732 | tok/s 2627946.07 | norm 0.4440 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9472 | train 3.114291 | val 3.367732 | tok/s 2628097.58 | norm 0.4065 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9473 | train 3.055100 | val 3.367732 | tok/s 2627750.34 | norm 0.3624 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9474 | train 2.959406 | val 3.367732 | tok/s 2628990.01 | norm 0.3618 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9475 | train 2.938221 | val 3.367732 | tok/s 2621637.85 | norm 0.4132 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9476 | train 2.859666 | val 3.367732 | tok/s 2622699.49 | norm 0.4396 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9477 | train 2.922051 | val 3.367732 | tok/s 2628127.69 | norm 0.4359 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9478 | train 3.046957 | val 3.367732 | tok/s 2627782.07 | norm 0.3897 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9479 | train 3.035659 | val 3.367732 | tok/s 2626190.94 | norm 0.3600 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9480 | train 2.958105 | val 3.367732 | tok/s 2634709.40 | norm 0.3595 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9481 | train 2.975787 | val 3.367732 | tok/s 2623742.52 | norm 0.3755 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9482 | train 3.152237 | val 3.367732 | tok/s 2626191.45 | norm 0.4199 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9483 | train 2.940130 | val 3.367732 | tok/s 2624156.13 | norm 0.3999 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9484 | train 2.938337 | val 3.367732 | tok/s 2624923.14 | norm 0.4013 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9485 | train 3.108453 | val 3.367732 | tok/s 2627548.43 | norm 0.3878 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9486 | train 2.918783 | val 3.367732 | tok/s 2626361.75 | norm 0.3608 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9487 | train 2.969878 | val 3.367732 | tok/s 2627050.22 | norm 0.3921 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9488 | train 3.028282 | val 3.367732 | tok/s 2627610.11 | norm 0.4247 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9489 | train 3.158546 | val 3.367732 | tok/s 2629567.30 | norm 0.4488 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9490 | train 2.971663 | val 3.367732 | tok/s 2629684.44 | norm 0.4738 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9491 | train 3.142617 | val 3.367732 | tok/s 2625941.13 | norm 0.4381 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9492 | train 2.956353 | val 3.367732 | tok/s 2630230.15 | norm 0.4492 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9493 | train 3.060944 | val 3.367732 | tok/s 2509356.48 | norm 0.3951 | dt 208.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9494 | train 3.063110 | val 3.367732 | tok/s 2618133.49 | norm 0.3801 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9495 | train 2.963258 | val 3.367732 | tok/s 2624539.82 | norm 0.4130 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9496 | train 2.907989 | val 3.367732 | tok/s 2622574.96 | norm 0.4052 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9497 | train 2.950841 | val 3.367732 | tok/s 2628179.52 | norm 0.4381 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9498 | train 3.017397 | val 3.367732 | tok/s 2618850.19 | norm 0.4123 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9499 | train 2.946914 | val 3.367732 | tok/s 2599858.51 | norm 0.3984 | dt 201.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 9500 | val_loss 3.370222 step 9500 | train 3.088250 | val 3.370222 | tok/s 2618037.43 | norm 0.4408 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9501 | train 3.045686 | val 3.370222 | tok/s 2478413.93 | norm 0.4341 | dt 211.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9502 | train 2.949310 | val 3.370222 | tok/s 2621531.50 | norm 0.4228 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9503 | train 3.050304 | val 3.370222 | tok/s 2528279.52 | norm 0.4411 | dt 207.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9504 | train 3.046699 | val 3.370222 | tok/s 2627882.53 | norm 0.4207 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9505 | train 3.002180 | val 3.370222 | tok/s 2627402.02 | norm 0.4117 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9506 | train 2.946251 | val 3.370222 | tok/s 2627044.63 | norm 0.3866 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9507 | train 2.990621 | val 3.370222 | tok/s 2627676.63 | norm 0.3836 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9508 | train 3.079893 | val 3.370222 | tok/s 2626422.38 | norm 1.1015 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9509 | train 3.102427 | val 3.370222 | tok/s 2627673.12 | norm 0.3317 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9510 | train 3.009531 | val 3.370222 | tok/s 2608836.15 | norm 0.3443 | dt 200.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9511 | train 3.014931 | val 3.370222 | tok/s 2624783.00 | norm 0.3547 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9512 | train 3.087677 | val 3.370222 | tok/s 2625543.96 | norm 0.3801 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9513 | train 3.115358 | val 3.370222 | tok/s 2627675.72 | norm 0.3715 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9514 | train 2.973284 | val 3.370222 | tok/s 2626348.96 | norm 0.3655 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9515 | train 3.003153 | val 3.370222 | tok/s 2626863.74 | norm 0.4207 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9516 | train 2.947552 | val 3.370222 | tok/s 2625397.25 | norm 0.4016 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9517 | train 2.917864 | val 3.370222 | tok/s 2628469.12 | norm 0.4015 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9518 | train 3.034045 | val 3.370222 | tok/s 2628960.33 | norm 0.4197 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9519 | train 3.023120 | val 3.370222 | tok/s 2623663.44 | norm 0.3669 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9520 | train 2.955233 | val 3.370222 | tok/s 2626008.61 | norm 0.3662 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9521 | train 2.924529 | val 3.370222 | tok/s 2624718.39 | norm 0.3995 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9522 | train 2.932550 | val 3.370222 | tok/s 2624928.69 | norm 0.4026 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9523 | train 2.928069 | val 3.370222 | tok/s 2631315.27 | norm 1.0717 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9524 | train 2.858790 | val 3.370222 | tok/s 2629635.02 | norm 1.6169 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9525 | train 3.050575 | val 3.370222 | tok/s 2627949.42 | norm 0.4040 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9526 | train 3.027828 | val 3.370222 | tok/s 2627138.35 | norm 0.3867 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9527 | train 3.038247 | val 3.370222 | tok/s 2624781.99 | norm 0.3542 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9528 | train 3.026067 | val 3.370222 | tok/s 2627736.56 | norm 0.4676 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9529 | train 3.043838 | val 3.370222 | tok/s 2623819.33 | norm 0.3960 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9530 | train 3.011307 | val 3.370222 | tok/s 2628417.19 | norm 0.3613 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9531 | train 2.969438 | val 3.370222 | tok/s 2628860.91 | norm 0.4167 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9532 | train 2.999216 | val 3.370222 | tok/s 2629004.66 | norm 0.3977 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9533 | train 3.156929 | val 3.370222 | tok/s 2628355.72 | norm 0.4924 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9534 | train 3.018475 | val 3.370222 | tok/s 2627222.03 | norm 0.4356 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9535 | train 2.965976 | val 3.370222 | tok/s 2629936.80 | norm 0.3559 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9536 | train 3.173359 | val 3.370222 | tok/s 2628708.12 | norm 0.3949 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9537 | train 2.908536 | val 3.370222 | tok/s 2628218.03 | norm 0.3877 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9538 | train 2.897532 | val 3.370222 | tok/s 2627080.43 | norm 0.4105 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9539 | train 2.933821 | val 3.370222 | tok/s 2632244.20 | norm 0.4187 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9540 | train 2.851636 | val 3.370222 | tok/s 2631058.74 | norm 0.3941 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9541 | train 3.050659 | val 3.370222 | tok/s 2627011.98 | norm 0.3877 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9542 | train 3.055482 | val 3.370222 | tok/s 2627454.08 | norm 0.4019 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9543 | train 3.063278 | val 3.370222 | tok/s 2632063.36 | norm 0.3922 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9544 | train 3.052180 | val 3.370222 | tok/s 2627324.08 | norm 0.4258 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9545 | train 3.250341 | val 3.370222 | tok/s 2627253.64 | norm 0.4601 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9546 | train 3.238042 | val 3.370222 | tok/s 2628869.57 | norm 0.4426 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9547 | train 3.047978 | val 3.370222 | tok/s 2631672.60 | norm 0.3674 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9548 | train 3.123954 | val 3.370222 | tok/s 2630591.23 | norm 0.3561 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9549 | train 3.002213 | val 3.370222 | tok/s 2631287.51 | norm 0.3651 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9550 | train 3.047957 | val 3.370222 | tok/s 2632203.96 | norm 0.4171 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9551 | train 3.036568 | val 3.370222 | tok/s 2629423.34 | norm 0.4059 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9552 | train 3.078477 | val 3.370222 | tok/s 2631173.13 | norm 0.3607 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9553 | train 3.065179 | val 3.370222 | tok/s 2626101.07 | norm 0.3905 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9554 | train 3.192159 | val 3.370222 | tok/s 2626575.63 | norm 0.4402 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9555 | train 2.913883 | val 3.370222 | tok/s 2627872.94 | norm 0.3824 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9556 | train 3.036364 | val 3.370222 | tok/s 2626776.95 | norm 0.3780 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9557 | train 2.863715 | val 3.370222 | tok/s 2486776.67 | norm 0.4049 | dt 210.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9558 | train 3.052703 | val 3.370222 | tok/s 2623280.13 | norm 0.3698 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9559 | train 2.942990 | val 3.370222 | tok/s 2626001.14 | norm 0.3730 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9560 | train 2.926832 | val 3.370222 | tok/s 2626404.59 | norm 0.4156 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9561 | train 2.982047 | val 3.370222 | tok/s 2628764.09 | norm 0.3990 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9562 | train 3.059671 | val 3.370222 | tok/s 2625617.50 | norm 0.4428 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9563 | train 3.075044 | val 3.370222 | tok/s 2630050.11 | norm 0.4649 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9564 | train 2.900001 | val 3.370222 | tok/s 2624413.29 | norm 0.4249 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9565 | train 3.114334 | val 3.370222 | tok/s 2631739.13 | norm 0.4258 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9566 | train 2.924586 | val 3.370222 | tok/s 2630743.81 | norm 0.3502 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9567 | train 3.024652 | val 3.370222 | tok/s 2626846.40 | norm 0.3801 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9568 | train 2.934293 | val 3.370222 | tok/s 2626526.64 | norm 0.3886 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9569 | train 2.932344 | val 3.370222 | tok/s 2627579.97 | norm 0.3848 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9570 | train 2.954673 | val 3.370222 | tok/s 2623682.55 | norm 0.4068 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9571 | train 2.943309 | val 3.370222 | tok/s 2626781.67 | norm 0.3912 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9572 | train 3.044987 | val 3.370222 | tok/s 2626671.28 | norm 0.3931 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9573 | train 3.108739 | val 3.370222 | tok/s 2621330.06 | norm 0.4368 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9574 | train 3.098532 | val 3.370222 | tok/s 2624788.04 | norm 0.4413 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9575 | train 2.992366 | val 3.370222 | tok/s 2623073.22 | norm 0.4079 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9576 | train 2.979848 | val 3.370222 | tok/s 2626734.74 | norm 0.4250 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9577 | train 3.019996 | val 3.370222 | tok/s 2625828.53 | norm 0.4127 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9578 | train 3.016395 | val 3.370222 | tok/s 2628901.40 | norm 0.3836 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9579 | train 3.087953 | val 3.370222 | tok/s 2624635.72 | norm 0.4179 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9580 | train 3.032798 | val 3.370222 | tok/s 2622381.78 | norm 0.4032 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9581 | train 3.079726 | val 3.370222 | tok/s 2629657.51 | norm 0.4234 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9582 | train 3.057489 | val 3.370222 | tok/s 2626672.05 | norm 0.4360 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9583 | train 2.968757 | val 3.370222 | tok/s 2630737.25 | norm 0.3945 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9584 | train 2.988411 | val 3.370222 | tok/s 2628529.18 | norm 0.3959 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9585 | train 2.960347 | val 3.370222 | tok/s 2627277.10 | norm 0.4468 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9586 | train 2.995162 | val 3.370222 | tok/s 2627034.07 | norm 0.4144 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9587 | train 3.020279 | val 3.370222 | tok/s 2629335.57 | norm 0.3959 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9588 | train 2.969875 | val 3.370222 | tok/s 2631081.74 | norm 0.4438 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9589 | train 2.976201 | val 3.370222 | tok/s 2627518.65 | norm 0.4314 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9590 | train 3.010380 | val 3.370222 | tok/s 2627115.81 | norm 0.3868 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9591 | train 2.958189 | val 3.370222 | tok/s 2619432.36 | norm 0.3527 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9592 | train 3.012933 | val 3.370222 | tok/s 2626261.33 | norm 0.3700 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9593 | train 3.109935 | val 3.370222 | tok/s 2629028.32 | norm 0.4162 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9594 | train 2.989522 | val 3.370222 | tok/s 2627490.88 | norm 0.3813 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9595 | train 3.064870 | val 3.370222 | tok/s 2625775.18 | norm 0.3872 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9596 | train 2.892706 | val 3.370222 | tok/s 2624037.31 | norm 0.4262 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9597 | train 2.963821 | val 3.370222 | tok/s 2627577.40 | norm 0.4316 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9598 | train 2.959432 | val 3.370222 | tok/s 2631952.35 | norm 0.4405 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9599 | train 3.039648 | val 3.370222 | tok/s 2621289.03 | norm 0.4508 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9600 | train 3.005617 | val 3.370222 | tok/s 2630571.93 | norm 0.4075 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9601 | train 3.025266 | val 3.370222 | tok/s 2628888.17 | norm 0.3793 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9602 | train 2.957888 | val 3.370222 | tok/s 2627876.72 | norm 0.3884 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9603 | train 3.105673 | val 3.370222 | tok/s 2630242.18 | norm 0.4219 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9604 | train 3.036086 | val 3.370222 | tok/s 2628854.50 | norm 0.4227 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9605 | train 3.001859 | val 3.370222 | tok/s 2593569.94 | norm 0.4114 | dt 202.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9606 | train 2.871663 | val 3.370222 | tok/s 2539501.12 | norm 0.3795 | dt 206.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9607 | train 2.799873 | val 3.370222 | tok/s 2629551.00 | norm 0.3749 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9608 | train 3.117257 | val 3.370222 | tok/s 2631006.67 | norm 0.4523 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9609 | train 3.234565 | val 3.370222 | tok/s 2630683.17 | norm 0.4313 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9610 | train 2.954165 | val 3.370222 | tok/s 2628220.53 | norm 0.3955 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9611 | train 3.114958 | val 3.370222 | tok/s 2626037.40 | norm 0.3982 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9612 | train 2.985124 | val 3.370222 | tok/s 2627420.40 | norm 0.4133 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9613 | train 3.092033 | val 3.370222 | tok/s 2472577.80 | norm 0.4358 | dt 212.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9614 | train 2.877017 | val 3.370222 | tok/s 2626093.06 | norm 0.4057 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9615 | train 3.012793 | val 3.370222 | tok/s 2631335.29 | norm 0.3909 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9616 | train 2.972669 | val 3.370222 | tok/s 2627649.92 | norm 0.3632 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9617 | train 2.949586 | val 3.370222 | tok/s 2628691.48 | norm 0.3824 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9618 | train 2.857856 | val 3.370222 | tok/s 2627440.26 | norm 0.4399 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9619 | train 2.956706 | val 3.370222 | tok/s 2628832.92 | norm 0.4320 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9620 | train 3.164549 | val 3.370222 | tok/s 2628085.06 | norm 0.4415 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9621 | train 2.995866 | val 3.370222 | tok/s 2580424.45 | norm 0.3943 | dt 203.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9622 | train 3.186511 | val 3.370222 | tok/s 2583303.59 | norm 0.4055 | dt 202.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9623 | train 2.878543 | val 3.370222 | tok/s 2627757.67 | norm 0.4061 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9624 | train 3.023468 | val 3.370222 | tok/s 2630810.18 | norm 0.4416 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9625 | train 3.047711 | val 3.370222 | tok/s 2630700.02 | norm 0.4034 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9626 | train 2.939842 | val 3.370222 | tok/s 2627861.36 | norm 0.3856 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9627 | train 3.002714 | val 3.370222 | tok/s 2625310.22 | norm 0.3834 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9628 | train 2.944977 | val 3.370222 | tok/s 2627494.40 | norm 0.4037 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9629 | train 3.077679 | val 3.370222 | tok/s 2625994.91 | norm 0.4148 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9630 | train 3.127185 | val 3.370222 | tok/s 2623325.08 | norm 0.4790 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9631 | train 2.992694 | val 3.370222 | tok/s 2627612.18 | norm 0.4295 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9632 | train 3.000284 | val 3.370222 | tok/s 2628798.10 | norm 0.3998 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9633 | train 3.096022 | val 3.370222 | tok/s 2627020.47 | norm 0.4000 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9634 | train 2.945098 | val 3.370222 | tok/s 2628690.55 | norm 0.3620 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9635 | train 3.087692 | val 3.370222 | tok/s 2625669.98 | norm 0.3923 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9636 | train 3.011549 | val 3.370222 | tok/s 2625316.75 | norm 0.3863 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9637 | train 3.013970 | val 3.370222 | tok/s 2626374.26 | norm 0.3508 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9638 | train 3.065443 | val 3.370222 | tok/s 2627488.81 | norm 0.4070 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9639 | train 3.030589 | val 3.370222 | tok/s 2627505.91 | norm 0.4530 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9640 | train 2.977691 | val 3.370222 | tok/s 2625238.13 | norm 0.4367 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9641 | train 3.137808 | val 3.370222 | tok/s 2626093.96 | norm 0.4290 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9642 | train 2.903893 | val 3.370222 | tok/s 2633206.34 | norm 0.4258 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9643 | train 2.897770 | val 3.370222 | tok/s 2621568.80 | norm 0.4499 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9644 | train 2.971205 | val 3.370222 | tok/s 2627380.19 | norm 0.4070 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9645 | train 2.992147 | val 3.370222 | tok/s 2625881.82 | norm 0.3882 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9646 | train 2.998867 | val 3.370222 | tok/s 2623343.84 | norm 0.3802 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9647 | train 2.954541 | val 3.370222 | tok/s 2615679.74 | norm 0.3769 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9648 | train 3.010788 | val 3.370222 | tok/s 2625682.03 | norm 0.3961 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9649 | train 3.015539 | val 3.370222 | tok/s 2624632.68 | norm 0.3919 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9650 | train 3.130663 | val 3.370222 | tok/s 2624902.72 | norm 0.3738 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9651 | train 3.045415 | val 3.370222 | tok/s 2627762.59 | norm 0.3899 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9652 | train 3.010962 | val 3.370222 | tok/s 2627405.09 | norm 0.4036 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9653 | train 2.928073 | val 3.370222 | tok/s 2629856.40 | norm 0.4099 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9654 | train 3.011751 | val 3.370222 | tok/s 2626512.23 | norm 0.4130 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9655 | train 2.852346 | val 3.370222 | tok/s 2625452.35 | norm 0.4317 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9656 | train 3.047736 | val 3.370222 | tok/s 2625623.25 | norm 0.4100 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9657 | train 2.989827 | val 3.370222 | tok/s 2626655.29 | norm 0.4042 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9658 | train 2.928601 | val 3.370222 | tok/s 2627218.07 | norm 0.4158 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9659 | train 3.020555 | val 3.370222 | tok/s 2628164.96 | norm 0.3995 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9660 | train 2.846411 | val 3.370222 | tok/s 2626585.63 | norm 0.4014 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9661 | train 3.073661 | val 3.370222 | tok/s 2626173.32 | norm 0.3970 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9662 | train 2.987414 | val 3.370222 | tok/s 2628383.24 | norm 0.4218 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9663 | train 2.869653 | val 3.370222 | tok/s 2623217.20 | norm 0.4507 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9664 | train 3.041471 | val 3.370222 | tok/s 2623873.18 | norm 0.4534 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9665 | train 3.094429 | val 3.370222 | tok/s 2624766.21 | norm 0.4199 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9666 | train 2.992484 | val 3.370222 | tok/s 2628353.13 | norm 0.3890 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9667 | train 2.898279 | val 3.370222 | tok/s 2628356.73 | norm 0.3907 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9668 | train 2.965806 | val 3.370222 | tok/s 2625617.15 | norm 0.4166 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9669 | train 3.118707 | val 3.370222 | tok/s 2631416.81 | norm 0.4432 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9670 | train 3.140304 | val 3.370222 | tok/s 2460608.27 | norm 0.3939 | dt 213.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9671 | train 2.960967 | val 3.370222 | tok/s 2625429.54 | norm 0.3645 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9672 | train 3.045428 | val 3.370222 | tok/s 2626490.49 | norm 0.3564 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9673 | train 2.962919 | val 3.370222 | tok/s 2630922.95 | norm 0.3794 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9674 | train 3.050925 | val 3.370222 | tok/s 2628736.74 | norm 0.3635 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9675 | train 2.932449 | val 3.370222 | tok/s 2628696.05 | norm 0.3762 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9676 | train 3.011478 | val 3.370222 | tok/s 2631624.66 | norm 0.4027 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9677 | train 2.982387 | val 3.370222 | tok/s 2629113.51 | norm 0.4246 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9678 | train 2.982478 | val 3.370222 | tok/s 2627526.10 | norm 0.3979 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9679 | train 3.015150 | val 3.370222 | tok/s 2629770.92 | norm 0.3605 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9680 | train 2.993147 | val 3.370222 | tok/s 2629845.06 | norm 0.3916 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9681 | train 2.906314 | val 3.370222 | tok/s 2630682.46 | norm 0.4168 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9682 | train 2.821005 | val 3.370222 | tok/s 2630259.40 | norm 0.4224 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9683 | train 3.086602 | val 3.370222 | tok/s 2630404.17 | norm 0.4271 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9684 | train 3.025547 | val 3.370222 | tok/s 2628209.23 | norm 0.4443 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9685 | train 3.041528 | val 3.370222 | tok/s 2628584.52 | norm 0.4425 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9686 | train 2.950344 | val 3.370222 | tok/s 2629007.23 | norm 0.4185 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9687 | train 3.168295 | val 3.370222 | tok/s 2628627.47 | norm 0.4144 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9688 | train 2.935309 | val 3.370222 | tok/s 2627936.07 | norm 0.4408 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9689 | train 2.910990 | val 3.370222 | tok/s 2626709.65 | norm 0.4102 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9690 | train 3.120644 | val 3.370222 | tok/s 2629217.33 | norm 0.3837 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9691 | train 2.929524 | val 3.370222 | tok/s 2619826.12 | norm 0.3797 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9692 | train 2.998714 | val 3.370222 | tok/s 2498588.12 | norm 0.4176 | dt 209.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9693 | train 3.042568 | val 3.370222 | tok/s 2629088.44 | norm 0.4226 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9694 | train 2.932119 | val 3.370222 | tok/s 2633757.36 | norm 0.3740 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9695 | train 3.149671 | val 3.370222 | tok/s 2629212.33 | norm 0.3742 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9696 | train 2.922596 | val 3.370222 | tok/s 2633113.54 | norm 0.3488 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9697 | train 3.029775 | val 3.370222 | tok/s 2632807.71 | norm 0.3883 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9698 | train 3.053960 | val 3.370222 | tok/s 2623719.21 | norm 0.4134 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9699 | train 3.188491 | val 3.370222 | tok/s 2625300.50 | norm 0.4558 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9700 | train 2.985524 | val 3.370222 | tok/s 2631220.52 | norm 0.4508 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9701 | train 2.980078 | val 3.370222 | tok/s 2630392.00 | norm 0.4519 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9702 | train 3.014984 | val 3.370222 | tok/s 2631875.25 | norm 0.4580 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9703 | train 2.942935 | val 3.370222 | tok/s 2632747.52 | norm 0.4329 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9704 | train 2.935516 | val 3.370222 | tok/s 2626876.28 | norm 0.4307 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9705 | train 2.857696 | val 3.370222 | tok/s 2630945.38 | norm 0.3743 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9706 | train 3.044176 | val 3.370222 | tok/s 2630671.47 | norm 0.4356 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9707 | train 3.038355 | val 3.370222 | tok/s 2630336.88 | norm 0.3761 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9708 | train 2.965466 | val 3.370222 | tok/s 2576132.45 | norm 0.3591 | dt 203.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9709 | train 3.064800 | val 3.370222 | tok/s 2628449.79 | norm 0.3945 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9710 | train 2.944451 | val 3.370222 | tok/s 2630394.01 | norm 0.3933 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9711 | train 2.915604 | val 3.370222 | tok/s 2631089.42 | norm 0.3759 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9712 | train 3.020118 | val 3.370222 | tok/s 2630844.39 | norm 0.3614 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9713 | train 3.046486 | val 3.370222 | tok/s 2630514.39 | norm 0.4036 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9714 | train 3.054268 | val 3.370222 | tok/s 2515774.91 | norm 0.4198 | dt 208.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9715 | train 3.137780 | val 3.370222 | tok/s 2616023.82 | norm 0.4414 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9716 | train 2.912212 | val 3.370222 | tok/s 2630144.27 | norm 0.3955 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9717 | train 3.099515 | val 3.370222 | tok/s 2633179.83 | norm 0.3999 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9718 | train 3.069128 | val 3.370222 | tok/s 2628918.44 | norm 0.4554 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9719 | train 3.064353 | val 3.370222 | tok/s 2629146.90 | norm 0.4129 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9720 | train 2.899670 | val 3.370222 | tok/s 2628312.17 | norm 0.4076 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9721 | train 3.067310 | val 3.370222 | tok/s 2510078.15 | norm 0.3964 | dt 208.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9722 | train 2.896878 | val 3.370222 | tok/s 2628396.84 | norm 0.4403 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9723 | train 3.015222 | val 3.370222 | tok/s 2628583.20 | norm 0.4455 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9724 | train 3.009206 | val 3.370222 | tok/s 2629546.22 | norm 0.3932 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9725 | train 2.848903 | val 3.370222 | tok/s 2624981.57 | norm 0.4104 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9726 | train 3.037599 | val 3.370222 | tok/s 2624537.85 | norm 0.4535 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9727 | train 2.942146 | val 3.370222 | tok/s 2629427.78 | norm 0.4077 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9728 | train 2.894842 | val 3.370222 | tok/s 2628570.84 | norm 0.4263 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9729 | train 3.025310 | val 3.370222 | tok/s 2626002.96 | norm 0.4003 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9730 | train 3.062732 | val 3.370222 | tok/s 2629172.79 | norm 0.3853 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9731 | train 2.934562 | val 3.370222 | tok/s 2626453.33 | norm 0.4052 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9732 | train 2.846509 | val 3.370222 | tok/s 2627034.31 | norm 0.3962 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9733 | train 2.959269 | val 3.370222 | tok/s 2628086.46 | norm 0.4115 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9734 | train 3.110606 | val 3.370222 | tok/s 2628302.48 | norm 0.4257 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9735 | train 3.103062 | val 3.370222 | tok/s 2626248.11 | norm 0.4103 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9736 | train 2.971401 | val 3.370222 | tok/s 2631105.74 | norm 0.3850 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9737 | train 3.039686 | val 3.370222 | tok/s 2629904.20 | norm 0.3791 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9738 | train 3.048165 | val 3.370222 | tok/s 2627359.07 | norm 0.3936 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9739 | train 3.080844 | val 3.370222 | tok/s 2629261.78 | norm 0.4301 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9740 | train 2.972614 | val 3.370222 | tok/s 2631766.82 | norm 0.3834 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9741 | train 3.026123 | val 3.370222 | tok/s 2629892.57 | norm 0.3730 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9742 | train 2.987370 | val 3.370222 | tok/s 2628141.37 | norm 0.4216 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9743 | train 2.865942 | val 3.370222 | tok/s 2627400.13 | norm 0.4380 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9744 | train 3.047482 | val 3.370222 | tok/s 2631854.00 | norm 0.4334 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9745 | train 2.944636 | val 3.370222 | tok/s 2622945.12 | norm 0.4259 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9746 | train 3.026536 | val 3.370222 | tok/s 2628641.87 | norm 0.4101 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9747 | train 3.131505 | val 3.370222 | tok/s 2627757.00 | norm 0.4058 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9748 | train 2.970519 | val 3.370222 | tok/s 2634449.99 | norm 0.3813 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9749 | train 2.947774 | val 3.370222 | tok/s 2628747.15 | norm 0.3961 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 9750 | val_loss 3.367506 step 9750 | train 3.015173 | val 3.367506 | tok/s 2614563.97 | norm 0.4157 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9751 | train 3.025334 | val 3.367506 | tok/s 2627858.78 | norm 0.4371 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9752 | train 3.006359 | val 3.367506 | tok/s 2631993.91 | norm 0.4314 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9753 | train 3.026072 | val 3.367506 | tok/s 2622211.89 | norm 0.4335 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9754 | train 3.065868 | val 3.367506 | tok/s 2625958.28 | norm 0.4286 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9755 | train 2.906707 | val 3.367506 | tok/s 2629114.79 | norm 0.4243 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9756 | train 3.033286 | val 3.367506 | tok/s 2629624.93 | norm 0.4296 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9757 | train 2.964982 | val 3.367506 | tok/s 2629849.96 | norm 0.4314 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9758 | train 2.954834 | val 3.367506 | tok/s 2627466.59 | norm 0.3887 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9759 | train 2.957594 | val 3.367506 | tok/s 2627448.82 | norm 0.3776 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9760 | train 3.008650 | val 3.367506 | tok/s 2632272.00 | norm 0.3836 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9761 | train 3.009895 | val 3.367506 | tok/s 2628539.29 | norm 0.3787 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9762 | train 3.017290 | val 3.367506 | tok/s 2628806.80 | norm 0.4045 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9763 | train 3.097339 | val 3.367506 | tok/s 2623177.40 | norm 0.4315 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9764 | train 2.920426 | val 3.367506 | tok/s 2628026.16 | norm 0.4480 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9765 | train 2.954967 | val 3.367506 | tok/s 2627751.64 | norm 0.4061 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9766 | train 2.898901 | val 3.367506 | tok/s 2628211.35 | norm 0.4173 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9767 | train 2.961712 | val 3.367506 | tok/s 2627521.12 | norm 0.4375 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9768 | train 3.110550 | val 3.367506 | tok/s 2630021.36 | norm 0.3728 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9769 | train 3.032075 | val 3.367506 | tok/s 2628880.19 | norm 0.3672 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9770 | train 2.957319 | val 3.367506 | tok/s 2622588.61 | norm 0.3977 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9771 | train 3.073681 | val 3.367506 | tok/s 2628264.68 | norm 0.3860 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9772 | train 2.940947 | val 3.367506 | tok/s 2629258.77 | norm 0.4204 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9773 | train 3.220240 | val 3.367506 | tok/s 2629092.57 | norm 0.5261 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9774 | train 3.046525 | val 3.367506 | tok/s 2629486.29 | norm 0.3539 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9775 | train 2.947471 | val 3.367506 | tok/s 2627930.59 | norm 0.3555 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9776 | train 3.248880 | val 3.367506 | tok/s 2626476.47 | norm 0.5419 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9777 | train 2.850535 | val 3.367506 | tok/s 2629053.34 | norm 0.3902 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9778 | train 3.117408 | val 3.367506 | tok/s 2629237.24 | norm 0.3769 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9779 | train 2.916870 | val 3.367506 | tok/s 2629333.24 | norm 0.3885 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9780 | train 3.029318 | val 3.367506 | tok/s 2630770.77 | norm 0.4172 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9781 | train 2.942306 | val 3.367506 | tok/s 2624659.46 | norm 0.3400 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9782 | train 3.041041 | val 3.367506 | tok/s 2629330.45 | norm 0.2972 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9783 | train 3.004716 | val 3.367506 | tok/s 2624442.82 | norm 0.3902 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9784 | train 3.121604 | val 3.367506 | tok/s 2628305.67 | norm 0.5960 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9785 | train 3.029304 | val 3.367506 | tok/s 2628151.56 | norm 0.3830 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9786 | train 2.893021 | val 3.367506 | tok/s 2630417.83 | norm 0.3564 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9787 | train 2.967481 | val 3.367506 | tok/s 2628950.91 | norm 0.3824 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9788 | train 2.930694 | val 3.367506 | tok/s 2627713.02 | norm 0.3935 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9789 | train 2.980950 | val 3.367506 | tok/s 2627242.79 | norm 0.4164 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9790 | train 3.103464 | val 3.367506 | tok/s 2626932.31 | norm 0.4517 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9791 | train 3.011671 | val 3.367506 | tok/s 2631348.16 | norm 0.3996 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9792 | train 3.100576 | val 3.367506 | tok/s 2628282.51 | norm 0.3575 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9793 | train 2.998122 | val 3.367506 | tok/s 2629511.18 | norm 0.3804 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9794 | train 2.965086 | val 3.367506 | tok/s 2632644.47 | norm 0.4262 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9795 | train 2.951498 | val 3.367506 | tok/s 2628980.68 | norm 0.3919 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9796 | train 2.933599 | val 3.367506 | tok/s 2629007.62 | norm 0.3845 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9797 | train 3.067097 | val 3.367506 | tok/s 2595137.85 | norm 0.3849 | dt 202.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9798 | train 2.993090 | val 3.367506 | tok/s 2628963.40 | norm 1.5614 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9799 | train 3.054162 | val 3.367506 | tok/s 2627699.76 | norm 0.2896 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9800 | train 2.992148 | val 3.367506 | tok/s 2633971.88 | norm 0.3481 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9801 | train 3.146420 | val 3.367506 | tok/s 2629550.57 | norm 0.4150 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9802 | train 3.118305 | val 3.367506 | tok/s 2628818.61 | norm 0.3945 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9803 | train 3.055462 | val 3.367506 | tok/s 2620181.07 | norm 0.3836 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9804 | train 3.019141 | val 3.367506 | tok/s 2627881.11 | norm 0.4100 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9805 | train 3.041734 | val 3.367506 | tok/s 2629562.98 | norm 0.5564 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9806 | train 2.962090 | val 3.367506 | tok/s 2628851.24 | norm 0.4298 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9807 | train 3.086780 | val 3.367506 | tok/s 2628705.62 | norm 0.4299 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9808 | train 3.077486 | val 3.367506 | tok/s 2631815.35 | norm 0.4434 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9809 | train 2.959763 | val 3.367506 | tok/s 2625647.84 | norm 0.3705 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9810 | train 3.107128 | val 3.367506 | tok/s 2631937.72 | norm 0.3944 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9811 | train 2.910171 | val 3.367506 | tok/s 2629310.95 | norm 0.3958 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9812 | train 2.890226 | val 3.367506 | tok/s 2624631.76 | norm 0.3802 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9813 | train 3.077402 | val 3.367506 | tok/s 2629696.50 | norm 0.4045 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9814 | train 3.009142 | val 3.367506 | tok/s 2630173.87 | norm 0.3880 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9815 | train 3.084778 | val 3.367506 | tok/s 2630310.67 | norm 0.3743 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9816 | train 2.965737 | val 3.367506 | tok/s 2631103.21 | norm 0.3465 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9817 | train 3.071732 | val 3.367506 | tok/s 2625451.41 | norm 0.3839 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9818 | train 3.084789 | val 3.367506 | tok/s 2626347.59 | norm 0.4205 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9819 | train 3.062719 | val 3.367506 | tok/s 2628191.90 | norm 0.4453 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9820 | train 3.059430 | val 3.367506 | tok/s 2631470.52 | norm 0.4329 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9821 | train 3.092986 | val 3.367506 | tok/s 2628989.00 | norm 0.4421 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9822 | train 3.082296 | val 3.367506 | tok/s 2628061.97 | norm 0.4267 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9823 | train 3.017310 | val 3.367506 | tok/s 2627369.35 | norm 0.4029 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9824 | train 2.992655 | val 3.367506 | tok/s 2628035.53 | norm 0.4126 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9825 | train 2.924108 | val 3.367506 | tok/s 2626291.55 | norm 0.4238 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9826 | train 2.987377 | val 3.367506 | tok/s 2633923.49 | norm 0.4251 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9827 | train 3.291697 | val 3.367506 | tok/s 2629664.71 | norm 0.4355 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9828 | train 2.929249 | val 3.367506 | tok/s 2629984.40 | norm 0.3944 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9829 | train 2.984555 | val 3.367506 | tok/s 2629868.66 | norm 0.3624 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9830 | train 3.004220 | val 3.367506 | tok/s 2631151.79 | norm 0.3594 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9831 | train 2.965442 | val 3.367506 | tok/s 2626333.10 | norm 0.3805 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9832 | train 3.040764 | val 3.367506 | tok/s 2626092.23 | norm 0.3723 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9833 | train 3.058643 | val 3.367506 | tok/s 2627204.49 | norm 0.3963 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9834 | train 3.053107 | val 3.367506 | tok/s 2628283.08 | norm 0.4263 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9835 | train 2.822023 | val 3.367506 | tok/s 2629719.37 | norm 0.3921 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9836 | train 3.204804 | val 3.367506 | tok/s 2631603.49 | norm 0.4167 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9837 | train 2.852570 | val 3.367506 | tok/s 2631152.41 | norm 0.4024 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9838 | train 3.046329 | val 3.367506 | tok/s 2632183.73 | norm 0.4011 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9839 | train 3.052088 | val 3.367506 | tok/s 2631741.17 | norm 0.4375 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9840 | train 3.078377 | val 3.367506 | tok/s 2630540.43 | norm 0.4371 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9841 | train 3.012967 | val 3.367506 | tok/s 2624138.81 | norm 0.4120 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9842 | train 2.899320 | val 3.367506 | tok/s 2626642.62 | norm 0.4208 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9843 | train 2.983078 | val 3.367506 | tok/s 2625187.11 | norm 0.4011 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9844 | train 3.022503 | val 3.367506 | tok/s 2627026.53 | norm 0.3824 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9845 | train 3.014394 | val 3.367506 | tok/s 2629905.60 | norm 0.3911 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9846 | train 3.019329 | val 3.367506 | tok/s 2630360.26 | norm 0.3863 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9847 | train 3.101967 | val 3.367506 | tok/s 2627376.20 | norm 0.3903 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9848 | train 2.942511 | val 3.367506 | tok/s 2628121.26 | norm 0.3871 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9849 | train 2.963087 | val 3.367506 | tok/s 2628703.96 | norm 0.3593 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9850 | train 3.291274 | val 3.367506 | tok/s 2628870.04 | norm 0.4650 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9851 | train 3.065364 | val 3.367506 | tok/s 2627508.72 | norm 0.4572 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9852 | train 3.171418 | val 3.367506 | tok/s 2621755.20 | norm 0.4225 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9853 | train 2.934568 | val 3.367506 | tok/s 2616990.51 | norm 0.4297 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9854 | train 2.999062 | val 3.367506 | tok/s 2622842.37 | norm 0.4130 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9855 | train 3.025846 | val 3.367506 | tok/s 2624559.60 | norm 0.3757 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9856 | train 3.198975 | val 3.367506 | tok/s 2625442.47 | norm 0.4365 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9857 | train 2.937006 | val 3.367506 | tok/s 2628881.76 | norm 0.4474 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9858 | train 3.015535 | val 3.367506 | tok/s 2629534.32 | norm 0.3617 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9859 | train 3.054434 | val 3.367506 | tok/s 2628679.51 | norm 0.3859 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9860 | train 3.107515 | val 3.367506 | tok/s 2629824.41 | norm 0.4360 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9861 | train 3.178190 | val 3.367506 | tok/s 2627298.24 | norm 0.4750 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9862 | train 3.189162 | val 3.367506 | tok/s 2628779.22 | norm 0.4894 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9863 | train 2.961743 | val 3.367506 | tok/s 2630722.85 | norm 0.4643 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9864 | train 3.039522 | val 3.367506 | tok/s 2626878.07 | norm 0.3682 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9865 | train 2.965737 | val 3.367506 | tok/s 2628919.82 | norm 0.4156 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9866 | train 2.918373 | val 3.367506 | tok/s 2627672.50 | norm 0.3958 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9867 | train 2.984753 | val 3.367506 | tok/s 2629795.10 | norm 0.3944 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9868 | train 3.135446 | val 3.367506 | tok/s 2623437.03 | norm 0.4416 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9869 | train 2.948424 | val 3.367506 | tok/s 2628406.82 | norm 0.5455 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9870 | train 2.956206 | val 3.367506 | tok/s 2628828.81 | norm 0.3848 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9871 | train 2.901677 | val 3.367506 | tok/s 2632458.76 | norm 0.3468 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9872 | train 2.998711 | val 3.367506 | tok/s 2607027.21 | norm 0.3881 | dt 201.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9873 | train 2.973190 | val 3.367506 | tok/s 2629514.44 | norm 0.3885 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9874 | train 2.951055 | val 3.367506 | tok/s 2628219.32 | norm 0.4063 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9875 | train 3.167070 | val 3.367506 | tok/s 2622231.18 | norm 0.3871 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9876 | train 3.129734 | val 3.367506 | tok/s 2625284.06 | norm 0.3744 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9877 | train 2.934798 | val 3.367506 | tok/s 2630400.74 | norm 0.3762 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9878 | train 2.972356 | val 3.367506 | tok/s 2627348.05 | norm 0.3684 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9879 | train 2.923272 | val 3.367506 | tok/s 2630618.04 | norm 0.3930 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9880 | train 2.999083 | val 3.367506 | tok/s 2632822.24 | norm 0.4226 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9881 | train 3.003773 | val 3.367506 | tok/s 2629739.56 | norm 0.3983 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9882 | train 2.803535 | val 3.367506 | tok/s 2630968.31 | norm 0.3913 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9883 | train 2.992360 | val 3.367506 | tok/s 2627698.64 | norm 0.4020 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9884 | train 3.054802 | val 3.367506 | tok/s 2630354.71 | norm 0.3805 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9885 | train 2.969560 | val 3.367506 | tok/s 2630155.89 | norm 0.3834 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9886 | train 3.067234 | val 3.367506 | tok/s 2626907.08 | norm 0.3590 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9887 | train 2.991436 | val 3.367506 | tok/s 2627932.62 | norm 0.3483 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9888 | train 3.013853 | val 3.367506 | tok/s 2632475.39 | norm 0.3552 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9889 | train 2.998971 | val 3.367506 | tok/s 2631632.42 | norm 0.3809 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9890 | train 2.947972 | val 3.367506 | tok/s 2631105.04 | norm 0.3819 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9891 | train 2.970873 | val 3.367506 | tok/s 2624773.25 | norm 0.4140 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9892 | train 3.183865 | val 3.367506 | tok/s 2623347.86 | norm 0.4119 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9893 | train 3.075432 | val 3.367506 | tok/s 2627842.73 | norm 0.3762 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9894 | train 2.946896 | val 3.367506 | tok/s 2623819.16 | norm 0.3988 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9895 | train 2.980792 | val 3.367506 | tok/s 2633224.93 | norm 0.4336 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9896 | train 2.876407 | val 3.367506 | tok/s 2632884.58 | norm 0.4209 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9897 | train 3.085392 | val 3.367506 | tok/s 2632650.09 | norm 0.4176 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9898 | train 2.997218 | val 3.367506 | tok/s 2630359.19 | norm 0.4031 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9899 | train 3.073250 | val 3.367506 | tok/s 2622855.70 | norm 0.4159 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9900 | train 2.936704 | val 3.367506 | tok/s 2621886.17 | norm 0.4231 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9901 | train 3.062659 | val 3.367506 | tok/s 2629460.97 | norm 0.4391 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9902 | train 3.028748 | val 3.367506 | tok/s 2622659.69 | norm 0.4280 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9903 | train 3.026263 | val 3.367506 | tok/s 2627886.72 | norm 0.4375 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9904 | train 2.841510 | val 3.367506 | tok/s 2630046.13 | norm 0.4129 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9905 | train 2.935812 | val 3.367506 | tok/s 2626149.84 | norm 0.4112 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9906 | train 2.961725 | val 3.367506 | tok/s 2630757.00 | norm 0.4171 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9907 | train 2.959715 | val 3.367506 | tok/s 2628102.88 | norm 0.3943 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9908 | train 3.034401 | val 3.367506 | tok/s 2630759.33 | norm 0.4280 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9909 | train 2.970057 | val 3.367506 | tok/s 2631541.57 | norm 0.4246 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9910 | train 3.092493 | val 3.367506 | tok/s 2630397.50 | norm 0.4251 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9911 | train 2.979494 | val 3.367506 | tok/s 2509688.46 | norm 0.4106 | dt 208.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9912 | train 3.026829 | val 3.367506 | tok/s 2465563.87 | norm 0.4152 | dt 212.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9913 | train 3.034595 | val 3.367506 | tok/s 2624126.66 | norm 0.4536 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9914 | train 3.100179 | val 3.367506 | tok/s 2627637.62 | norm 0.4124 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9915 | train 2.999919 | val 3.367506 | tok/s 2632927.11 | norm 0.3655 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9916 | train 2.899679 | val 3.367506 | tok/s 2625550.72 | norm 0.3714 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9917 | train 3.083978 | val 3.367506 | tok/s 2628003.69 | norm 0.3836 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9918 | train 2.989976 | val 3.367506 | tok/s 2630173.03 | norm 0.4104 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9919 | train 3.056050 | val 3.367506 | tok/s 2630809.75 | norm 0.4087 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9920 | train 3.062198 | val 3.367506 | tok/s 2632449.26 | norm 0.4105 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9921 | train 3.039873 | val 3.367506 | tok/s 2628312.01 | norm 0.3874 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9922 | train 2.996523 | val 3.367506 | tok/s 2626056.58 | norm 0.3921 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9923 | train 3.039544 | val 3.367506 | tok/s 2632104.23 | norm 0.4063 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9924 | train 2.988832 | val 3.367506 | tok/s 2630318.89 | norm 0.4163 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9925 | train 3.038745 | val 3.367506 | tok/s 2630550.25 | norm 0.4183 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9926 | train 2.999258 | val 3.367506 | tok/s 2626780.27 | norm 0.4218 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9927 | train 3.030632 | val 3.367506 | tok/s 2628701.10 | norm 0.4299 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9928 | train 2.999797 | val 3.367506 | tok/s 2630015.47 | norm 0.3932 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9929 | train 2.985291 | val 3.367506 | tok/s 2632996.76 | norm 0.4175 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9930 | train 2.987171 | val 3.367506 | tok/s 2633002.21 | norm 0.4418 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9931 | train 2.944438 | val 3.367506 | tok/s 2632800.89 | norm 0.3825 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9932 | train 3.060965 | val 3.367506 | tok/s 2630412.47 | norm 0.3614 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9933 | train 3.055670 | val 3.367506 | tok/s 2630150.22 | norm 0.4107 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9934 | train 2.997578 | val 3.367506 | tok/s 2624546.45 | norm 0.4174 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9935 | train 2.898027 | val 3.367506 | tok/s 2626451.03 | norm 0.3664 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9936 | train 2.966212 | val 3.367506 | tok/s 2631559.91 | norm 0.4042 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9937 | train 3.006655 | val 3.367506 | tok/s 2628545.43 | norm 0.3965 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9938 | train 3.083822 | val 3.367506 | tok/s 2633573.45 | norm 0.4044 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9939 | train 3.052333 | val 3.367506 | tok/s 2632527.06 | norm 0.4102 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9940 | train 2.933202 | val 3.367506 | tok/s 2623114.03 | norm 0.4350 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9941 | train 2.948962 | val 3.367506 | tok/s 2629494.52 | norm 0.4057 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9942 | train 3.011773 | val 3.367506 | tok/s 2632434.79 | norm 0.4401 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9943 | train 3.039555 | val 3.367506 | tok/s 2630714.93 | norm 0.4230 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9944 | train 2.982552 | val 3.367506 | tok/s 2624677.33 | norm 0.3965 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9945 | train 2.898588 | val 3.367506 | tok/s 2630337.96 | norm 0.4136 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9946 | train 3.069673 | val 3.367506 | tok/s 2626855.95 | norm 0.3811 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9947 | train 3.007662 | val 3.367506 | tok/s 2628126.35 | norm 0.3850 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9948 | train 2.959058 | val 3.367506 | tok/s 2627650.60 | norm 0.4004 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9949 | train 3.161546 | val 3.367506 | tok/s 2630721.20 | norm 0.3966 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9950 | train 3.024633 | val 3.367506 | tok/s 2626016.02 | norm 0.3656 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9951 | train 2.951447 | val 3.367506 | tok/s 2631059.28 | norm 0.3802 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9952 | train 2.950544 | val 3.367506 | tok/s 2626392.80 | norm 0.4107 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9953 | train 3.153502 | val 3.367506 | tok/s 2628005.47 | norm 0.4168 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9954 | train 2.997329 | val 3.367506 | tok/s 2632990.34 | norm 0.4122 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9955 | train 3.016649 | val 3.367506 | tok/s 2629776.30 | norm 0.4023 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9956 | train 3.156543 | val 3.367506 | tok/s 2629881.64 | norm 0.4093 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9957 | train 2.967096 | val 3.367506 | tok/s 2581256.28 | norm 0.3970 | dt 203.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9958 | train 2.978732 | val 3.367506 | tok/s 2595196.03 | norm 0.4225 | dt 202.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9959 | train 2.942861 | val 3.367506 | tok/s 2594545.74 | norm 0.4054 | dt 202.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9960 | train 3.013721 | val 3.367506 | tok/s 2508020.86 | norm 0.4001 | dt 209.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9961 | train 2.962720 | val 3.367506 | tok/s 2575940.20 | norm 0.4197 | dt 203.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9962 | train 2.932254 | val 3.367506 | tok/s 2595991.05 | norm 0.4169 | dt 201.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9963 | train 2.949381 | val 3.367506 | tok/s 2591559.44 | norm 0.4344 | dt 202.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9964 | train 3.006999 | val 3.367506 | tok/s 2617083.45 | norm 0.4267 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9965 | train 2.997167 | val 3.367506 | tok/s 2622691.18 | norm 0.4430 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9966 | train 2.999377 | val 3.367506 | tok/s 2617776.85 | norm 0.3936 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9967 | train 2.972952 | val 3.367506 | tok/s 2624202.64 | norm 0.4114 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9968 | train 3.074519 | val 3.367506 | tok/s 2619988.42 | norm 0.3891 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9969 | train 3.034663 | val 3.367506 | tok/s 2626808.09 | norm 0.3710 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9970 | train 2.851740 | val 3.367506 | tok/s 2629919.06 | norm 0.4123 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9971 | train 3.095480 | val 3.367506 | tok/s 2618445.06 | norm 0.4420 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9972 | train 3.129178 | val 3.367506 | tok/s 2623886.52 | norm 0.4224 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9973 | train 2.976721 | val 3.367506 | tok/s 2625552.40 | norm 0.4221 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9974 | train 3.023434 | val 3.367506 | tok/s 2624298.46 | norm 0.4179 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9975 | train 3.159385 | val 3.367506 | tok/s 2624373.35 | norm 0.4629 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9976 | train 3.022779 | val 3.367506 | tok/s 2620790.48 | norm 0.4770 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9977 | train 3.046259 | val 3.367506 | tok/s 2617119.49 | norm 0.4119 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9978 | train 3.077325 | val 3.367506 | tok/s 2624905.33 | norm 0.3849 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9979 | train 3.077567 | val 3.367506 | tok/s 2627775.03 | norm 0.3984 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9980 | train 2.969027 | val 3.367506 | tok/s 2628582.70 | norm 0.3790 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9981 | train 2.936663 | val 3.367506 | tok/s 2624183.28 | norm 0.3800 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9982 | train 2.936282 | val 3.367506 | tok/s 2622895.69 | norm 0.4415 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9983 | train 2.968750 | val 3.367506 | tok/s 2630845.98 | norm 0.3975 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9984 | train 2.987312 | val 3.367506 | tok/s 2633184.87 | norm 0.3735 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9985 | train 2.969898 | val 3.367506 | tok/s 2632525.80 | norm 0.4024 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9986 | train 3.127229 | val 3.367506 | tok/s 2631247.91 | norm 0.4390 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9987 | train 2.886059 | val 3.367506 | tok/s 2628033.70 | norm 0.4741 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9988 | train 2.980833 | val 3.367506 | tok/s 2631549.78 | norm 0.4036 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9989 | train 3.044106 | val 3.367506 | tok/s 2630316.79 | norm 0.4142 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9990 | train 2.961752 | val 3.367506 | tok/s 2626542.77 | norm 0.3677 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9991 | train 3.030535 | val 3.367506 | tok/s 2626539.68 | norm 0.3449 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9992 | train 2.893906 | val 3.367506 | tok/s 2625243.32 | norm 0.3956 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9993 | train 2.968715 | val 3.367506 | tok/s 2625559.20 | norm 0.3865 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9994 | train 3.104202 | val 3.367506 | tok/s 2626457.81 | norm 0.3820 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9995 | train 3.034205 | val 3.367506 | tok/s 2628015.90 | norm 0.3995 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9996 | train 3.080948 | val 3.367506 | tok/s 2629666.50 | norm 0.3976 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9997 | train 3.120761 | val 3.367506 | tok/s 2627567.33 | norm 0.4155 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9998 | train 3.118257 | val 3.367506 | tok/s 2629158.59 | norm 0.4466 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 9999 | train 3.015435 | val 3.367506 | tok/s 2627666.89 | norm 0.4043 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 10000 | val_loss 3.364277 step 10000 | train 3.057643 | val 3.364277 | tok/s 2615735.63 | norm 0.3768 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10001 | train 2.975780 | val 3.364277 | tok/s 2625923.04 | norm 0.4145 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10002 | train 3.131015 | val 3.364277 | tok/s 2630274.43 | norm 0.5704 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10003 | train 2.850502 | val 3.364277 | tok/s 2627630.57 | norm 0.3390 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10004 | train 3.046948 | val 3.364277 | tok/s 2629492.40 | norm 0.3488 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10005 | train 3.072710 | val 3.364277 | tok/s 2626640.85 | norm 0.3575 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10006 | train 2.939781 | val 3.364277 | tok/s 2632870.42 | norm 0.3511 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10007 | train 2.978503 | val 3.364277 | tok/s 2624414.03 | norm 0.3546 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10008 | train 3.058531 | val 3.364277 | tok/s 2626982.58 | norm 0.3989 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10009 | train 2.971004 | val 3.364277 | tok/s 2628189.52 | norm 0.3690 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10010 | train 2.891223 | val 3.364277 | tok/s 2628279.80 | norm 0.4079 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10011 | train 3.137101 | val 3.364277 | tok/s 2632830.36 | norm 0.4092 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10012 | train 2.961189 | val 3.364277 | tok/s 2627422.19 | norm 0.3852 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10013 | train 3.068665 | val 3.364277 | tok/s 2626948.69 | norm 0.4303 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10014 | train 3.046779 | val 3.364277 | tok/s 2625815.49 | norm 0.3956 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10015 | train 3.009007 | val 3.364277 | tok/s 2630398.65 | norm 0.4010 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10016 | train 2.982665 | val 3.364277 | tok/s 2629687.81 | norm 0.4199 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10017 | train 3.045234 | val 3.364277 | tok/s 2625939.43 | norm 0.4438 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10018 | train 3.083346 | val 3.364277 | tok/s 2626495.41 | norm 0.4113 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10019 | train 3.036210 | val 3.364277 | tok/s 2627638.59 | norm 0.3730 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10020 | train 3.014717 | val 3.364277 | tok/s 2627478.16 | norm 0.3915 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10021 | train 2.995787 | val 3.364277 | tok/s 2576492.32 | norm 0.4185 | dt 203.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10022 | train 2.893691 | val 3.364277 | tok/s 2624278.88 | norm 0.3860 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10023 | train 3.064591 | val 3.364277 | tok/s 2627329.05 | norm 0.3747 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10024 | train 2.974491 | val 3.364277 | tok/s 2623806.17 | norm 0.3679 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10025 | train 2.878110 | val 3.364277 | tok/s 2627507.27 | norm 0.4149 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10026 | train 3.035750 | val 3.364277 | tok/s 2627488.72 | norm 0.3869 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10027 | train 3.074302 | val 3.364277 | tok/s 2512283.01 | norm 0.4356 | dt 208.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10028 | train 3.053755 | val 3.364277 | tok/s 2619696.58 | norm 0.3776 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10029 | train 2.953946 | val 3.364277 | tok/s 2629976.26 | norm 0.3233 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10030 | train 3.010144 | val 3.364277 | tok/s 2631909.38 | norm 0.3561 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10031 | train 3.075054 | val 3.364277 | tok/s 2632059.41 | norm 0.3961 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10032 | train 3.015684 | val 3.364277 | tok/s 2622278.36 | norm 0.3832 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10033 | train 2.975712 | val 3.364277 | tok/s 2627596.51 | norm 0.3971 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10034 | train 3.020024 | val 3.364277 | tok/s 2496921.88 | norm 0.3931 | dt 209.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10035 | train 3.175306 | val 3.364277 | tok/s 2624055.09 | norm 0.4843 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10036 | train 2.978836 | val 3.364277 | tok/s 2629916.23 | norm 0.3780 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10037 | train 3.039157 | val 3.364277 | tok/s 2627277.48 | norm 0.3346 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10038 | train 2.991010 | val 3.364277 | tok/s 2629094.11 | norm 0.3632 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10039 | train 2.992919 | val 3.364277 | tok/s 2628445.64 | norm 0.4050 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10040 | train 3.037610 | val 3.364277 | tok/s 2628422.17 | norm 0.3760 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10041 | train 3.083001 | val 3.364277 | tok/s 2626837.64 | norm 0.3852 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10042 | train 2.976363 | val 3.364277 | tok/s 2626793.46 | norm 0.4319 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10043 | train 3.047687 | val 3.364277 | tok/s 2629310.87 | norm 0.4022 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10044 | train 2.857499 | val 3.364277 | tok/s 2630661.53 | norm 0.3994 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10045 | train 3.105341 | val 3.364277 | tok/s 2605504.53 | norm 0.4158 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10046 | train 2.857409 | val 3.364277 | tok/s 2626221.42 | norm 0.4229 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10047 | train 2.952710 | val 3.364277 | tok/s 2629591.81 | norm 0.4065 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10048 | train 2.975602 | val 3.364277 | tok/s 2625021.59 | norm 0.3951 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10049 | train 2.932265 | val 3.364277 | tok/s 2618658.92 | norm 0.3834 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10050 | train 3.044462 | val 3.364277 | tok/s 2630354.93 | norm 0.3997 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10051 | train 2.932311 | val 3.364277 | tok/s 2626210.08 | norm 0.3875 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10052 | train 2.919991 | val 3.364277 | tok/s 2624235.61 | norm 0.3919 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10053 | train 3.003608 | val 3.364277 | tok/s 2627598.22 | norm 0.4176 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10054 | train 3.040081 | val 3.364277 | tok/s 2625958.83 | norm 0.3779 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10055 | train 3.072886 | val 3.364277 | tok/s 2630327.22 | norm 0.3554 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10056 | train 2.907154 | val 3.364277 | tok/s 2623028.66 | norm 0.4003 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10057 | train 3.196337 | val 3.364277 | tok/s 2628141.93 | norm 0.4793 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10058 | train 2.942961 | val 3.364277 | tok/s 2632557.55 | norm 0.3918 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10059 | train 3.032645 | val 3.364277 | tok/s 2623066.03 | norm 0.3706 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10060 | train 3.019591 | val 3.364277 | tok/s 2626223.60 | norm 0.4261 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10061 | train 3.125999 | val 3.364277 | tok/s 2626711.71 | norm 0.4145 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10062 | train 3.122337 | val 3.364277 | tok/s 2626316.65 | norm 0.3792 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10063 | train 2.999754 | val 3.364277 | tok/s 2627361.59 | norm 0.4033 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10064 | train 2.996305 | val 3.364277 | tok/s 2629224.48 | norm 0.4050 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10065 | train 2.994455 | val 3.364277 | tok/s 2630933.61 | norm 0.4011 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10066 | train 2.964496 | val 3.364277 | tok/s 2629333.52 | norm 0.4067 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10067 | train 3.059413 | val 3.364277 | tok/s 2626054.04 | norm 0.4439 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10068 | train 2.821208 | val 3.364277 | tok/s 2626415.89 | norm 0.3880 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10069 | train 3.033288 | val 3.364277 | tok/s 2626465.85 | norm 0.3726 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10070 | train 2.935070 | val 3.364277 | tok/s 2627176.64 | norm 0.4165 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10071 | train 2.944828 | val 3.364277 | tok/s 2629098.52 | norm 0.4373 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10072 | train 3.036060 | val 3.364277 | tok/s 2627454.41 | norm 0.4254 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10073 | train 3.049032 | val 3.364277 | tok/s 2626722.21 | norm 0.4126 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10074 | train 3.041222 | val 3.364277 | tok/s 2624907.24 | norm 0.4177 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10075 | train 3.087651 | val 3.364277 | tok/s 2624796.63 | norm 0.4050 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10076 | train 3.054072 | val 3.364277 | tok/s 2628247.95 | norm 0.4129 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10077 | train 3.012048 | val 3.364277 | tok/s 2626893.43 | norm 0.4075 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10078 | train 2.930508 | val 3.364277 | tok/s 2627335.06 | norm 0.3521 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10079 | train 2.985360 | val 3.364277 | tok/s 2619835.53 | norm 0.3909 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10080 | train 3.097063 | val 3.364277 | tok/s 2623803.13 | norm 0.4266 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10081 | train 2.992251 | val 3.364277 | tok/s 2626106.47 | norm 0.4165 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10082 | train 2.919890 | val 3.364277 | tok/s 2538697.83 | norm 0.4063 | dt 206.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10083 | train 3.021554 | val 3.364277 | tok/s 2629233.90 | norm 0.4115 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10084 | train 3.043423 | val 3.364277 | tok/s 2631559.15 | norm 0.3953 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10085 | train 3.166285 | val 3.364277 | tok/s 2626440.37 | norm 0.4274 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10086 | train 3.051584 | val 3.364277 | tok/s 2627925.84 | norm 0.4287 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10087 | train 3.127133 | val 3.364277 | tok/s 2627402.05 | norm 0.4517 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10088 | train 3.000969 | val 3.364277 | tok/s 2630579.96 | norm 0.3413 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10089 | train 2.867091 | val 3.364277 | tok/s 2629947.75 | norm 0.3638 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10090 | train 2.982534 | val 3.364277 | tok/s 2633938.86 | norm 0.3818 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10091 | train 2.954359 | val 3.364277 | tok/s 2627862.40 | norm 0.3563 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10092 | train 3.037785 | val 3.364277 | tok/s 2627180.22 | norm 0.3644 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10093 | train 2.966421 | val 3.364277 | tok/s 2630094.33 | norm 0.4141 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10094 | train 2.945432 | val 3.364277 | tok/s 2633534.07 | norm 0.3824 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10095 | train 2.979401 | val 3.364277 | tok/s 2633092.74 | norm 0.3947 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10096 | train 3.213429 | val 3.364277 | tok/s 2628706.90 | norm 0.5088 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10097 | train 3.112441 | val 3.364277 | tok/s 2521262.00 | norm 0.3790 | dt 207.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10098 | train 3.030617 | val 3.364277 | tok/s 2576943.76 | norm 0.3461 | dt 203.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10099 | train 2.953548 | val 3.364277 | tok/s 2628445.38 | norm 0.3932 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10100 | train 3.141769 | val 3.364277 | tok/s 2627454.20 | norm 0.3823 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10101 | train 2.985392 | val 3.364277 | tok/s 2629497.21 | norm 0.3645 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10102 | train 3.094506 | val 3.364277 | tok/s 2627022.09 | norm 0.4086 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10103 | train 3.134308 | val 3.364277 | tok/s 2626769.79 | norm 0.4537 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10104 | train 2.929974 | val 3.364277 | tok/s 2628139.59 | norm 0.4319 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10105 | train 2.949533 | val 3.364277 | tok/s 2629205.50 | norm 0.3981 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10106 | train 3.000934 | val 3.364277 | tok/s 2625222.12 | norm 0.4256 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10107 | train 3.032295 | val 3.364277 | tok/s 2626388.12 | norm 0.4045 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10108 | train 2.875790 | val 3.364277 | tok/s 2602758.27 | norm 0.3929 | dt 201.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10109 | train 2.947791 | val 3.364277 | tok/s 2624956.37 | norm 0.3995 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10110 | train 3.023936 | val 3.364277 | tok/s 2627107.09 | norm 0.4127 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10111 | train 2.997209 | val 3.364277 | tok/s 2624075.42 | norm 0.3998 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10112 | train 2.969658 | val 3.364277 | tok/s 2624553.28 | norm 0.3670 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10113 | train 3.005947 | val 3.364277 | tok/s 2628206.09 | norm 0.3675 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10114 | train 3.016899 | val 3.364277 | tok/s 2628074.30 | norm 0.3659 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10115 | train 3.035392 | val 3.364277 | tok/s 2628117.28 | norm 0.4253 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10116 | train 3.006161 | val 3.364277 | tok/s 2625237.52 | norm 0.4141 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10117 | train 3.004484 | val 3.364277 | tok/s 2628194.04 | norm 0.3931 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10118 | train 2.942724 | val 3.364277 | tok/s 2628604.81 | norm 0.3821 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10119 | train 2.989159 | val 3.364277 | tok/s 2631053.84 | norm 0.3796 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10120 | train 2.908588 | val 3.364277 | tok/s 2628181.15 | norm 0.3842 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10121 | train 2.864408 | val 3.364277 | tok/s 2623836.72 | norm 0.4083 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10122 | train 3.081000 | val 3.364277 | tok/s 2620057.19 | norm 0.4344 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10123 | train 3.000988 | val 3.364277 | tok/s 2626796.63 | norm 0.4110 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10124 | train 3.047238 | val 3.364277 | tok/s 2625426.97 | norm 0.3657 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10125 | train 2.949376 | val 3.364277 | tok/s 2627811.43 | norm 0.4600 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10126 | train 3.037246 | val 3.364277 | tok/s 2627106.01 | norm 0.4323 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10127 | train 2.935394 | val 3.364277 | tok/s 2627236.88 | norm 0.3665 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10128 | train 3.090952 | val 3.364277 | tok/s 2628453.99 | norm 0.4166 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10129 | train 3.002596 | val 3.364277 | tok/s 2627532.18 | norm 0.3595 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10130 | train 3.071556 | val 3.364277 | tok/s 2626911.67 | norm 0.3707 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10131 | train 2.897688 | val 3.364277 | tok/s 2626259.51 | norm 0.4112 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10132 | train 2.822944 | val 3.364277 | tok/s 2625712.87 | norm 0.4069 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10133 | train 3.056027 | val 3.364277 | tok/s 2625546.55 | norm 0.4524 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10134 | train 3.141293 | val 3.364277 | tok/s 2627358.59 | norm 0.5019 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10135 | train 3.092544 | val 3.364277 | tok/s 2629323.42 | norm 0.3922 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10136 | train 3.000661 | val 3.364277 | tok/s 2625191.66 | norm 0.3868 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10137 | train 3.005860 | val 3.364277 | tok/s 2627268.86 | norm 0.3921 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10138 | train 3.002141 | val 3.364277 | tok/s 2630551.20 | norm 0.3743 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10139 | train 2.929292 | val 3.364277 | tok/s 2626822.06 | norm 0.3580 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10140 | train 3.021174 | val 3.364277 | tok/s 2625612.49 | norm 0.4004 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10141 | train 2.991689 | val 3.364277 | tok/s 2623878.38 | norm 0.3700 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10142 | train 3.272932 | val 3.364277 | tok/s 2625474.07 | norm 0.4131 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10143 | train 3.011269 | val 3.364277 | tok/s 2627191.68 | norm 0.3728 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10144 | train 3.073464 | val 3.364277 | tok/s 2629066.67 | norm 0.3186 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10145 | train 2.980212 | val 3.364277 | tok/s 2628044.85 | norm 0.3466 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10146 | train 2.868542 | val 3.364277 | tok/s 2630736.17 | norm 0.3932 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10147 | train 2.986889 | val 3.364277 | tok/s 2632372.89 | norm 0.4271 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10148 | train 3.054759 | val 3.364277 | tok/s 2627153.74 | norm 0.4275 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10149 | train 2.957430 | val 3.364277 | tok/s 2630633.87 | norm 0.3964 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10150 | train 3.047940 | val 3.364277 | tok/s 2626195.38 | norm 0.3767 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10151 | train 3.114462 | val 3.364277 | tok/s 2621619.10 | norm 0.4270 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10152 | train 2.993301 | val 3.364277 | tok/s 2627986.99 | norm 0.4193 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10153 | train 2.942465 | val 3.364277 | tok/s 2626156.97 | norm 0.3886 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10154 | train 2.942223 | val 3.364277 | tok/s 2627428.41 | norm 0.3973 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10155 | train 2.902527 | val 3.364277 | tok/s 2626703.85 | norm 0.3920 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10156 | train 2.983600 | val 3.364277 | tok/s 2529736.57 | norm 0.3676 | dt 207.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10157 | train 3.165088 | val 3.364277 | tok/s 2473039.06 | norm 0.4511 | dt 212.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10158 | train 2.973300 | val 3.364277 | tok/s 2625782.41 | norm 0.4083 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10159 | train 3.037820 | val 3.364277 | tok/s 2625698.44 | norm 0.3770 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10160 | train 2.946025 | val 3.364277 | tok/s 2626624.50 | norm 0.4154 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10161 | train 2.968367 | val 3.364277 | tok/s 2627278.73 | norm 0.4225 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10162 | train 3.078894 | val 3.364277 | tok/s 2628297.53 | norm 0.3844 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10163 | train 2.931281 | val 3.364277 | tok/s 2624286.67 | norm 0.3778 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10164 | train 2.978023 | val 3.364277 | tok/s 2611778.82 | norm 0.4041 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10165 | train 2.879564 | val 3.364277 | tok/s 2597780.35 | norm 0.3813 | dt 201.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10166 | train 2.965527 | val 3.364277 | tok/s 2599923.31 | norm 0.3616 | dt 201.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10167 | train 2.896018 | val 3.364277 | tok/s 2627920.56 | norm 0.3880 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10168 | train 3.006818 | val 3.364277 | tok/s 2628440.98 | norm 0.3975 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10169 | train 2.999432 | val 3.364277 | tok/s 2629233.81 | norm 0.4145 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10170 | train 2.938922 | val 3.364277 | tok/s 2630112.28 | norm 0.4097 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10171 | train 3.055690 | val 3.364277 | tok/s 2627735.12 | norm 0.4283 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10172 | train 3.032804 | val 3.364277 | tok/s 2627041.69 | norm 0.4110 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10173 | train 2.944041 | val 3.364277 | tok/s 2622130.73 | norm 0.4162 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10174 | train 2.955975 | val 3.364277 | tok/s 2625026.53 | norm 0.4290 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10175 | train 3.164763 | val 3.364277 | tok/s 2626082.23 | norm 0.4822 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10176 | train 2.976378 | val 3.364277 | tok/s 2629635.80 | norm 0.4475 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10177 | train 3.104731 | val 3.364277 | tok/s 2629846.49 | norm 0.4613 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10178 | train 3.066922 | val 3.364277 | tok/s 2629956.36 | norm 0.4431 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10179 | train 3.015316 | val 3.364277 | tok/s 2628946.19 | norm 0.4020 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10180 | train 3.080835 | val 3.364277 | tok/s 2629339.51 | norm 0.4197 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10181 | train 2.920659 | val 3.364277 | tok/s 2624380.78 | norm 0.4472 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10182 | train 2.958601 | val 3.364277 | tok/s 2624335.42 | norm 0.4107 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10183 | train 3.166247 | val 3.364277 | tok/s 2624791.08 | norm 0.3880 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10184 | train 3.014763 | val 3.364277 | tok/s 2609999.07 | norm 0.4086 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10185 | train 2.939597 | val 3.364277 | tok/s 2629869.60 | norm 0.4007 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10186 | train 3.006272 | val 3.364277 | tok/s 2625944.70 | norm 0.3947 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10187 | train 2.886780 | val 3.364277 | tok/s 2629825.10 | norm 0.3881 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10188 | train 3.086136 | val 3.364277 | tok/s 2628309.74 | norm 0.4178 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10189 | train 2.992653 | val 3.364277 | tok/s 2626744.50 | norm 0.4318 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10190 | train 2.981786 | val 3.364277 | tok/s 2630208.10 | norm 0.4049 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10191 | train 2.979702 | val 3.364277 | tok/s 2627110.30 | norm 0.3865 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10192 | train 3.149211 | val 3.364277 | tok/s 2626090.75 | norm 0.3834 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10193 | train 3.045025 | val 3.364277 | tok/s 2628491.06 | norm 0.3903 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10194 | train 2.864151 | val 3.364277 | tok/s 2633322.88 | norm 0.3926 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10195 | train 2.895277 | val 3.364277 | tok/s 2628937.76 | norm 0.4367 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10196 | train 2.872312 | val 3.364277 | tok/s 2627435.26 | norm 0.4170 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10197 | train 2.908143 | val 3.364277 | tok/s 2631140.92 | norm 0.4079 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10198 | train 3.119513 | val 3.364277 | tok/s 2631702.10 | norm 0.4434 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10199 | train 2.928981 | val 3.364277 | tok/s 2626499.57 | norm 0.3922 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10200 | train 2.978625 | val 3.364277 | tok/s 2629291.27 | norm 0.3787 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10201 | train 3.060314 | val 3.364277 | tok/s 2625572.39 | norm 0.4211 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10202 | train 2.886571 | val 3.364277 | tok/s 2628772.53 | norm 0.3868 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10203 | train 3.056368 | val 3.364277 | tok/s 2629370.22 | norm 0.3680 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10204 | train 3.041591 | val 3.364277 | tok/s 2628324.26 | norm 0.3756 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10205 | train 3.200560 | val 3.364277 | tok/s 2628306.96 | norm 0.4175 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10206 | train 2.959322 | val 3.364277 | tok/s 2625544.25 | norm 0.4309 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10207 | train 3.005862 | val 3.364277 | tok/s 2632096.61 | norm 0.4161 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10208 | train 2.989533 | val 3.364277 | tok/s 2633610.57 | norm 0.3950 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10209 | train 2.997276 | val 3.364277 | tok/s 2631594.97 | norm 0.3788 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10210 | train 2.862980 | val 3.364277 | tok/s 2627031.62 | norm 0.4743 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10211 | train 2.955886 | val 3.364277 | tok/s 2630055.21 | norm 0.4376 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10212 | train 2.997201 | val 3.364277 | tok/s 2626865.99 | norm 0.4249 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10213 | train 2.864293 | val 3.364277 | tok/s 2627902.00 | norm 0.4578 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10214 | train 2.996686 | val 3.364277 | tok/s 2627267.13 | norm 0.4012 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10215 | train 2.899650 | val 3.364277 | tok/s 2629210.55 | norm 0.3879 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10216 | train 3.111803 | val 3.364277 | tok/s 2632533.54 | norm 0.4133 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10217 | train 3.003642 | val 3.364277 | tok/s 2610203.56 | norm 0.4099 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10218 | train 3.073018 | val 3.364277 | tok/s 2424874.52 | norm 0.3692 | dt 216.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10219 | train 3.143842 | val 3.364277 | tok/s 2623170.45 | norm 0.3561 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10220 | train 2.999672 | val 3.364277 | tok/s 2632146.63 | norm 0.3960 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10221 | train 3.161114 | val 3.364277 | tok/s 2632601.31 | norm 0.4010 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10222 | train 3.004292 | val 3.364277 | tok/s 2625056.20 | norm 0.3561 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10223 | train 3.077270 | val 3.364277 | tok/s 2627961.13 | norm 0.3577 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10224 | train 2.981717 | val 3.364277 | tok/s 2628067.30 | norm 0.3692 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10225 | train 3.021788 | val 3.364277 | tok/s 2522916.60 | norm 0.4075 | dt 207.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10226 | train 2.943255 | val 3.364277 | tok/s 2630751.44 | norm 0.4241 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10227 | train 3.029426 | val 3.364277 | tok/s 2568200.65 | norm 0.4253 | dt 204.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10228 | train 2.901187 | val 3.364277 | tok/s 2630226.31 | norm 0.4184 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10229 | train 2.877607 | val 3.364277 | tok/s 2624575.00 | norm 0.4219 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10230 | train 3.057538 | val 3.364277 | tok/s 2633554.26 | norm 0.4561 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10231 | train 3.069971 | val 3.364277 | tok/s 2629680.35 | norm 0.3882 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10232 | train 2.997423 | val 3.364277 | tok/s 2626179.89 | norm 0.3737 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10233 | train 3.083348 | val 3.364277 | tok/s 2631594.74 | norm 0.4108 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10234 | train 2.953739 | val 3.364277 | tok/s 2627239.22 | norm 0.3779 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10235 | train 3.020228 | val 3.364277 | tok/s 2630403.36 | norm 0.3731 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10236 | train 3.072350 | val 3.364277 | tok/s 2624307.59 | norm 0.4014 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10237 | train 2.893897 | val 3.364277 | tok/s 2633094.56 | norm 0.4194 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10238 | train 2.962370 | val 3.364277 | tok/s 2623856.22 | norm 0.3924 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10239 | train 3.046510 | val 3.364277 | tok/s 2631348.43 | norm 0.3796 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10240 | train 2.946172 | val 3.364277 | tok/s 2632826.95 | norm 0.3815 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10241 | train 3.039776 | val 3.364277 | tok/s 2632087.85 | norm 0.3903 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10242 | train 2.962050 | val 3.364277 | tok/s 2617771.10 | norm 0.3571 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10243 | train 2.999630 | val 3.364277 | tok/s 2635552.13 | norm 0.3711 | dt 198.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10244 | train 2.852621 | val 3.364277 | tok/s 2632024.90 | norm 0.3751 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10245 | train 3.045840 | val 3.364277 | tok/s 2626797.18 | norm 0.3883 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10246 | train 2.970934 | val 3.364277 | tok/s 2625275.83 | norm 0.4049 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10247 | train 3.090422 | val 3.364277 | tok/s 2629539.82 | norm 0.3708 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10248 | train 3.036589 | val 3.364277 | tok/s 2630255.14 | norm 0.3601 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10249 | train 3.042463 | val 3.364277 | tok/s 2630372.65 | norm 0.3772 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 10250 | val_loss 3.360761 step 10250 | train 3.108864 | val 3.360761 | tok/s 2594465.10 | norm 0.4221 | dt 202.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10251 | train 3.057718 | val 3.360761 | tok/s 2628868.78 | norm 0.4678 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10252 | train 2.959618 | val 3.360761 | tok/s 2630584.69 | norm 0.3830 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10253 | train 2.976722 | val 3.360761 | tok/s 2626643.83 | norm 0.3970 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10254 | train 2.923276 | val 3.360761 | tok/s 2630228.51 | norm 0.4105 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10255 | train 3.224504 | val 3.360761 | tok/s 2631299.23 | norm 0.3969 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10256 | train 3.000383 | val 3.360761 | tok/s 2625549.79 | norm 0.3554 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10257 | train 2.975747 | val 3.360761 | tok/s 2627988.78 | norm 0.3735 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10258 | train 3.019144 | val 3.360761 | tok/s 2632076.59 | norm 0.4065 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10259 | train 2.846278 | val 3.360761 | tok/s 2633780.33 | norm 0.3565 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10260 | train 2.854629 | val 3.360761 | tok/s 2624325.70 | norm 0.3581 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10261 | train 3.073925 | val 3.360761 | tok/s 2632317.90 | norm 0.4303 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10262 | train 2.969363 | val 3.360761 | tok/s 2632728.43 | norm 0.4023 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10263 | train 3.025425 | val 3.360761 | tok/s 2631110.98 | norm 0.3664 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10264 | train 2.942849 | val 3.360761 | tok/s 2625668.76 | norm 0.3803 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10265 | train 2.956449 | val 3.360761 | tok/s 2629005.32 | norm 0.4182 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10266 | train 3.216978 | val 3.360761 | tok/s 2628470.09 | norm 0.4786 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10267 | train 3.005121 | val 3.360761 | tok/s 2630091.94 | norm 0.4245 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10268 | train 2.989262 | val 3.360761 | tok/s 2630598.56 | norm 0.3979 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10269 | train 2.999162 | val 3.360761 | tok/s 2632301.27 | norm 0.4665 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10270 | train 2.958280 | val 3.360761 | tok/s 2587497.99 | norm 0.4419 | dt 202.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10271 | train 2.934141 | val 3.360761 | tok/s 2632217.31 | norm 0.4956 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10272 | train 2.984000 | val 3.360761 | tok/s 2634342.60 | norm 0.4551 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10273 | train 3.130448 | val 3.360761 | tok/s 2624051.15 | norm 0.4347 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10274 | train 2.909591 | val 3.360761 | tok/s 2625968.31 | norm 0.4121 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10275 | train 3.100170 | val 3.360761 | tok/s 2629856.00 | norm 0.3811 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10276 | train 2.909820 | val 3.360761 | tok/s 2629534.86 | norm 0.3491 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10277 | train 3.008329 | val 3.360761 | tok/s 2627077.63 | norm 0.3691 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10278 | train 2.962485 | val 3.360761 | tok/s 2629683.60 | norm 0.3968 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10279 | train 3.011252 | val 3.360761 | tok/s 2627039.61 | norm 0.3916 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10280 | train 2.997814 | val 3.360761 | tok/s 2628742.09 | norm 0.4071 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10281 | train 2.983673 | val 3.360761 | tok/s 2625210.75 | norm 0.4131 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10282 | train 3.153118 | val 3.360761 | tok/s 2628707.13 | norm 0.5485 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10283 | train 3.131397 | val 3.360761 | tok/s 2631816.24 | norm 0.5255 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10284 | train 3.056247 | val 3.360761 | tok/s 2627854.03 | norm 0.4214 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10285 | train 3.009260 | val 3.360761 | tok/s 2412504.73 | norm 0.3826 | dt 217.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10286 | train 3.072052 | val 3.360761 | tok/s 2627217.51 | norm 0.4421 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10287 | train 2.938593 | val 3.360761 | tok/s 2629462.50 | norm 0.4079 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10288 | train 3.040527 | val 3.360761 | tok/s 2631197.72 | norm 0.3835 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10289 | train 3.074745 | val 3.360761 | tok/s 2630466.80 | norm 0.3843 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10290 | train 3.170928 | val 3.360761 | tok/s 2631451.03 | norm 0.3819 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10291 | train 2.896623 | val 3.360761 | tok/s 2628124.86 | norm 0.3979 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10292 | train 3.078285 | val 3.360761 | tok/s 2631651.29 | norm 0.3968 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10293 | train 3.045136 | val 3.360761 | tok/s 2626035.65 | norm 0.3787 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10294 | train 3.008939 | val 3.360761 | tok/s 2632047.86 | norm 0.4082 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10295 | train 2.958236 | val 3.360761 | tok/s 2630164.56 | norm 0.3940 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10296 | train 2.814198 | val 3.360761 | tok/s 2630897.00 | norm 0.4230 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10297 | train 2.932176 | val 3.360761 | tok/s 2631297.58 | norm 0.4089 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10298 | train 2.940390 | val 3.360761 | tok/s 2630751.15 | norm 0.4076 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10299 | train 3.002297 | val 3.360761 | tok/s 2629268.45 | norm 0.3841 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10300 | train 2.977271 | val 3.360761 | tok/s 2579471.60 | norm 0.3578 | dt 203.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10301 | train 3.051225 | val 3.360761 | tok/s 2603485.58 | norm 0.3815 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10302 | train 3.086379 | val 3.360761 | tok/s 2629631.88 | norm 0.4022 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10303 | train 2.819209 | val 3.360761 | tok/s 2632217.43 | norm 0.4156 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10304 | train 3.220047 | val 3.360761 | tok/s 2631873.96 | norm 0.4832 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10305 | train 2.965782 | val 3.360761 | tok/s 2626474.89 | norm 0.3966 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10306 | train 3.021011 | val 3.360761 | tok/s 2623646.14 | norm 0.3500 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10307 | train 3.015423 | val 3.360761 | tok/s 2628229.62 | norm 0.4009 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10308 | train 3.014911 | val 3.360761 | tok/s 2626357.17 | norm 0.4067 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10309 | train 3.005001 | val 3.360761 | tok/s 2629451.10 | norm 0.4303 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10310 | train 3.186084 | val 3.360761 | tok/s 2624280.93 | norm 0.4953 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10311 | train 2.809986 | val 3.360761 | tok/s 2629104.68 | norm 0.3591 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10312 | train 2.915183 | val 3.360761 | tok/s 2628630.36 | norm 0.3641 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10313 | train 2.936194 | val 3.360761 | tok/s 2629738.42 | norm 0.4912 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10314 | train 3.142500 | val 3.360761 | tok/s 2634967.81 | norm 0.4044 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10315 | train 3.220759 | val 3.360761 | tok/s 2629018.33 | norm 0.3620 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10316 | train 3.103127 | val 3.360761 | tok/s 2631541.53 | norm 0.4068 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10317 | train 2.919634 | val 3.360761 | tok/s 2629313.11 | norm 0.4158 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10318 | train 2.879717 | val 3.360761 | tok/s 2624386.07 | norm 0.4145 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10319 | train 2.908508 | val 3.360761 | tok/s 2629363.06 | norm 0.4115 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10320 | train 3.217659 | val 3.360761 | tok/s 2632257.37 | norm 0.4180 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10321 | train 2.880125 | val 3.360761 | tok/s 2625928.67 | norm 0.3773 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10322 | train 3.191959 | val 3.360761 | tok/s 2630202.45 | norm 0.4731 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10323 | train 3.085673 | val 3.360761 | tok/s 2630154.64 | norm 0.4406 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10324 | train 3.041863 | val 3.360761 | tok/s 2629967.68 | norm 0.3421 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10325 | train 2.999676 | val 3.360761 | tok/s 2628132.28 | norm 0.3793 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10326 | train 3.044012 | val 3.360761 | tok/s 2630693.62 | norm 0.4174 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10327 | train 3.088964 | val 3.360761 | tok/s 2632178.51 | norm 0.4052 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10328 | train 3.103959 | val 3.360761 | tok/s 2632399.48 | norm 0.4215 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10329 | train 3.018693 | val 3.360761 | tok/s 2631878.78 | norm 0.4154 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10330 | train 2.909156 | val 3.360761 | tok/s 2629700.32 | norm 0.3966 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10331 | train 3.184663 | val 3.360761 | tok/s 2628744.61 | norm 0.4496 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10332 | train 2.865586 | val 3.360761 | tok/s 2632358.02 | norm 0.3969 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10333 | train 2.945339 | val 3.360761 | tok/s 2630178.47 | norm 0.3749 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10334 | train 3.119768 | val 3.360761 | tok/s 2632026.66 | norm 0.4271 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10335 | train 3.018675 | val 3.360761 | tok/s 2632895.40 | norm 0.4258 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10336 | train 2.990295 | val 3.360761 | tok/s 2630344.85 | norm 0.4006 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10337 | train 3.081883 | val 3.360761 | tok/s 2629883.95 | norm 0.4446 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10338 | train 2.918147 | val 3.360761 | tok/s 2627099.18 | norm 0.3603 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10339 | train 2.916302 | val 3.360761 | tok/s 2627069.73 | norm 0.3887 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10340 | train 2.943589 | val 3.360761 | tok/s 2629756.99 | norm 0.4090 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10341 | train 2.917148 | val 3.360761 | tok/s 2630144.95 | norm 0.3577 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10342 | train 2.902754 | val 3.360761 | tok/s 2627448.83 | norm 0.3720 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10343 | train 3.042816 | val 3.360761 | tok/s 2629284.05 | norm 0.3634 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10344 | train 2.957103 | val 3.360761 | tok/s 2626269.76 | norm 0.3160 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10345 | train 3.029516 | val 3.360761 | tok/s 2629800.91 | norm 0.3351 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10346 | train 3.122485 | val 3.360761 | tok/s 2626953.77 | norm 0.3656 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10347 | train 3.042866 | val 3.360761 | tok/s 2630152.14 | norm 0.4013 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10348 | train 3.022279 | val 3.360761 | tok/s 2626615.38 | norm 0.4174 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10349 | train 3.030969 | val 3.360761 | tok/s 2627777.81 | norm 0.3897 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10350 | train 3.110122 | val 3.360761 | tok/s 2629839.37 | norm 0.4129 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10351 | train 2.949753 | val 3.360761 | tok/s 2627163.10 | norm 0.4173 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10352 | train 2.960318 | val 3.360761 | tok/s 2629753.44 | norm 0.4199 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10353 | train 2.985173 | val 3.360761 | tok/s 2621642.32 | norm 0.4131 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10354 | train 2.899671 | val 3.360761 | tok/s 2630008.96 | norm 0.4042 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10355 | train 3.084283 | val 3.360761 | tok/s 2631676.02 | norm 0.4239 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10356 | train 3.019419 | val 3.360761 | tok/s 2629703.58 | norm 0.3927 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10357 | train 3.001843 | val 3.360761 | tok/s 2633176.78 | norm 0.4063 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10358 | train 2.945636 | val 3.360761 | tok/s 2628965.94 | norm 0.4420 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10359 | train 3.039174 | val 3.360761 | tok/s 2628769.36 | norm 0.4177 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10360 | train 3.040960 | val 3.360761 | tok/s 2631010.72 | norm 0.4332 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10361 | train 3.028350 | val 3.360761 | tok/s 2482206.78 | norm 0.3984 | dt 211.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10362 | train 3.010556 | val 3.360761 | tok/s 2633979.32 | norm 0.4280 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10363 | train 2.829451 | val 3.360761 | tok/s 2550308.16 | norm 0.3728 | dt 205.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10364 | train 3.057787 | val 3.360761 | tok/s 2633438.91 | norm 0.3494 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10365 | train 2.882024 | val 3.360761 | tok/s 2623452.41 | norm 0.4188 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10366 | train 3.113963 | val 3.360761 | tok/s 2631353.86 | norm 0.4315 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10367 | train 3.011572 | val 3.360761 | tok/s 2624799.69 | norm 0.4031 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10368 | train 3.141261 | val 3.360761 | tok/s 2633722.69 | norm 0.3979 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10369 | train 2.893444 | val 3.360761 | tok/s 2616891.80 | norm 0.3827 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10370 | train 2.975073 | val 3.360761 | tok/s 2629161.80 | norm 0.4036 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10371 | train 3.003073 | val 3.360761 | tok/s 2623663.45 | norm 0.4385 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10372 | train 2.971556 | val 3.360761 | tok/s 2634256.98 | norm 0.4214 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10373 | train 2.984936 | val 3.360761 | tok/s 2631385.84 | norm 0.3777 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10374 | train 2.970952 | val 3.360761 | tok/s 2623393.68 | norm 0.3594 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10375 | train 3.205780 | val 3.360761 | tok/s 2615660.19 | norm 0.4047 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10376 | train 3.044041 | val 3.360761 | tok/s 2629664.13 | norm 0.3665 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10377 | train 2.933114 | val 3.360761 | tok/s 2625906.85 | norm 0.3451 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10378 | train 3.009891 | val 3.360761 | tok/s 2626091.40 | norm 0.3766 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10379 | train 2.857183 | val 3.360761 | tok/s 2627647.75 | norm 0.4192 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10380 | train 2.976100 | val 3.360761 | tok/s 2626627.22 | norm 0.3953 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10381 | train 2.933880 | val 3.360761 | tok/s 2626518.65 | norm 0.3784 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10382 | train 3.076300 | val 3.360761 | tok/s 2626191.02 | norm 0.3872 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10383 | train 2.958786 | val 3.360761 | tok/s 2621042.61 | norm 0.3888 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10384 | train 2.994791 | val 3.360761 | tok/s 2626861.25 | norm 0.3384 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10385 | train 3.037476 | val 3.360761 | tok/s 2631868.85 | norm 0.3828 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10386 | train 2.953308 | val 3.360761 | tok/s 2629822.43 | norm 0.4021 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10387 | train 2.961968 | val 3.360761 | tok/s 2627705.42 | norm 0.3918 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10388 | train 3.010072 | val 3.360761 | tok/s 2619631.34 | norm 0.4134 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10389 | train 3.059002 | val 3.360761 | tok/s 2627483.22 | norm 0.4095 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10390 | train 2.977788 | val 3.360761 | tok/s 2630637.25 | norm 0.4034 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10391 | train 3.255260 | val 3.360761 | tok/s 2628799.42 | norm 0.4336 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10392 | train 2.984306 | val 3.360761 | tok/s 2629340.59 | norm 0.4070 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10393 | train 3.022760 | val 3.360761 | tok/s 2632886.47 | norm 0.3772 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10394 | train 3.150362 | val 3.360761 | tok/s 2628103.32 | norm 0.3763 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10395 | train 2.948372 | val 3.360761 | tok/s 2629993.38 | norm 0.3842 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10396 | train 2.908908 | val 3.360761 | tok/s 2629994.49 | norm 0.3937 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10397 | train 2.996198 | val 3.360761 | tok/s 2629703.04 | norm 0.3986 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10398 | train 3.017030 | val 3.360761 | tok/s 2630002.93 | norm 0.4314 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10399 | train 3.077585 | val 3.360761 | tok/s 2625874.25 | norm 0.3982 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10400 | train 2.931173 | val 3.360761 | tok/s 2630093.14 | norm 0.3671 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10401 | train 2.988379 | val 3.360761 | tok/s 2627002.86 | norm 0.3936 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10402 | train 2.945090 | val 3.360761 | tok/s 2627582.49 | norm 0.4065 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10403 | train 3.027204 | val 3.360761 | tok/s 2630625.64 | norm 0.3991 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10404 | train 2.982307 | val 3.360761 | tok/s 2630288.90 | norm 0.4081 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10405 | train 2.929283 | val 3.360761 | tok/s 2626792.67 | norm 0.3930 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10406 | train 2.980589 | val 3.360761 | tok/s 2630832.44 | norm 0.3781 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10407 | train 3.127753 | val 3.360761 | tok/s 2629164.10 | norm 0.4041 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10408 | train 2.974375 | val 3.360761 | tok/s 2622073.31 | norm 0.4169 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10409 | train 2.949735 | val 3.360761 | tok/s 2633364.48 | norm 0.4025 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10410 | train 2.939426 | val 3.360761 | tok/s 2624943.52 | norm 0.4143 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10411 | train 2.958457 | val 3.360761 | tok/s 2629336.53 | norm 0.3979 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10412 | train 3.049210 | val 3.360761 | tok/s 2627479.00 | norm 0.4138 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10413 | train 3.172578 | val 3.360761 | tok/s 2632196.41 | norm 0.4378 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10414 | train 2.976876 | val 3.360761 | tok/s 2556122.78 | norm 0.3829 | dt 205.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10415 | train 2.979178 | val 3.360761 | tok/s 2455609.98 | norm 0.3945 | dt 213.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10416 | train 3.000222 | val 3.360761 | tok/s 2630157.87 | norm 0.4145 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10417 | train 3.061029 | val 3.360761 | tok/s 2634852.76 | norm 0.4086 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10418 | train 3.042752 | val 3.360761 | tok/s 2632555.23 | norm 0.4189 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10419 | train 2.922747 | val 3.360761 | tok/s 2627307.41 | norm 0.4306 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10420 | train 2.821494 | val 3.360761 | tok/s 2631286.30 | norm 0.4356 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10421 | train 3.039122 | val 3.360761 | tok/s 2625481.25 | norm 0.4253 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10422 | train 2.945659 | val 3.360761 | tok/s 2552659.66 | norm 0.4102 | dt 205.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10423 | train 3.037493 | val 3.360761 | tok/s 2561619.80 | norm 0.4234 | dt 204.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10424 | train 3.059006 | val 3.360761 | tok/s 2621315.43 | norm 0.4170 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10425 | train 2.958814 | val 3.360761 | tok/s 2625467.91 | norm 0.4233 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10426 | train 2.970619 | val 3.360761 | tok/s 2631466.47 | norm 0.4196 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10427 | train 2.890089 | val 3.360761 | tok/s 2628331.50 | norm 0.3894 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10428 | train 2.929891 | val 3.360761 | tok/s 2627754.20 | norm 0.3969 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10429 | train 2.968924 | val 3.360761 | tok/s 2618293.40 | norm 0.4328 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10430 | train 2.918470 | val 3.360761 | tok/s 2628916.03 | norm 0.3983 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10431 | train 2.957729 | val 3.360761 | tok/s 2627775.50 | norm 0.3797 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10432 | train 3.068740 | val 3.360761 | tok/s 2620443.79 | norm 0.3989 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10433 | train 2.956521 | val 3.360761 | tok/s 2625957.57 | norm 0.4153 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10434 | train 2.972081 | val 3.360761 | tok/s 2630924.09 | norm 0.4092 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10435 | train 2.986464 | val 3.360761 | tok/s 2633448.91 | norm 0.4268 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10436 | train 3.092062 | val 3.360761 | tok/s 2629418.26 | norm 0.3640 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10437 | train 3.012307 | val 3.360761 | tok/s 2628871.23 | norm 0.3905 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10438 | train 3.010816 | val 3.360761 | tok/s 2627689.84 | norm 0.4708 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10439 | train 3.051715 | val 3.360761 | tok/s 2613204.92 | norm 0.4097 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10440 | train 3.003204 | val 3.360761 | tok/s 2632164.11 | norm 0.3719 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10441 | train 3.025816 | val 3.360761 | tok/s 2629398.33 | norm 0.3815 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10442 | train 3.119689 | val 3.360761 | tok/s 2630900.06 | norm 0.3772 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10443 | train 3.058061 | val 3.360761 | tok/s 2628918.60 | norm 0.3615 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10444 | train 3.118490 | val 3.360761 | tok/s 2633530.41 | norm 0.4307 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10445 | train 2.941949 | val 3.360761 | tok/s 2629611.61 | norm 0.3647 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10446 | train 2.989014 | val 3.360761 | tok/s 2621977.41 | norm 0.3605 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10447 | train 2.985883 | val 3.360761 | tok/s 2631106.61 | norm 0.3696 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10448 | train 2.926021 | val 3.360761 | tok/s 2631422.04 | norm 0.3821 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10449 | train 3.007523 | val 3.360761 | tok/s 2628154.42 | norm 0.3756 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10450 | train 3.036416 | val 3.360761 | tok/s 2630500.65 | norm 0.3775 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10451 | train 2.905543 | val 3.360761 | tok/s 2629482.32 | norm 0.4102 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10452 | train 2.936202 | val 3.360761 | tok/s 2628525.45 | norm 0.3800 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10453 | train 3.088352 | val 3.360761 | tok/s 2629061.52 | norm 0.3967 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10454 | train 2.871410 | val 3.360761 | tok/s 2630682.81 | norm 0.4119 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10455 | train 3.073314 | val 3.360761 | tok/s 2632326.58 | norm 0.4178 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10456 | train 3.148961 | val 3.360761 | tok/s 2629147.36 | norm 0.4897 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10457 | train 2.971395 | val 3.360761 | tok/s 2628833.51 | norm 0.3944 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10458 | train 2.993918 | val 3.360761 | tok/s 2624079.05 | norm 0.3759 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10459 | train 2.943079 | val 3.360761 | tok/s 2624169.29 | norm 0.3810 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10460 | train 2.969788 | val 3.360761 | tok/s 2625321.01 | norm 0.4165 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10461 | train 3.137462 | val 3.360761 | tok/s 2627755.35 | norm 0.4729 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10462 | train 3.073182 | val 3.360761 | tok/s 2628746.89 | norm 0.4461 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10463 | train 2.988594 | val 3.360761 | tok/s 2630727.22 | norm 0.3595 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10464 | train 3.039921 | val 3.360761 | tok/s 2627421.18 | norm 0.3956 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10465 | train 2.855474 | val 3.360761 | tok/s 2631149.55 | norm 0.4282 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10466 | train 3.150690 | val 3.360761 | tok/s 2628438.15 | norm 0.4731 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10467 | train 3.068331 | val 3.360761 | tok/s 2632068.20 | norm 0.4447 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10468 | train 2.995337 | val 3.360761 | tok/s 2628381.95 | norm 0.4033 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10469 | train 3.023638 | val 3.360761 | tok/s 2626868.88 | norm 0.4143 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10470 | train 2.871680 | val 3.360761 | tok/s 2626190.31 | norm 0.4013 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10471 | train 2.827909 | val 3.360761 | tok/s 2525940.91 | norm 0.4572 | dt 207.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10472 | train 2.999940 | val 3.360761 | tok/s 2476211.81 | norm 0.4236 | dt 211.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10473 | train 2.968822 | val 3.360761 | tok/s 2627769.46 | norm 0.4509 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10474 | train 2.971712 | val 3.360761 | tok/s 2628296.44 | norm 0.4545 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10475 | train 3.121308 | val 3.360761 | tok/s 2602279.86 | norm 0.3834 | dt 201.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10476 | train 3.054315 | val 3.360761 | tok/s 2631205.77 | norm 0.3758 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10477 | train 3.028028 | val 3.360761 | tok/s 2633168.06 | norm 0.4004 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10478 | train 3.100951 | val 3.360761 | tok/s 2633405.67 | norm 0.4000 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10479 | train 2.981699 | val 3.360761 | tok/s 2626613.67 | norm 0.3995 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10480 | train 3.057458 | val 3.360761 | tok/s 2625665.44 | norm 0.3719 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10481 | train 3.098192 | val 3.360761 | tok/s 2629456.28 | norm 0.3758 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10482 | train 2.982483 | val 3.360761 | tok/s 2611925.90 | norm 0.3887 | dt 200.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10483 | train 2.960467 | val 3.360761 | tok/s 2630604.92 | norm 0.3740 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10484 | train 2.910510 | val 3.360761 | tok/s 2628987.30 | norm 0.4122 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10485 | train 2.954136 | val 3.360761 | tok/s 2632576.17 | norm 0.4263 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10486 | train 3.003203 | val 3.360761 | tok/s 2628247.53 | norm 0.4276 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10487 | train 2.989912 | val 3.360761 | tok/s 2630884.37 | norm 0.4066 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10488 | train 3.002718 | val 3.360761 | tok/s 2634490.89 | norm 0.4017 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10489 | train 2.928020 | val 3.360761 | tok/s 2633561.63 | norm 0.3611 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10490 | train 2.889927 | val 3.360761 | tok/s 2616791.30 | norm 0.3616 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10491 | train 3.050092 | val 3.360761 | tok/s 2627578.90 | norm 0.4179 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10492 | train 2.907888 | val 3.360761 | tok/s 2627115.02 | norm 0.4205 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10493 | train 2.849336 | val 3.360761 | tok/s 2629281.74 | norm 0.3939 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10494 | train 2.986691 | val 3.360761 | tok/s 2623860.35 | norm 0.3508 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10495 | train 3.006162 | val 3.360761 | tok/s 2631945.42 | norm 0.4123 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10496 | train 2.928475 | val 3.360761 | tok/s 2631810.23 | norm 0.4403 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10497 | train 2.951702 | val 3.360761 | tok/s 2629397.63 | norm 0.3936 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10498 | train 3.042593 | val 3.360761 | tok/s 2631135.66 | norm 0.3775 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10499 | train 2.988119 | val 3.360761 | tok/s 2630728.80 | norm 0.3919 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 10500 | val_loss 3.365266 step 10500 | train 2.922074 | val 3.365266 | tok/s 2606808.21 | norm 0.3614 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10501 | train 3.142486 | val 3.365266 | tok/s 2573361.54 | norm 0.3578 | dt 203.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10502 | train 2.917157 | val 3.365266 | tok/s 2631256.57 | norm 0.3397 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10503 | train 3.035653 | val 3.365266 | tok/s 2626654.59 | norm 0.3865 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10504 | train 2.983563 | val 3.365266 | tok/s 2627939.45 | norm 0.4059 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10505 | train 3.050487 | val 3.365266 | tok/s 2619851.75 | norm 0.4100 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10506 | train 3.003383 | val 3.365266 | tok/s 2630085.96 | norm 0.4212 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10507 | train 2.984537 | val 3.365266 | tok/s 2627458.44 | norm 0.4068 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10508 | train 3.017823 | val 3.365266 | tok/s 2622794.94 | norm 0.3912 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10509 | train 2.991121 | val 3.365266 | tok/s 2625707.61 | norm 0.3865 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10510 | train 3.010942 | val 3.365266 | tok/s 2618512.99 | norm 0.3817 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10511 | train 2.987373 | val 3.365266 | tok/s 2626071.42 | norm 0.3924 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10512 | train 2.929497 | val 3.365266 | tok/s 2629826.60 | norm 0.4166 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10513 | train 3.064090 | val 3.365266 | tok/s 2633444.78 | norm 0.4407 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10514 | train 3.002089 | val 3.365266 | tok/s 2631469.32 | norm 0.4208 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10515 | train 3.018412 | val 3.365266 | tok/s 2630220.29 | norm 0.4210 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10516 | train 3.030405 | val 3.365266 | tok/s 2624058.72 | norm 0.3979 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10517 | train 2.930180 | val 3.365266 | tok/s 2626963.91 | norm 0.3627 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10518 | train 2.886542 | val 3.365266 | tok/s 2630242.08 | norm 0.3976 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10519 | train 3.024327 | val 3.365266 | tok/s 2628097.46 | norm 0.4284 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10520 | train 3.055974 | val 3.365266 | tok/s 2624775.85 | norm 0.3872 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10521 | train 2.891235 | val 3.365266 | tok/s 2628871.44 | norm 0.3753 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10522 | train 3.083986 | val 3.365266 | tok/s 2633120.85 | norm 0.3876 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10523 | train 2.978497 | val 3.365266 | tok/s 2628722.09 | norm 0.4073 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10524 | train 2.979559 | val 3.365266 | tok/s 2628547.53 | norm 0.3677 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10525 | train 3.146264 | val 3.365266 | tok/s 2630786.06 | norm 0.3789 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10526 | train 3.031894 | val 3.365266 | tok/s 2628718.14 | norm 0.4047 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10527 | train 2.897619 | val 3.365266 | tok/s 2628452.47 | norm 0.3839 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10528 | train 2.998297 | val 3.365266 | tok/s 2628064.88 | norm 0.3508 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10529 | train 3.011445 | val 3.365266 | tok/s 2629419.28 | norm 0.4211 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10530 | train 3.038710 | val 3.365266 | tok/s 2627293.37 | norm 0.4313 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10531 | train 2.893280 | val 3.365266 | tok/s 2623356.30 | norm 0.4213 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10532 | train 3.033149 | val 3.365266 | tok/s 2626685.67 | norm 0.3922 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10533 | train 2.999405 | val 3.365266 | tok/s 2630290.68 | norm 0.3997 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10534 | train 2.973050 | val 3.365266 | tok/s 2630684.90 | norm 0.4432 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10535 | train 2.968720 | val 3.365266 | tok/s 2625937.04 | norm 0.4247 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10536 | train 2.928180 | val 3.365266 | tok/s 2627885.77 | norm 0.3880 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10537 | train 2.982005 | val 3.365266 | tok/s 2626708.34 | norm 0.4096 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10538 | train 3.081232 | val 3.365266 | tok/s 2568200.01 | norm 0.3870 | dt 204.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10539 | train 3.220770 | val 3.365266 | tok/s 2545141.04 | norm 0.4603 | dt 206.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10540 | train 3.115912 | val 3.365266 | tok/s 2634247.79 | norm 0.4227 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10541 | train 2.992488 | val 3.365266 | tok/s 2629599.83 | norm 0.3707 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10542 | train 2.936856 | val 3.365266 | tok/s 2633477.90 | norm 0.3921 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10543 | train 3.029421 | val 3.365266 | tok/s 2629335.14 | norm 0.4547 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10544 | train 2.999640 | val 3.365266 | tok/s 2632561.22 | norm 0.4089 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10545 | train 2.987741 | val 3.365266 | tok/s 2627081.28 | norm 0.3682 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10546 | train 2.957739 | val 3.365266 | tok/s 2629780.88 | norm 0.4073 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10547 | train 2.895995 | val 3.365266 | tok/s 2482531.52 | norm 0.4325 | dt 211.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10548 | train 3.084985 | val 3.365266 | tok/s 2630950.25 | norm 0.3788 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10549 | train 2.990825 | val 3.365266 | tok/s 2631019.67 | norm 0.3518 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10550 | train 2.959851 | val 3.365266 | tok/s 2625460.05 | norm 0.3931 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10551 | train 2.907222 | val 3.365266 | tok/s 2609599.49 | norm 0.3906 | dt 200.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10552 | train 2.996583 | val 3.365266 | tok/s 2628745.01 | norm 0.3822 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10553 | train 2.949212 | val 3.365266 | tok/s 2626046.07 | norm 0.4274 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10554 | train 3.059161 | val 3.365266 | tok/s 2583016.86 | norm 0.3976 | dt 202.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10555 | train 2.975718 | val 3.365266 | tok/s 2633670.19 | norm 0.3961 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10556 | train 3.047942 | val 3.365266 | tok/s 2629945.80 | norm 0.4281 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10557 | train 2.987026 | val 3.365266 | tok/s 2622864.28 | norm 0.3857 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10558 | train 2.992338 | val 3.365266 | tok/s 2631151.65 | norm 0.3887 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10559 | train 3.044475 | val 3.365266 | tok/s 2628494.34 | norm 0.3817 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10560 | train 2.942203 | val 3.365266 | tok/s 2625590.74 | norm 0.3959 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10561 | train 3.067728 | val 3.365266 | tok/s 2628079.21 | norm 0.4176 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10562 | train 3.017596 | val 3.365266 | tok/s 2628350.53 | norm 0.3680 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10563 | train 3.357692 | val 3.365266 | tok/s 2631213.06 | norm 0.3996 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10564 | train 3.055360 | val 3.365266 | tok/s 2624751.08 | norm 0.4660 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10565 | train 3.057792 | val 3.365266 | tok/s 2631304.05 | norm 0.4504 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10566 | train 2.976276 | val 3.365266 | tok/s 2623310.29 | norm 0.4111 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10567 | train 2.999554 | val 3.365266 | tok/s 2625816.42 | norm 0.3942 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10568 | train 3.071512 | val 3.365266 | tok/s 2629563.04 | norm 0.4133 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10569 | train 2.936801 | val 3.365266 | tok/s 2630584.94 | norm 0.4273 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10570 | train 3.136185 | val 3.365266 | tok/s 2632764.36 | norm 0.4098 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10571 | train 2.953283 | val 3.365266 | tok/s 2629830.92 | norm 0.3911 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10572 | train 3.064578 | val 3.365266 | tok/s 2633395.38 | norm 0.4075 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10573 | train 2.966799 | val 3.365266 | tok/s 2633989.92 | norm 0.4069 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10574 | train 3.009370 | val 3.365266 | tok/s 2625833.36 | norm 0.3923 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10575 | train 2.957250 | val 3.365266 | tok/s 2624284.32 | norm 0.3915 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10576 | train 2.994243 | val 3.365266 | tok/s 2626760.21 | norm 0.3949 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10577 | train 2.893601 | val 3.365266 | tok/s 2626476.41 | norm 0.3756 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10578 | train 2.945683 | val 3.365266 | tok/s 2626776.41 | norm 0.4005 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10579 | train 3.040797 | val 3.365266 | tok/s 2630753.71 | norm 0.4048 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10580 | train 3.038815 | val 3.365266 | tok/s 2629074.23 | norm 0.3925 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10581 | train 2.900095 | val 3.365266 | tok/s 2629793.92 | norm 0.3799 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10582 | train 3.036064 | val 3.365266 | tok/s 2629551.11 | norm 0.3660 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10583 | train 3.019247 | val 3.365266 | tok/s 2617506.03 | norm 0.4157 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10584 | train 2.908482 | val 3.365266 | tok/s 2633575.90 | norm 0.4004 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10585 | train 2.970460 | val 3.365266 | tok/s 2629087.23 | norm 0.3748 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10586 | train 2.940827 | val 3.365266 | tok/s 2631031.13 | norm 0.3776 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10587 | train 3.002668 | val 3.365266 | tok/s 2631778.81 | norm 0.4063 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10588 | train 2.947324 | val 3.365266 | tok/s 2609761.57 | norm 0.3997 | dt 200.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10589 | train 2.963425 | val 3.365266 | tok/s 2627645.61 | norm 0.4336 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10590 | train 2.979600 | val 3.365266 | tok/s 2629471.76 | norm 0.4439 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10591 | train 3.056698 | val 3.365266 | tok/s 2627160.22 | norm 0.3926 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10592 | train 3.012417 | val 3.365266 | tok/s 2628965.76 | norm 0.4013 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10593 | train 3.015483 | val 3.365266 | tok/s 2628730.64 | norm 0.3855 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10594 | train 3.077680 | val 3.365266 | tok/s 2622058.14 | norm 0.3671 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10595 | train 3.045800 | val 3.365266 | tok/s 2624922.38 | norm 0.3849 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10596 | train 3.103088 | val 3.365266 | tok/s 2550317.16 | norm 0.4351 | dt 205.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10597 | train 2.865907 | val 3.365266 | tok/s 2633739.76 | norm 0.3962 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10598 | train 3.107872 | val 3.365266 | tok/s 2550445.22 | norm 0.3865 | dt 205.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10599 | train 3.004100 | val 3.365266 | tok/s 2627811.34 | norm 0.4136 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10600 | train 3.045066 | val 3.365266 | tok/s 2631917.24 | norm 0.3677 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10601 | train 2.984286 | val 3.365266 | tok/s 2630127.80 | norm 0.3765 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10602 | train 2.927188 | val 3.365266 | tok/s 2632368.42 | norm 0.3729 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10603 | train 3.037763 | val 3.365266 | tok/s 2628879.80 | norm 0.3551 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10604 | train 2.991284 | val 3.365266 | tok/s 2628786.99 | norm 0.3767 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10605 | train 3.007388 | val 3.365266 | tok/s 2630097.86 | norm 0.3812 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10606 | train 2.976384 | val 3.365266 | tok/s 2434760.80 | norm 0.3507 | dt 215.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10607 | train 3.001062 | val 3.365266 | tok/s 2631373.28 | norm 0.3746 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10608 | train 3.000313 | val 3.365266 | tok/s 2631162.24 | norm 0.4141 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10609 | train 2.857265 | val 3.365266 | tok/s 2630041.64 | norm 0.4284 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10610 | train 2.907803 | val 3.365266 | tok/s 2627718.96 | norm 0.4150 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10611 | train 3.038234 | val 3.365266 | tok/s 2635223.84 | norm 0.4759 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10612 | train 2.906509 | val 3.365266 | tok/s 2629937.63 | norm 0.4107 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10613 | train 3.025517 | val 3.365266 | tok/s 2626820.32 | norm 0.3730 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10614 | train 2.933685 | val 3.365266 | tok/s 2630494.87 | norm 0.4495 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10615 | train 3.101254 | val 3.365266 | tok/s 2626094.78 | norm 0.4239 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10616 | train 2.937103 | val 3.365266 | tok/s 2628050.66 | norm 0.4385 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10617 | train 2.970673 | val 3.365266 | tok/s 2630668.46 | norm 0.4101 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10618 | train 3.140291 | val 3.365266 | tok/s 2625597.06 | norm 0.3856 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10619 | train 3.081157 | val 3.365266 | tok/s 2627353.21 | norm 0.4200 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10620 | train 2.968446 | val 3.365266 | tok/s 2629700.20 | norm 0.4121 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10621 | train 2.848103 | val 3.365266 | tok/s 2626642.47 | norm 0.3784 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10622 | train 3.013238 | val 3.365266 | tok/s 2632509.48 | norm 0.3700 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10623 | train 2.931193 | val 3.365266 | tok/s 2630878.36 | norm 0.3836 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10624 | train 2.851968 | val 3.365266 | tok/s 2630358.84 | norm 0.4937 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10625 | train 3.073387 | val 3.365266 | tok/s 2629375.43 | norm 0.3650 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10626 | train 3.024759 | val 3.365266 | tok/s 2631811.96 | norm 0.3844 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10627 | train 2.941845 | val 3.365266 | tok/s 2631281.27 | norm 0.4040 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10628 | train 2.895718 | val 3.365266 | tok/s 2627323.55 | norm 0.4318 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10629 | train 2.986772 | val 3.365266 | tok/s 2634004.52 | norm 0.3912 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10630 | train 3.025166 | val 3.365266 | tok/s 2627677.38 | norm 0.4182 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10631 | train 3.011623 | val 3.365266 | tok/s 2630122.38 | norm 0.4462 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10632 | train 2.916453 | val 3.365266 | tok/s 2630566.16 | norm 0.3911 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10633 | train 2.938800 | val 3.365266 | tok/s 2629954.24 | norm 0.3890 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10634 | train 2.964519 | val 3.365266 | tok/s 2631747.11 | norm 0.4083 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10635 | train 2.952510 | val 3.365266 | tok/s 2629454.59 | norm 0.3840 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10636 | train 3.117362 | val 3.365266 | tok/s 2629299.59 | norm 0.4995 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10637 | train 2.970188 | val 3.365266 | tok/s 2631844.93 | norm 0.3809 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10638 | train 3.046941 | val 3.365266 | tok/s 2634791.74 | norm 0.3668 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10639 | train 2.984742 | val 3.365266 | tok/s 2633751.97 | norm 0.3768 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10640 | train 2.931758 | val 3.365266 | tok/s 2634382.44 | norm 0.3944 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10641 | train 3.003783 | val 3.365266 | tok/s 2633255.52 | norm 0.4317 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10642 | train 3.079197 | val 3.365266 | tok/s 2627003.34 | norm 0.4026 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10643 | train 2.988396 | val 3.365266 | tok/s 2629074.58 | norm 0.3842 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10644 | train 3.028764 | val 3.365266 | tok/s 2629033.57 | norm 0.3862 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10645 | train 3.041046 | val 3.365266 | tok/s 2634470.78 | norm 0.3909 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10646 | train 3.070202 | val 3.365266 | tok/s 2631115.39 | norm 0.4382 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10647 | train 2.986081 | val 3.365266 | tok/s 2627316.61 | norm 0.4485 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10648 | train 3.003259 | val 3.365266 | tok/s 2626641.24 | norm 0.4399 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10649 | train 3.009964 | val 3.365266 | tok/s 2629148.53 | norm 0.3934 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10650 | train 2.906663 | val 3.365266 | tok/s 2626759.10 | norm 0.3736 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10651 | train 3.023771 | val 3.365266 | tok/s 2628160.57 | norm 0.3956 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10652 | train 3.108767 | val 3.365266 | tok/s 2627459.83 | norm 0.4294 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10653 | train 3.083402 | val 3.365266 | tok/s 2628932.80 | norm 0.3828 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10654 | train 3.055178 | val 3.365266 | tok/s 2627271.41 | norm 0.4125 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10655 | train 3.070562 | val 3.365266 | tok/s 2627480.54 | norm 0.4267 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10656 | train 3.143746 | val 3.365266 | tok/s 2622351.96 | norm 0.4321 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10657 | train 3.041216 | val 3.365266 | tok/s 2629529.33 | norm 0.3952 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10658 | train 2.966455 | val 3.365266 | tok/s 2545551.32 | norm 0.4098 | dt 205.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10659 | train 3.060595 | val 3.365266 | tok/s 2612494.01 | norm 0.4080 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10660 | train 3.054453 | val 3.365266 | tok/s 2627301.42 | norm 0.4169 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10661 | train 2.974723 | val 3.365266 | tok/s 2630957.54 | norm 0.4166 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10662 | train 2.953227 | val 3.365266 | tok/s 2635175.97 | norm 0.4307 | dt 198.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10663 | train 2.923090 | val 3.365266 | tok/s 2628349.37 | norm 0.4400 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10664 | train 3.035568 | val 3.365266 | tok/s 2629085.97 | norm 0.4048 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10665 | train 2.922639 | val 3.365266 | tok/s 2629002.85 | norm 0.3674 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10666 | train 2.941128 | val 3.365266 | tok/s 2478037.51 | norm 0.3937 | dt 211.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10667 | train 3.030596 | val 3.365266 | tok/s 2627313.61 | norm 0.3945 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10668 | train 2.963854 | val 3.365266 | tok/s 2630885.34 | norm 0.4083 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10669 | train 3.119805 | val 3.365266 | tok/s 2629989.27 | norm 0.4994 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10670 | train 3.060904 | val 3.365266 | tok/s 2628919.38 | norm 0.3524 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10671 | train 3.032274 | val 3.365266 | tok/s 2626625.37 | norm 0.3428 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10672 | train 3.088209 | val 3.365266 | tok/s 2631503.67 | norm 0.3917 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10673 | train 2.965272 | val 3.365266 | tok/s 2629670.42 | norm 0.4422 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10674 | train 3.135566 | val 3.365266 | tok/s 2633537.50 | norm 0.4422 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10675 | train 2.912609 | val 3.365266 | tok/s 2628601.90 | norm 0.4260 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10676 | train 2.940895 | val 3.365266 | tok/s 2598687.04 | norm 0.4095 | dt 201.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10677 | train 3.060825 | val 3.365266 | tok/s 2619931.20 | norm 0.3831 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10678 | train 2.930274 | val 3.365266 | tok/s 2606875.09 | norm 0.3649 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10679 | train 3.111942 | val 3.365266 | tok/s 2631505.31 | norm 0.3434 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10680 | train 2.983244 | val 3.365266 | tok/s 2629183.10 | norm 0.3501 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10681 | train 2.964739 | val 3.365266 | tok/s 2630687.70 | norm 0.3916 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10682 | train 2.965303 | val 3.365266 | tok/s 2629017.42 | norm 0.3729 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10683 | train 2.948235 | val 3.365266 | tok/s 2615838.43 | norm 0.3430 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10684 | train 2.970700 | val 3.365266 | tok/s 2628765.77 | norm 0.3642 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10685 | train 3.064073 | val 3.365266 | tok/s 2631155.99 | norm 0.4276 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10686 | train 2.981065 | val 3.365266 | tok/s 2631756.64 | norm 0.3862 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10687 | train 2.900908 | val 3.365266 | tok/s 2628751.49 | norm 0.4062 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10688 | train 2.922221 | val 3.365266 | tok/s 2607229.79 | norm 0.3763 | dt 201.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10689 | train 3.049750 | val 3.365266 | tok/s 2620571.20 | norm 0.3550 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10690 | train 2.994634 | val 3.365266 | tok/s 2576798.45 | norm 0.3746 | dt 203.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10691 | train 3.057485 | val 3.365266 | tok/s 2534273.12 | norm 0.3974 | dt 206.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10692 | train 3.172096 | val 3.365266 | tok/s 2629023.98 | norm 0.3902 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10693 | train 3.173895 | val 3.365266 | tok/s 2626444.80 | norm 0.3953 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10694 | train 2.926999 | val 3.365266 | tok/s 2628996.97 | norm 0.3372 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10695 | train 3.032926 | val 3.365266 | tok/s 2626903.00 | norm 0.3670 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10696 | train 2.986104 | val 3.365266 | tok/s 2630653.03 | norm 0.4082 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10697 | train 2.958525 | val 3.365266 | tok/s 2631513.00 | norm 0.4638 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10698 | train 2.995854 | val 3.365266 | tok/s 2630677.18 | norm 0.4205 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10699 | train 2.978450 | val 3.365266 | tok/s 2625437.66 | norm 0.3796 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10700 | train 2.934378 | val 3.365266 | tok/s 2630294.74 | norm 0.4050 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10701 | train 3.086938 | val 3.365266 | tok/s 2628351.63 | norm 0.4392 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10702 | train 3.022448 | val 3.365266 | tok/s 2628212.38 | norm 0.4130 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10703 | train 3.029340 | val 3.365266 | tok/s 2631913.83 | norm 0.4218 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10704 | train 3.080263 | val 3.365266 | tok/s 2630513.22 | norm 0.3872 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10705 | train 3.087100 | val 3.365266 | tok/s 2627399.41 | norm 0.3737 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10706 | train 3.308519 | val 3.365266 | tok/s 2625582.17 | norm 0.6558 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10707 | train 2.984545 | val 3.365266 | tok/s 2629289.36 | norm 0.4311 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10708 | train 3.017136 | val 3.365266 | tok/s 2546044.53 | norm 0.3222 | dt 205.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10709 | train 2.992539 | val 3.365266 | tok/s 2615813.06 | norm 0.3715 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10710 | train 2.975117 | val 3.365266 | tok/s 2565068.69 | norm 0.3857 | dt 204.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10711 | train 3.001602 | val 3.365266 | tok/s 2629469.85 | norm 0.3974 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10712 | train 3.034966 | val 3.365266 | tok/s 2628423.34 | norm 0.4331 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10713 | train 3.160191 | val 3.365266 | tok/s 2629079.22 | norm 0.4104 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10714 | train 3.067665 | val 3.365266 | tok/s 2628978.19 | norm 0.3873 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10715 | train 3.049952 | val 3.365266 | tok/s 2631294.57 | norm 0.3854 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10716 | train 3.143496 | val 3.365266 | tok/s 2625481.88 | norm 0.3989 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10717 | train 2.949716 | val 3.365266 | tok/s 2619412.00 | norm 0.4054 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10718 | train 3.029412 | val 3.365266 | tok/s 2627822.68 | norm 0.4282 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10719 | train 3.009572 | val 3.365266 | tok/s 2632363.43 | norm 0.4134 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10720 | train 3.006105 | val 3.365266 | tok/s 2630887.79 | norm 0.3791 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10721 | train 3.092490 | val 3.365266 | tok/s 2625950.21 | norm 0.3759 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10722 | train 3.282576 | val 3.365266 | tok/s 2611173.71 | norm 0.4033 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10723 | train 3.258314 | val 3.365266 | tok/s 2629037.33 | norm 0.3942 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10724 | train 2.916509 | val 3.365266 | tok/s 2627115.62 | norm 0.4218 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10725 | train 3.066849 | val 3.365266 | tok/s 2612398.03 | norm 0.4165 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10726 | train 3.005198 | val 3.365266 | tok/s 2629486.35 | norm 0.4122 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10727 | train 3.041263 | val 3.365266 | tok/s 2629852.42 | norm 0.4075 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10728 | train 3.051275 | val 3.365266 | tok/s 2628258.08 | norm 0.3936 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10729 | train 2.917898 | val 3.365266 | tok/s 2630742.29 | norm 0.3966 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10730 | train 3.014355 | val 3.365266 | tok/s 2627749.65 | norm 0.3470 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10731 | train 3.101384 | val 3.365266 | tok/s 2629074.54 | norm 0.3970 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10732 | train 3.189354 | val 3.365266 | tok/s 2633015.38 | norm 0.4094 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10733 | train 2.908274 | val 3.365266 | tok/s 2628883.46 | norm 0.3610 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10734 | train 3.067603 | val 3.365266 | tok/s 2631073.37 | norm 0.3945 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10735 | train 2.956984 | val 3.365266 | tok/s 2632685.13 | norm 0.3900 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10736 | train 2.932956 | val 3.365266 | tok/s 2631438.60 | norm 0.4327 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10737 | train 3.120041 | val 3.365266 | tok/s 2632715.72 | norm 0.4106 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10738 | train 3.014481 | val 3.365266 | tok/s 2629499.01 | norm 0.3792 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10739 | train 3.054650 | val 3.365266 | tok/s 2631403.92 | norm 0.4163 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10740 | train 2.988863 | val 3.365266 | tok/s 2625956.22 | norm 0.4155 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10741 | train 3.044766 | val 3.365266 | tok/s 2630018.36 | norm 0.4611 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10742 | train 3.120622 | val 3.365266 | tok/s 2631131.80 | norm 0.4233 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10743 | train 2.978892 | val 3.365266 | tok/s 2629671.98 | norm 0.3821 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10744 | train 3.135563 | val 3.365266 | tok/s 2629611.58 | norm 0.3953 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10745 | train 3.186613 | val 3.365266 | tok/s 2626758.94 | norm 0.4223 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10746 | train 3.215889 | val 3.365266 | tok/s 2627822.60 | norm 0.4527 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10747 | train 3.040986 | val 3.365266 | tok/s 2627309.19 | norm 0.3709 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10748 | train 2.930222 | val 3.365266 | tok/s 2627897.37 | norm 0.4142 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10749 | train 2.983224 | val 3.365266 | tok/s 2630784.44 | norm 0.4266 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 10750 | val_loss 3.362560 step 10750 | train 3.022764 | val 3.362560 | tok/s 2609344.44 | norm 0.4022 | dt 200.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10751 | train 3.032024 | val 3.362560 | tok/s 2630159.14 | norm 0.3729 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10752 | train 3.089385 | val 3.362560 | tok/s 2625092.13 | norm 0.4468 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10753 | train 2.961493 | val 3.362560 | tok/s 2624980.17 | norm 0.4728 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10754 | train 3.043360 | val 3.362560 | tok/s 2622505.14 | norm 0.3699 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10755 | train 2.968186 | val 3.362560 | tok/s 2622296.35 | norm 0.4026 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10756 | train 2.923788 | val 3.362560 | tok/s 2623981.09 | norm 0.3980 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10757 | train 2.935616 | val 3.362560 | tok/s 2631859.41 | norm 0.4186 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10758 | train 2.959552 | val 3.362560 | tok/s 2619908.19 | norm 0.3771 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10759 | train 3.036597 | val 3.362560 | tok/s 2620841.74 | norm 0.3502 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10760 | train 2.942202 | val 3.362560 | tok/s 2623996.90 | norm 0.3893 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10761 | train 2.953872 | val 3.362560 | tok/s 2629426.07 | norm 0.3993 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10762 | train 3.119713 | val 3.362560 | tok/s 2627236.54 | norm 0.3743 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10763 | train 3.122549 | val 3.362560 | tok/s 2634504.14 | norm 0.4301 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10764 | train 2.974447 | val 3.362560 | tok/s 2629699.90 | norm 0.4121 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10765 | train 2.993947 | val 3.362560 | tok/s 2629016.01 | norm 0.3651 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10766 | train 3.045524 | val 3.362560 | tok/s 2630287.67 | norm 0.3775 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10767 | train 3.068483 | val 3.362560 | tok/s 2630778.14 | norm 0.3791 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10768 | train 2.973598 | val 3.362560 | tok/s 2625103.13 | norm 0.4268 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10769 | train 3.058742 | val 3.362560 | tok/s 2581911.71 | norm 0.3963 | dt 203.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10770 | train 3.026667 | val 3.362560 | tok/s 2543289.64 | norm 0.3744 | dt 206.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10771 | train 3.027862 | val 3.362560 | tok/s 2632712.88 | norm 0.4027 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10772 | train 2.991247 | val 3.362560 | tok/s 2628964.99 | norm 0.4146 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10773 | train 3.021197 | val 3.362560 | tok/s 2629128.61 | norm 0.4093 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10774 | train 2.979589 | val 3.362560 | tok/s 2629798.30 | norm 0.4137 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10775 | train 2.953552 | val 3.362560 | tok/s 2499663.54 | norm 0.4169 | dt 209.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10776 | train 3.028392 | val 3.362560 | tok/s 2500593.80 | norm 0.4157 | dt 209.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10777 | train 2.878759 | val 3.362560 | tok/s 2631583.93 | norm 0.4250 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10778 | train 2.912596 | val 3.362560 | tok/s 2634075.16 | norm 0.3998 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10779 | train 3.056974 | val 3.362560 | tok/s 2633318.46 | norm 0.3535 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10780 | train 2.994231 | val 3.362560 | tok/s 2630885.18 | norm 0.3882 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10781 | train 3.002571 | val 3.362560 | tok/s 2624937.22 | norm 0.4287 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10782 | train 3.002330 | val 3.362560 | tok/s 2630063.66 | norm 0.4044 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10783 | train 2.902249 | val 3.362560 | tok/s 2617012.47 | norm 0.3664 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10784 | train 3.066207 | val 3.362560 | tok/s 2628485.91 | norm 0.4271 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10785 | train 3.065617 | val 3.362560 | tok/s 2623360.26 | norm 0.4026 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10786 | train 2.975110 | val 3.362560 | tok/s 2634962.23 | norm 0.4043 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10787 | train 3.115461 | val 3.362560 | tok/s 2634404.88 | norm 0.4114 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10788 | train 2.976699 | val 3.362560 | tok/s 2633698.11 | norm 0.4160 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10789 | train 3.026784 | val 3.362560 | tok/s 2630756.96 | norm 0.4319 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10790 | train 3.048427 | val 3.362560 | tok/s 2628627.55 | norm 0.4181 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10791 | train 2.994103 | val 3.362560 | tok/s 2622738.12 | norm 0.3847 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10792 | train 3.058228 | val 3.362560 | tok/s 2626886.68 | norm 0.3882 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10793 | train 3.067736 | val 3.362560 | tok/s 2621546.64 | norm 0.3578 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10794 | train 2.888263 | val 3.362560 | tok/s 2630088.82 | norm 0.3644 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10795 | train 2.928513 | val 3.362560 | tok/s 2623907.55 | norm 0.3793 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10796 | train 3.019450 | val 3.362560 | tok/s 2624306.50 | norm 0.3863 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10797 | train 3.002973 | val 3.362560 | tok/s 2629415.37 | norm 0.3748 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10798 | train 2.968951 | val 3.362560 | tok/s 2628882.54 | norm 0.4087 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10799 | train 3.034165 | val 3.362560 | tok/s 2627157.09 | norm 0.4327 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10800 | train 3.066165 | val 3.362560 | tok/s 2628464.95 | norm 0.4459 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10801 | train 3.046166 | val 3.362560 | tok/s 2631922.69 | norm 0.4398 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10802 | train 2.924357 | val 3.362560 | tok/s 2634134.47 | norm 0.3854 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10803 | train 3.101182 | val 3.362560 | tok/s 2632503.05 | norm 0.4340 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10804 | train 3.077056 | val 3.362560 | tok/s 2629104.87 | norm 0.4042 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10805 | train 2.914404 | val 3.362560 | tok/s 2632004.74 | norm 0.3631 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10806 | train 3.062745 | val 3.362560 | tok/s 2630590.10 | norm 0.3500 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10807 | train 2.963975 | val 3.362560 | tok/s 2631436.91 | norm 0.3207 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10808 | train 3.089483 | val 3.362560 | tok/s 2628457.75 | norm 0.3848 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10809 | train 3.017322 | val 3.362560 | tok/s 2629663.11 | norm 0.3766 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10810 | train 2.862944 | val 3.362560 | tok/s 2633205.80 | norm 0.4666 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10811 | train 2.953939 | val 3.362560 | tok/s 2628084.01 | norm 0.3691 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10812 | train 3.076189 | val 3.362560 | tok/s 2632704.14 | norm 0.4177 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10813 | train 2.845633 | val 3.362560 | tok/s 2628760.78 | norm 0.3980 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10814 | train 2.922159 | val 3.362560 | tok/s 2632445.09 | norm 0.5150 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10815 | train 2.969455 | val 3.362560 | tok/s 2622694.12 | norm 0.3718 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10816 | train 2.993097 | val 3.362560 | tok/s 2628653.56 | norm 0.3577 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10817 | train 3.047863 | val 3.362560 | tok/s 2629884.77 | norm 0.3837 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10818 | train 2.936031 | val 3.362560 | tok/s 2630335.23 | norm 0.3758 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10819 | train 2.980760 | val 3.362560 | tok/s 2630429.78 | norm 0.3865 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10820 | train 3.018239 | val 3.362560 | tok/s 2626368.30 | norm 0.3994 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10821 | train 3.050690 | val 3.362560 | tok/s 2630332.83 | norm 0.4104 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10822 | train 3.035643 | val 3.362560 | tok/s 2629275.25 | norm 0.3997 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10823 | train 3.087469 | val 3.362560 | tok/s 2625259.73 | norm 0.3919 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10824 | train 2.969187 | val 3.362560 | tok/s 2626071.54 | norm 0.3699 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10825 | train 2.964411 | val 3.362560 | tok/s 2554934.87 | norm 0.3979 | dt 205.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10826 | train 3.041183 | val 3.362560 | tok/s 2632654.03 | norm 0.4309 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10827 | train 3.083000 | val 3.362560 | tok/s 2629313.66 | norm 0.4552 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10828 | train 2.903816 | val 3.362560 | tok/s 2626074.61 | norm 0.4379 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10829 | train 2.984666 | val 3.362560 | tok/s 2630873.17 | norm 0.4208 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10830 | train 2.986384 | val 3.362560 | tok/s 2629651.97 | norm 0.3848 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10831 | train 2.981202 | val 3.362560 | tok/s 2627718.28 | norm 0.4078 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10832 | train 3.029941 | val 3.362560 | tok/s 2500295.17 | norm 0.4188 | dt 209.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10833 | train 3.104730 | val 3.362560 | tok/s 2485977.83 | norm 0.4301 | dt 210.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10834 | train 3.127371 | val 3.362560 | tok/s 2629895.05 | norm 0.4125 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10835 | train 2.967597 | val 3.362560 | tok/s 2626393.39 | norm 0.3742 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10836 | train 2.986605 | val 3.362560 | tok/s 2629644.78 | norm 0.3523 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10837 | train 2.997095 | val 3.362560 | tok/s 2631785.81 | norm 0.4034 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10838 | train 3.122449 | val 3.362560 | tok/s 2628079.16 | norm 0.4738 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10839 | train 3.027543 | val 3.362560 | tok/s 2627276.98 | norm 0.4096 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10840 | train 2.870414 | val 3.362560 | tok/s 2631410.19 | norm 0.3888 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10841 | train 3.096022 | val 3.362560 | tok/s 2633231.47 | norm 0.4225 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10842 | train 2.956395 | val 3.362560 | tok/s 2621702.14 | norm 0.3617 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10843 | train 3.020566 | val 3.362560 | tok/s 2623517.68 | norm 0.3671 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10844 | train 3.037844 | val 3.362560 | tok/s 2626318.72 | norm 0.3782 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10845 | train 3.074961 | val 3.362560 | tok/s 2625047.92 | norm 0.4036 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10846 | train 2.953751 | val 3.362560 | tok/s 2629265.30 | norm 0.4259 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10847 | train 3.062856 | val 3.362560 | tok/s 2630115.45 | norm 0.4174 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10848 | train 2.944869 | val 3.362560 | tok/s 2633109.04 | norm 0.4422 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10849 | train 2.920500 | val 3.362560 | tok/s 2627736.87 | norm 0.4152 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10850 | train 3.043616 | val 3.362560 | tok/s 2622784.48 | norm 0.3960 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10851 | train 3.138821 | val 3.362560 | tok/s 2631526.11 | norm 0.4156 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10852 | train 3.165118 | val 3.362560 | tok/s 2622855.88 | norm 0.3785 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10853 | train 3.087254 | val 3.362560 | tok/s 2625773.06 | norm 0.4195 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10854 | train 3.035913 | val 3.362560 | tok/s 2632707.82 | norm 0.4365 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10855 | train 3.083093 | val 3.362560 | tok/s 2624520.83 | norm 0.4299 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10856 | train 2.932104 | val 3.362560 | tok/s 2629408.94 | norm 0.3727 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10857 | train 3.110127 | val 3.362560 | tok/s 2628269.64 | norm 0.3553 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10858 | train 2.939693 | val 3.362560 | tok/s 2630425.34 | norm 0.3886 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10859 | train 2.906409 | val 3.362560 | tok/s 2630808.08 | norm 0.4043 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10860 | train 3.082904 | val 3.362560 | tok/s 2629135.58 | norm 0.3623 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10861 | train 3.064251 | val 3.362560 | tok/s 2629018.57 | norm 0.4184 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10862 | train 2.974813 | val 3.362560 | tok/s 2627425.34 | norm 0.3894 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10863 | train 2.922885 | val 3.362560 | tok/s 2629315.47 | norm 0.3329 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10864 | train 2.956630 | val 3.362560 | tok/s 2627061.69 | norm 0.3641 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10865 | train 3.033792 | val 3.362560 | tok/s 2628726.62 | norm 0.3760 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10866 | train 2.948264 | val 3.362560 | tok/s 2628054.89 | norm 0.3546 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10867 | train 3.027899 | val 3.362560 | tok/s 2630920.03 | norm 0.3274 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10868 | train 2.982787 | val 3.362560 | tok/s 2629435.98 | norm 0.3770 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10869 | train 2.998989 | val 3.362560 | tok/s 2625637.91 | norm 0.4114 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10870 | train 2.964211 | val 3.362560 | tok/s 2626512.60 | norm 0.3765 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10871 | train 3.057027 | val 3.362560 | tok/s 2628392.95 | norm 0.3691 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10872 | train 3.004078 | val 3.362560 | tok/s 2628867.44 | norm 0.3807 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10873 | train 2.981744 | val 3.362560 | tok/s 2624334.34 | norm 0.3451 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10874 | train 3.056492 | val 3.362560 | tok/s 2630711.18 | norm 0.3821 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10875 | train 3.103722 | val 3.362560 | tok/s 2627140.98 | norm 0.3999 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10876 | train 3.009935 | val 3.362560 | tok/s 2630178.62 | norm 0.4027 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10877 | train 3.146244 | val 3.362560 | tok/s 2632892.20 | norm 0.3968 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10878 | train 2.962409 | val 3.362560 | tok/s 2633930.07 | norm 0.3855 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10879 | train 2.969123 | val 3.362560 | tok/s 2633556.27 | norm 0.3915 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10880 | train 2.993847 | val 3.362560 | tok/s 2630907.56 | norm 0.3959 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10881 | train 3.053060 | val 3.362560 | tok/s 2629186.15 | norm 0.3473 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10882 | train 2.932546 | val 3.362560 | tok/s 2634257.85 | norm 0.4119 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10883 | train 2.974622 | val 3.362560 | tok/s 2539582.49 | norm 0.4130 | dt 206.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10884 | train 3.081721 | val 3.362560 | tok/s 2472266.46 | norm 0.4116 | dt 212.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10885 | train 2.883452 | val 3.362560 | tok/s 2550757.32 | norm 0.4038 | dt 205.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10886 | train 3.128136 | val 3.362560 | tok/s 2631943.51 | norm 0.4023 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10887 | train 3.104965 | val 3.362560 | tok/s 2622751.03 | norm 0.4108 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10888 | train 2.931198 | val 3.362560 | tok/s 2628691.56 | norm 0.4088 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10889 | train 2.925067 | val 3.362560 | tok/s 2631452.85 | norm 0.3912 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10890 | train 3.108541 | val 3.362560 | tok/s 2630400.18 | norm 0.3683 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10891 | train 3.011825 | val 3.362560 | tok/s 2626798.96 | norm 0.3838 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10892 | train 3.122349 | val 3.362560 | tok/s 2626718.06 | norm 0.4285 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10893 | train 2.958374 | val 3.362560 | tok/s 2629070.11 | norm 0.4284 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10894 | train 3.100542 | val 3.362560 | tok/s 2628835.30 | norm 0.4933 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10895 | train 3.003510 | val 3.362560 | tok/s 2624188.56 | norm 0.4435 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10896 | train 3.010859 | val 3.362560 | tok/s 2628182.48 | norm 0.4127 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10897 | train 3.033682 | val 3.362560 | tok/s 2629760.34 | norm 0.4160 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10898 | train 3.103363 | val 3.362560 | tok/s 2632044.89 | norm 0.4952 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10899 | train 3.005998 | val 3.362560 | tok/s 2626080.32 | norm 0.3810 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10900 | train 3.069615 | val 3.362560 | tok/s 2629881.22 | norm 0.3340 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10901 | train 3.038611 | val 3.362560 | tok/s 2630793.84 | norm 0.3893 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10902 | train 2.943415 | val 3.362560 | tok/s 2631280.50 | norm 0.4070 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10903 | train 2.976540 | val 3.362560 | tok/s 2628670.84 | norm 0.4128 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10904 | train 2.994643 | val 3.362560 | tok/s 2631283.34 | norm 0.3870 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10905 | train 3.005708 | val 3.362560 | tok/s 2631604.29 | norm 0.3749 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10906 | train 2.957739 | val 3.362560 | tok/s 2626308.68 | norm 0.4080 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10907 | train 2.956524 | val 3.362560 | tok/s 2626877.00 | norm 0.3916 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10908 | train 3.014657 | val 3.362560 | tok/s 2629056.57 | norm 0.3775 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10909 | train 3.009177 | val 3.362560 | tok/s 2628923.77 | norm 0.4044 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10910 | train 2.983971 | val 3.362560 | tok/s 2630752.05 | norm 0.3883 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10911 | train 3.000708 | val 3.362560 | tok/s 2632461.19 | norm 0.4042 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10912 | train 2.977357 | val 3.362560 | tok/s 2628835.72 | norm 0.4011 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10913 | train 2.912508 | val 3.362560 | tok/s 2629934.66 | norm 0.3892 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10914 | train 2.921183 | val 3.362560 | tok/s 2629961.23 | norm 0.4089 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10915 | train 3.174785 | val 3.362560 | tok/s 2630984.07 | norm 0.4364 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10916 | train 3.054470 | val 3.362560 | tok/s 2630852.20 | norm 0.4046 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10917 | train 3.076655 | val 3.362560 | tok/s 2627592.10 | norm 0.3927 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10918 | train 2.895138 | val 3.362560 | tok/s 2630285.28 | norm 0.4178 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10919 | train 3.063367 | val 3.362560 | tok/s 2632895.17 | norm 0.4048 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10920 | train 2.835034 | val 3.362560 | tok/s 2632217.24 | norm 0.3891 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10921 | train 2.928666 | val 3.362560 | tok/s 2625195.01 | norm 0.4030 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10922 | train 2.889874 | val 3.362560 | tok/s 2623811.02 | norm 0.4207 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10923 | train 2.845086 | val 3.362560 | tok/s 2625936.10 | norm 0.4010 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10924 | train 2.978294 | val 3.362560 | tok/s 2625870.87 | norm 0.3879 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10925 | train 2.985481 | val 3.362560 | tok/s 2625350.64 | norm 0.4028 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10926 | train 3.064360 | val 3.362560 | tok/s 2627109.08 | norm 0.3709 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10927 | train 3.088053 | val 3.362560 | tok/s 2623577.93 | norm 0.3547 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10928 | train 2.931446 | val 3.362560 | tok/s 2628783.08 | norm 0.3677 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10929 | train 2.937260 | val 3.362560 | tok/s 2626406.73 | norm 0.3682 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10930 | train 2.955129 | val 3.362560 | tok/s 2625365.91 | norm 0.4152 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10931 | train 3.059804 | val 3.362560 | tok/s 2627916.83 | norm 0.4106 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10932 | train 2.989186 | val 3.362560 | tok/s 2626352.83 | norm 0.4313 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10933 | train 3.047221 | val 3.362560 | tok/s 2597113.08 | norm 0.4037 | dt 201.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10934 | train 2.975669 | val 3.362560 | tok/s 2627830.46 | norm 0.3950 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10935 | train 3.057897 | val 3.362560 | tok/s 2616551.37 | norm 0.4553 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10936 | train 3.032444 | val 3.362560 | tok/s 2629722.89 | norm 0.3755 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10937 | train 3.034342 | val 3.362560 | tok/s 2601038.21 | norm 0.3533 | dt 201.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10938 | train 2.942240 | val 3.362560 | tok/s 2585721.41 | norm 0.3939 | dt 202.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10939 | train 2.998102 | val 3.362560 | tok/s 2623761.54 | norm 0.3832 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10940 | train 3.035245 | val 3.362560 | tok/s 2626761.66 | norm 0.5567 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10941 | train 2.998413 | val 3.362560 | tok/s 2628575.39 | norm 0.4537 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10942 | train 2.922335 | val 3.362560 | tok/s 2622557.18 | norm 0.4083 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10943 | train 2.878387 | val 3.362560 | tok/s 2629196.02 | norm 0.3826 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10944 | train 2.914565 | val 3.362560 | tok/s 2629966.85 | norm 0.3913 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10945 | train 2.948377 | val 3.362560 | tok/s 2628284.26 | norm 0.3969 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10946 | train 3.087017 | val 3.362560 | tok/s 2625632.32 | norm 0.3838 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10947 | train 2.940445 | val 3.362560 | tok/s 2620391.87 | norm 0.4072 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10948 | train 2.945271 | val 3.362560 | tok/s 2628129.25 | norm 0.3996 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10949 | train 3.058696 | val 3.362560 | tok/s 2626584.39 | norm 0.3956 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10950 | train 2.890386 | val 3.362560 | tok/s 2521733.54 | norm 0.4172 | dt 207.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10951 | train 3.073113 | val 3.362560 | tok/s 2629616.96 | norm 0.4371 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10952 | train 3.314371 | val 3.362560 | tok/s 2632317.33 | norm 0.4531 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10953 | train 3.120662 | val 3.362560 | tok/s 2629940.23 | norm 0.4132 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10954 | train 3.031081 | val 3.362560 | tok/s 2631423.82 | norm 0.3755 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10955 | train 3.086262 | val 3.362560 | tok/s 2630030.22 | norm 0.4265 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10956 | train 2.978060 | val 3.362560 | tok/s 2621642.61 | norm 0.3656 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10957 | train 3.054962 | val 3.362560 | tok/s 2602879.48 | norm 0.3866 | dt 201.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10958 | train 3.020887 | val 3.362560 | tok/s 2616550.32 | norm 0.4910 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10959 | train 3.059561 | val 3.362560 | tok/s 2625690.60 | norm 0.4009 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10960 | train 3.026324 | val 3.362560 | tok/s 2629179.56 | norm 0.3538 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10961 | train 3.002863 | val 3.362560 | tok/s 2630611.32 | norm 0.3745 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10962 | train 3.065453 | val 3.362560 | tok/s 2627959.56 | norm 0.3698 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10963 | train 2.959575 | val 3.362560 | tok/s 2629809.47 | norm 0.4236 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10964 | train 3.015417 | val 3.362560 | tok/s 2629895.44 | norm 0.4018 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10965 | train 2.980211 | val 3.362560 | tok/s 2629660.13 | norm 0.4362 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10966 | train 2.945082 | val 3.362560 | tok/s 2623168.96 | norm 0.4272 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10967 | train 3.037422 | val 3.362560 | tok/s 2624992.46 | norm 0.3877 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10968 | train 2.994863 | val 3.362560 | tok/s 2628667.51 | norm 0.3721 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10969 | train 2.934061 | val 3.362560 | tok/s 2630676.46 | norm 0.3765 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10970 | train 3.023993 | val 3.362560 | tok/s 2633600.02 | norm 0.3630 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10971 | train 3.015682 | val 3.362560 | tok/s 2625693.01 | norm 0.3798 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10972 | train 3.033378 | val 3.362560 | tok/s 2627419.35 | norm 0.3792 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10973 | train 2.905454 | val 3.362560 | tok/s 2628944.92 | norm 0.3601 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10974 | train 2.968622 | val 3.362560 | tok/s 2628165.46 | norm 0.4061 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10975 | train 2.932214 | val 3.362560 | tok/s 2630765.23 | norm 0.4263 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10976 | train 2.910457 | val 3.362560 | tok/s 2629563.91 | norm 0.4145 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10977 | train 3.087554 | val 3.362560 | tok/s 2629884.35 | norm 0.3804 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10978 | train 3.002177 | val 3.362560 | tok/s 2630590.64 | norm 0.3703 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10979 | train 3.061504 | val 3.362560 | tok/s 2632039.84 | norm 0.3809 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10980 | train 3.003894 | val 3.362560 | tok/s 2633338.83 | norm 0.3763 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10981 | train 2.983289 | val 3.362560 | tok/s 2632370.35 | norm 0.3967 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10982 | train 2.961887 | val 3.362560 | tok/s 2629504.47 | norm 0.4173 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10983 | train 3.000153 | val 3.362560 | tok/s 2632045.58 | norm 0.4132 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10984 | train 2.996161 | val 3.362560 | tok/s 2630062.09 | norm 0.4119 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10985 | train 2.925455 | val 3.362560 | tok/s 2630552.60 | norm 0.4023 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10986 | train 2.936122 | val 3.362560 | tok/s 2630276.10 | norm 0.4094 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10987 | train 3.086517 | val 3.362560 | tok/s 2631265.78 | norm 0.4095 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10988 | train 3.006472 | val 3.362560 | tok/s 2627566.14 | norm 0.3562 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10989 | train 2.940632 | val 3.362560 | tok/s 2629727.40 | norm 0.3769 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10990 | train 2.998348 | val 3.362560 | tok/s 2630121.31 | norm 0.4219 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10991 | train 2.980490 | val 3.362560 | tok/s 2629011.46 | norm 0.3658 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10992 | train 3.037276 | val 3.362560 | tok/s 2630866.96 | norm 0.3617 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10993 | train 3.251311 | val 3.362560 | tok/s 2626567.12 | norm 0.4379 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10994 | train 2.990909 | val 3.362560 | tok/s 2632502.10 | norm 0.4101 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10995 | train 2.864375 | val 3.362560 | tok/s 2633984.17 | norm 0.4050 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10996 | train 2.996384 | val 3.362560 | tok/s 2632586.98 | norm 0.4268 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10997 | train 2.945904 | val 3.362560 | tok/s 2633081.68 | norm 0.3946 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10998 | train 3.116216 | val 3.362560 | tok/s 2634724.88 | norm 0.4484 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 10999 | train 3.031147 | val 3.362560 | tok/s 2625066.53 | norm 0.4349 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 11000 | val_loss 3.355891 step 11000 | train 2.972484 | val 3.355891 | tok/s 2613431.75 | norm 0.3949 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11001 | train 3.041126 | val 3.355891 | tok/s 2625552.05 | norm 0.4052 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11002 | train 2.972958 | val 3.355891 | tok/s 2629481.90 | norm 0.3961 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11003 | train 3.023425 | val 3.355891 | tok/s 2629048.90 | norm 0.4331 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11004 | train 2.948133 | val 3.355891 | tok/s 2631984.08 | norm 0.4204 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11005 | train 3.266044 | val 3.355891 | tok/s 2632587.72 | norm 0.4512 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11006 | train 3.082592 | val 3.355891 | tok/s 2632344.99 | norm 0.4444 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11007 | train 2.886532 | val 3.355891 | tok/s 2597792.75 | norm 0.3910 | dt 201.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11008 | train 3.177021 | val 3.355891 | tok/s 2629753.18 | norm 0.4226 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11009 | train 3.038218 | val 3.355891 | tok/s 2631864.18 | norm 0.4253 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11010 | train 3.005819 | val 3.355891 | tok/s 2631311.11 | norm 0.3795 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11011 | train 2.951104 | val 3.355891 | tok/s 2616230.61 | norm 0.4299 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11012 | train 2.926090 | val 3.355891 | tok/s 2623581.22 | norm 0.4382 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11013 | train 2.898870 | val 3.355891 | tok/s 2626101.20 | norm 0.4415 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11014 | train 3.190644 | val 3.355891 | tok/s 2626562.43 | norm 0.4003 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11015 | train 2.981608 | val 3.355891 | tok/s 2619303.36 | norm 0.3683 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11016 | train 2.902489 | val 3.355891 | tok/s 2628061.26 | norm 0.3929 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11017 | train 2.871371 | val 3.355891 | tok/s 2629163.93 | norm 0.4130 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11018 | train 2.976018 | val 3.355891 | tok/s 2626943.71 | norm 0.3753 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11019 | train 3.015681 | val 3.355891 | tok/s 2627819.24 | norm 0.3592 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11020 | train 2.945470 | val 3.355891 | tok/s 2627736.07 | norm 0.3565 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11021 | train 3.017783 | val 3.355891 | tok/s 2624303.25 | norm 0.4112 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11022 | train 3.009028 | val 3.355891 | tok/s 2556363.20 | norm 0.3858 | dt 205.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11023 | train 3.128859 | val 3.355891 | tok/s 2492409.14 | norm 0.3912 | dt 210.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11024 | train 3.111480 | val 3.355891 | tok/s 2619788.93 | norm 0.3853 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11025 | train 3.033620 | val 3.355891 | tok/s 2628050.73 | norm 0.3794 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11026 | train 2.919094 | val 3.355891 | tok/s 2626936.96 | norm 0.3548 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11027 | train 2.917721 | val 3.355891 | tok/s 2627584.55 | norm 0.3906 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11028 | train 2.950859 | val 3.355891 | tok/s 2628783.69 | norm 0.4233 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11029 | train 2.966529 | val 3.355891 | tok/s 2625896.80 | norm 0.4369 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11030 | train 2.964547 | val 3.355891 | tok/s 2628677.30 | norm 0.4532 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11031 | train 2.839518 | val 3.355891 | tok/s 2626689.04 | norm 0.4173 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11032 | train 2.971115 | val 3.355891 | tok/s 2623801.11 | norm 0.3960 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11033 | train 2.942743 | val 3.355891 | tok/s 2624022.29 | norm 0.3772 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11034 | train 3.107595 | val 3.355891 | tok/s 2629657.22 | norm 0.4257 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11035 | train 3.034404 | val 3.355891 | tok/s 2631393.23 | norm 0.4259 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11036 | train 3.302397 | val 3.355891 | tok/s 2627364.19 | norm 0.4527 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11037 | train 3.004285 | val 3.355891 | tok/s 2627416.47 | norm 0.4148 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11038 | train 3.051643 | val 3.355891 | tok/s 2625956.13 | norm 0.3703 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11039 | train 3.263368 | val 3.355891 | tok/s 2628151.52 | norm 0.5542 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11040 | train 2.943037 | val 3.355891 | tok/s 2628045.82 | norm 0.3821 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11041 | train 3.096682 | val 3.355891 | tok/s 2627492.84 | norm 0.4274 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11042 | train 2.912168 | val 3.355891 | tok/s 2619823.03 | norm 0.4037 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11043 | train 3.107375 | val 3.355891 | tok/s 2625210.18 | norm 0.3999 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11044 | train 2.962216 | val 3.355891 | tok/s 2626916.95 | norm 0.3934 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11045 | train 3.016798 | val 3.355891 | tok/s 2623686.43 | norm 0.3632 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11046 | train 2.995154 | val 3.355891 | tok/s 2625051.52 | norm 0.3889 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11047 | train 2.853642 | val 3.355891 | tok/s 2626879.30 | norm 0.4434 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11048 | train 3.058156 | val 3.355891 | tok/s 2625957.04 | norm 0.4110 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11049 | train 3.108111 | val 3.355891 | tok/s 2625129.72 | norm 0.4234 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11050 | train 3.025171 | val 3.355891 | tok/s 2628059.73 | norm 0.4571 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11051 | train 2.983286 | val 3.355891 | tok/s 2631365.89 | norm 0.4077 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11052 | train 2.985660 | val 3.355891 | tok/s 2623314.21 | norm 0.3949 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11053 | train 2.973916 | val 3.355891 | tok/s 2627593.89 | norm 0.4085 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11054 | train 2.976703 | val 3.355891 | tok/s 2622940.64 | norm 0.4777 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11055 | train 3.049103 | val 3.355891 | tok/s 2622510.95 | norm 0.3934 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11056 | train 3.000810 | val 3.355891 | tok/s 2630223.50 | norm 0.3443 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11057 | train 3.093569 | val 3.355891 | tok/s 2627421.48 | norm 0.3818 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11058 | train 3.044785 | val 3.355891 | tok/s 2630508.59 | norm 0.4094 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11059 | train 3.003319 | val 3.355891 | tok/s 2630075.33 | norm 0.3605 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11060 | train 3.005935 | val 3.355891 | tok/s 2630819.73 | norm 0.3796 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11061 | train 3.012400 | val 3.355891 | tok/s 2630810.51 | norm 0.3736 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11062 | train 3.086079 | val 3.355891 | tok/s 2623877.21 | norm 0.3828 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11063 | train 2.927136 | val 3.355891 | tok/s 2500712.79 | norm 0.4189 | dt 209.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11064 | train 2.942744 | val 3.355891 | tok/s 2581650.43 | norm 0.4035 | dt 203.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11065 | train 2.869593 | val 3.355891 | tok/s 2619677.17 | norm 0.4101 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11066 | train 2.975344 | val 3.355891 | tok/s 2620412.85 | norm 0.4037 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11067 | train 3.001737 | val 3.355891 | tok/s 2623282.29 | norm 0.3715 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11068 | train 3.040530 | val 3.355891 | tok/s 2624669.46 | norm 0.3450 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11069 | train 2.982325 | val 3.355891 | tok/s 2624993.91 | norm 0.4077 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11070 | train 3.064615 | val 3.355891 | tok/s 2616528.50 | norm 0.4058 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11071 | train 3.030563 | val 3.355891 | tok/s 2608550.71 | norm 0.3958 | dt 200.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11072 | train 3.136721 | val 3.355891 | tok/s 2623273.58 | norm 0.4106 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11073 | train 2.936545 | val 3.355891 | tok/s 2530551.80 | norm 0.3692 | dt 207.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11074 | train 3.014419 | val 3.355891 | tok/s 2624503.08 | norm 0.3632 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11075 | train 3.042207 | val 3.355891 | tok/s 2623413.12 | norm 0.4353 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11076 | train 2.986180 | val 3.355891 | tok/s 2624439.43 | norm 0.3900 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11077 | train 2.883063 | val 3.355891 | tok/s 2627991.45 | norm 0.3786 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11078 | train 2.962904 | val 3.355891 | tok/s 2623657.00 | norm 0.3965 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11079 | train 2.913834 | val 3.355891 | tok/s 2621132.86 | norm 0.3936 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11080 | train 2.996895 | val 3.355891 | tok/s 2619796.03 | norm 0.3935 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11081 | train 2.947509 | val 3.355891 | tok/s 2628468.71 | norm 0.3949 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11082 | train 3.011749 | val 3.355891 | tok/s 2630829.18 | norm 0.3979 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11083 | train 2.954732 | val 3.355891 | tok/s 2616878.82 | norm 0.4091 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11084 | train 3.060424 | val 3.355891 | tok/s 2629047.35 | norm 0.4052 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11085 | train 2.902631 | val 3.355891 | tok/s 2626772.66 | norm 0.3868 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11086 | train 2.941299 | val 3.355891 | tok/s 2627834.63 | norm 0.3999 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11087 | train 2.983073 | val 3.355891 | tok/s 2627114.51 | norm 0.7662 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11088 | train 3.085545 | val 3.355891 | tok/s 2625936.40 | norm 0.3470 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11089 | train 3.010781 | val 3.355891 | tok/s 2623227.78 | norm 0.3161 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11090 | train 3.069988 | val 3.355891 | tok/s 2626188.49 | norm 0.3697 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11091 | train 3.040317 | val 3.355891 | tok/s 2627839.96 | norm 0.3936 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11092 | train 3.225970 | val 3.355891 | tok/s 2627095.02 | norm 0.4187 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11093 | train 3.001938 | val 3.355891 | tok/s 2623541.23 | norm 0.4699 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11094 | train 2.900978 | val 3.355891 | tok/s 2623229.97 | norm 0.4273 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11095 | train 2.979222 | val 3.355891 | tok/s 2625248.82 | norm 0.4262 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11096 | train 2.946085 | val 3.355891 | tok/s 2627304.83 | norm 0.4281 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11097 | train 2.978000 | val 3.355891 | tok/s 2625301.49 | norm 0.4025 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11098 | train 2.900417 | val 3.355891 | tok/s 2626774.70 | norm 0.3940 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11099 | train 2.996169 | val 3.355891 | tok/s 2622541.11 | norm 0.3625 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11100 | train 2.881149 | val 3.355891 | tok/s 2626822.57 | norm 0.3932 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11101 | train 3.013530 | val 3.355891 | tok/s 2624088.76 | norm 0.4442 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11102 | train 3.082590 | val 3.355891 | tok/s 2626388.83 | norm 0.4165 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11103 | train 2.872884 | val 3.355891 | tok/s 2624963.76 | norm 0.3313 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11104 | train 2.934477 | val 3.355891 | tok/s 2629461.66 | norm 0.3695 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11105 | train 3.030927 | val 3.355891 | tok/s 2607908.02 | norm 0.3709 | dt 201.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11106 | train 2.970557 | val 3.355891 | tok/s 2618187.27 | norm 0.3558 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11107 | train 3.123678 | val 3.355891 | tok/s 2622881.00 | norm 0.3920 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11108 | train 3.040133 | val 3.355891 | tok/s 2627132.96 | norm 0.3759 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11109 | train 2.885952 | val 3.355891 | tok/s 2628698.12 | norm 0.3754 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11110 | train 2.948250 | val 3.355891 | tok/s 2631905.55 | norm 0.3847 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11111 | train 3.029785 | val 3.355891 | tok/s 2621182.46 | norm 0.3872 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11112 | train 2.881459 | val 3.355891 | tok/s 2622683.61 | norm 0.3730 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11113 | train 3.150068 | val 3.355891 | tok/s 2627682.65 | norm 0.3612 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11114 | train 2.975553 | val 3.355891 | tok/s 2625831.65 | norm 0.4010 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11115 | train 3.016442 | val 3.355891 | tok/s 2626152.51 | norm 0.3582 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11116 | train 2.971305 | val 3.355891 | tok/s 2629020.93 | norm 0.3838 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11117 | train 3.048791 | val 3.355891 | tok/s 2631508.90 | norm 0.4141 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11118 | train 3.003051 | val 3.355891 | tok/s 2625732.18 | norm 0.4074 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11119 | train 2.965396 | val 3.355891 | tok/s 2632275.92 | norm 0.3604 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11120 | train 2.842835 | val 3.355891 | tok/s 2632550.81 | norm 0.3889 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11121 | train 2.932553 | val 3.355891 | tok/s 2626664.34 | norm 0.4101 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11122 | train 3.222492 | val 3.355891 | tok/s 2506276.41 | norm 0.3935 | dt 209.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11123 | train 2.976709 | val 3.355891 | tok/s 2623815.27 | norm 0.3792 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11124 | train 3.060338 | val 3.355891 | tok/s 2629931.73 | norm 0.4098 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11125 | train 2.875877 | val 3.355891 | tok/s 2629673.05 | norm 0.4154 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11126 | train 2.965145 | val 3.355891 | tok/s 2632086.88 | norm 0.3941 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11127 | train 2.998459 | val 3.355891 | tok/s 2627982.93 | norm 0.4576 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11128 | train 2.945699 | val 3.355891 | tok/s 2622637.55 | norm 0.4210 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11129 | train 3.002552 | val 3.355891 | tok/s 2629387.68 | norm 0.3792 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11130 | train 2.956009 | val 3.355891 | tok/s 2629634.12 | norm 0.3933 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11131 | train 3.094473 | val 3.355891 | tok/s 2625221.36 | norm 0.4042 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11132 | train 2.934392 | val 3.355891 | tok/s 2624475.36 | norm 0.3927 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11133 | train 2.992890 | val 3.355891 | tok/s 2627136.79 | norm 0.3961 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11134 | train 2.870464 | val 3.355891 | tok/s 2626189.91 | norm 0.3971 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11135 | train 2.891452 | val 3.355891 | tok/s 2624712.02 | norm 0.3935 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11136 | train 2.833081 | val 3.355891 | tok/s 2629787.30 | norm 0.4002 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11137 | train 3.023515 | val 3.355891 | tok/s 2626591.63 | norm 0.3592 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11138 | train 2.998029 | val 3.355891 | tok/s 2630084.63 | norm 0.3862 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11139 | train 3.026817 | val 3.355891 | tok/s 2625301.31 | norm 0.3960 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11140 | train 2.885914 | val 3.355891 | tok/s 2629106.12 | norm 0.3609 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11141 | train 3.337587 | val 3.355891 | tok/s 2622448.44 | norm 0.3940 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11142 | train 3.144551 | val 3.355891 | tok/s 2494288.29 | norm 0.4327 | dt 210.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11143 | train 2.986275 | val 3.355891 | tok/s 2620902.61 | norm 0.3697 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11144 | train 2.964312 | val 3.355891 | tok/s 2627543.60 | norm 0.3744 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11145 | train 3.000491 | val 3.355891 | tok/s 2629392.91 | norm 0.3818 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11146 | train 3.029396 | val 3.355891 | tok/s 2628087.56 | norm 0.4424 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11147 | train 3.008263 | val 3.355891 | tok/s 2629568.35 | norm 0.4321 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11148 | train 2.852553 | val 3.355891 | tok/s 2622968.48 | norm 0.4187 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11149 | train 3.176674 | val 3.355891 | tok/s 2624141.34 | norm 0.4106 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11150 | train 3.032377 | val 3.355891 | tok/s 2628741.03 | norm 0.3883 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11151 | train 2.905922 | val 3.355891 | tok/s 2624541.08 | norm 0.3680 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11152 | train 2.951054 | val 3.355891 | tok/s 2627747.25 | norm 0.3837 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11153 | train 3.125953 | val 3.355891 | tok/s 2626848.77 | norm 0.3711 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11154 | train 2.988672 | val 3.355891 | tok/s 2628853.00 | norm 0.3814 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11155 | train 3.138019 | val 3.355891 | tok/s 2626290.15 | norm 0.4232 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11156 | train 2.986420 | val 3.355891 | tok/s 2629795.36 | norm 0.3664 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11157 | train 3.012938 | val 3.355891 | tok/s 2629287.67 | norm 0.3905 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11158 | train 3.025551 | val 3.355891 | tok/s 2627714.90 | norm 0.4290 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11159 | train 3.020382 | val 3.355891 | tok/s 2626479.52 | norm 0.3860 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11160 | train 2.936453 | val 3.355891 | tok/s 2626694.12 | norm 0.3566 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11161 | train 3.079231 | val 3.355891 | tok/s 2621399.67 | norm 0.3780 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11162 | train 2.926859 | val 3.355891 | tok/s 2627094.62 | norm 0.4934 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11163 | train 2.964156 | val 3.355891 | tok/s 2625871.18 | norm 0.4229 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11164 | train 3.051062 | val 3.355891 | tok/s 2625217.04 | norm 0.3826 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11165 | train 3.013202 | val 3.355891 | tok/s 2621858.45 | norm 0.4030 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11166 | train 3.011301 | val 3.355891 | tok/s 2618044.33 | norm 0.3798 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11167 | train 2.930080 | val 3.355891 | tok/s 2628232.81 | norm 0.3853 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11168 | train 3.007499 | val 3.355891 | tok/s 2620322.03 | norm 0.3757 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11169 | train 2.986040 | val 3.355891 | tok/s 2626823.85 | norm 0.3659 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11170 | train 2.844522 | val 3.355891 | tok/s 2625442.68 | norm 0.4039 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11171 | train 3.109742 | val 3.355891 | tok/s 2620378.13 | norm 0.4131 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11172 | train 2.908151 | val 3.355891 | tok/s 2623932.26 | norm 0.3860 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11173 | train 3.030389 | val 3.355891 | tok/s 2625728.56 | norm 0.4024 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11174 | train 2.960881 | val 3.355891 | tok/s 2623994.55 | norm 0.3861 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11175 | train 2.944723 | val 3.355891 | tok/s 2625712.10 | norm 0.4392 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11176 | train 2.910272 | val 3.355891 | tok/s 2628444.48 | norm 0.4294 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11177 | train 3.026019 | val 3.355891 | tok/s 2629089.49 | norm 0.3942 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11178 | train 2.873765 | val 3.355891 | tok/s 2626348.72 | norm 0.3946 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11179 | train 3.011939 | val 3.355891 | tok/s 2622579.06 | norm 0.4345 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11180 | train 2.924179 | val 3.355891 | tok/s 2508599.67 | norm 0.4093 | dt 209.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11181 | train 2.892088 | val 3.355891 | tok/s 2615524.66 | norm 0.3876 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11182 | train 2.858426 | val 3.355891 | tok/s 2627328.09 | norm 0.4436 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11183 | train 2.951920 | val 3.355891 | tok/s 2626949.87 | norm 0.4449 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11184 | train 3.010508 | val 3.355891 | tok/s 2629072.83 | norm 0.4022 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11185 | train 2.924268 | val 3.355891 | tok/s 2626038.36 | norm 0.4177 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11186 | train 3.005362 | val 3.355891 | tok/s 2620484.17 | norm 0.4531 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11187 | train 2.930430 | val 3.355891 | tok/s 2628246.59 | norm 0.4294 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11188 | train 3.157384 | val 3.355891 | tok/s 2623980.49 | norm 0.3691 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11189 | train 3.002170 | val 3.355891 | tok/s 2620289.92 | norm 0.3418 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11190 | train 2.920497 | val 3.355891 | tok/s 2625931.17 | norm 0.3824 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11191 | train 2.980792 | val 3.355891 | tok/s 2622530.09 | norm 0.3920 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11192 | train 3.306007 | val 3.355891 | tok/s 2625580.83 | norm 0.5242 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11193 | train 3.343159 | val 3.355891 | tok/s 2629023.51 | norm 0.8066 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11194 | train 3.173283 | val 3.355891 | tok/s 2624653.73 | norm 0.4222 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11195 | train 2.904847 | val 3.355891 | tok/s 2619917.70 | norm 0.3507 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11196 | train 3.199577 | val 3.355891 | tok/s 2609091.02 | norm 0.4714 | dt 200.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11197 | train 2.960207 | val 3.355891 | tok/s 2623701.33 | norm 0.4056 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11198 | train 3.050265 | val 3.355891 | tok/s 2625792.71 | norm 0.3709 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11199 | train 2.896354 | val 3.355891 | tok/s 2625166.13 | norm 0.3904 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11200 | train 3.064571 | val 3.355891 | tok/s 2617365.14 | norm 0.4798 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11201 | train 2.951465 | val 3.355891 | tok/s 2625980.32 | norm 0.4226 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11202 | train 3.060002 | val 3.355891 | tok/s 2607756.74 | norm 0.3557 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11203 | train 3.175898 | val 3.355891 | tok/s 2626262.66 | norm 0.4164 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11204 | train 3.008894 | val 3.355891 | tok/s 2510372.34 | norm 0.4012 | dt 208.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11205 | train 2.955296 | val 3.355891 | tok/s 2623806.44 | norm 0.3822 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11206 | train 2.880643 | val 3.355891 | tok/s 2618618.66 | norm 0.3785 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11207 | train 3.055064 | val 3.355891 | tok/s 2627854.86 | norm 0.4019 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11208 | train 3.040338 | val 3.355891 | tok/s 2625899.21 | norm 0.4040 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11209 | train 3.048532 | val 3.355891 | tok/s 2623231.14 | norm 0.3919 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11210 | train 2.853955 | val 3.355891 | tok/s 2622229.88 | norm 0.4170 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11211 | train 3.087728 | val 3.355891 | tok/s 2605211.50 | norm 0.4181 | dt 201.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11212 | train 3.044857 | val 3.355891 | tok/s 2630481.90 | norm 0.4190 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11213 | train 3.103912 | val 3.355891 | tok/s 2627177.12 | norm 0.4059 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11214 | train 2.892050 | val 3.355891 | tok/s 2626047.33 | norm 0.4265 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11215 | train 2.965147 | val 3.355891 | tok/s 2625788.23 | norm 0.4311 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11216 | train 2.981348 | val 3.355891 | tok/s 2628170.75 | norm 0.4305 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11217 | train 3.128367 | val 3.355891 | tok/s 2626514.12 | norm 0.3926 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11218 | train 2.969606 | val 3.355891 | tok/s 2626697.28 | norm 0.3578 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11219 | train 3.000230 | val 3.355891 | tok/s 2627567.17 | norm 0.4102 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11220 | train 3.034389 | val 3.355891 | tok/s 2562812.95 | norm 0.4125 | dt 204.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11221 | train 3.081324 | val 3.355891 | tok/s 2625229.02 | norm 0.4236 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11222 | train 3.186694 | val 3.355891 | tok/s 2630723.01 | norm 0.4073 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11223 | train 2.944057 | val 3.355891 | tok/s 2628865.84 | norm 0.4000 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11224 | train 3.015077 | val 3.355891 | tok/s 2629079.70 | norm 0.3368 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11225 | train 3.041934 | val 3.355891 | tok/s 2627051.76 | norm 0.3434 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11226 | train 2.969016 | val 3.355891 | tok/s 2516533.78 | norm 0.3596 | dt 208.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11227 | train 2.999048 | val 3.355891 | tok/s 2613851.15 | norm 0.3771 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11228 | train 3.023238 | val 3.355891 | tok/s 2622718.84 | norm 0.3740 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11229 | train 2.920814 | val 3.355891 | tok/s 2622290.08 | norm 0.3820 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11230 | train 3.019321 | val 3.355891 | tok/s 2621912.15 | norm 0.4333 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11231 | train 3.022907 | val 3.355891 | tok/s 2624270.28 | norm 0.4267 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11232 | train 2.987508 | val 3.355891 | tok/s 2624445.54 | norm 0.4376 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11233 | train 2.949001 | val 3.355891 | tok/s 2617667.22 | norm 0.4262 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11234 | train 3.014901 | val 3.355891 | tok/s 2628054.70 | norm 0.4019 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11235 | train 2.945109 | val 3.355891 | tok/s 2627432.50 | norm 0.3664 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11236 | train 3.012047 | val 3.355891 | tok/s 2628085.69 | norm 0.3749 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11237 | train 2.956250 | val 3.355891 | tok/s 2628092.08 | norm 0.3989 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11238 | train 3.021392 | val 3.355891 | tok/s 2626343.62 | norm 0.4207 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11239 | train 3.006906 | val 3.355891 | tok/s 2628710.55 | norm 0.3370 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11240 | train 3.057920 | val 3.355891 | tok/s 2624370.50 | norm 0.3678 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11241 | train 2.963733 | val 3.355891 | tok/s 2626156.24 | norm 0.3835 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11242 | train 2.931750 | val 3.355891 | tok/s 2585478.11 | norm 0.3494 | dt 202.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11243 | train 3.074432 | val 3.355891 | tok/s 2599569.20 | norm 0.3696 | dt 201.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11244 | train 3.008018 | val 3.355891 | tok/s 2628035.49 | norm 0.3483 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11245 | train 2.997864 | val 3.355891 | tok/s 2626235.24 | norm 0.4000 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11246 | train 3.158606 | val 3.355891 | tok/s 2625631.44 | norm 0.4110 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11247 | train 2.970872 | val 3.355891 | tok/s 2620308.85 | norm 0.3754 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11248 | train 3.114869 | val 3.355891 | tok/s 2625564.16 | norm 0.4088 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11249 | train 2.948825 | val 3.355891 | tok/s 2630984.41 | norm 0.3995 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 11250 | val_loss 3.355799 step 11250 | train 3.020218 | val 3.355799 | tok/s 2612821.98 | norm 0.3988 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11251 | train 2.972457 | val 3.355799 | tok/s 2627070.50 | norm 0.4207 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11252 | train 2.992453 | val 3.355799 | tok/s 2630840.85 | norm 0.3961 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11253 | train 2.935131 | val 3.355799 | tok/s 2617365.13 | norm 0.3755 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11254 | train 2.913134 | val 3.355799 | tok/s 2618013.99 | norm 0.3791 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11255 | train 3.046496 | val 3.355799 | tok/s 2632943.30 | norm 0.3908 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11256 | train 2.987411 | val 3.355799 | tok/s 2632183.96 | norm 0.4441 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11257 | train 2.914001 | val 3.355799 | tok/s 2629477.51 | norm 0.4812 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11258 | train 2.895725 | val 3.355799 | tok/s 2623474.44 | norm 0.3918 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11259 | train 3.021563 | val 3.355799 | tok/s 2626621.87 | norm 0.3982 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11260 | train 2.947526 | val 3.355799 | tok/s 2632394.29 | norm 0.4382 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11261 | train 2.962106 | val 3.355799 | tok/s 2631163.16 | norm 0.3735 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11262 | train 2.963838 | val 3.355799 | tok/s 2631083.42 | norm 0.3707 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11263 | train 2.952374 | val 3.355799 | tok/s 2630352.86 | norm 0.4193 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11264 | train 3.115270 | val 3.355799 | tok/s 2624731.43 | norm 0.3971 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11265 | train 3.020568 | val 3.355799 | tok/s 2628727.57 | norm 0.3828 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11266 | train 3.013107 | val 3.355799 | tok/s 2626263.24 | norm 0.4269 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11267 | train 3.003463 | val 3.355799 | tok/s 2629997.85 | norm 0.4436 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11268 | train 3.046890 | val 3.355799 | tok/s 2616131.86 | norm 0.4219 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11269 | train 2.890254 | val 3.355799 | tok/s 2599141.78 | norm 0.3999 | dt 201.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11270 | train 2.951789 | val 3.355799 | tok/s 2627358.56 | norm 0.3351 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11271 | train 2.952176 | val 3.355799 | tok/s 2625558.74 | norm 0.3156 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11272 | train 2.879118 | val 3.355799 | tok/s 2630275.42 | norm 0.3784 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11273 | train 2.966049 | val 3.355799 | tok/s 2624155.19 | norm 0.4081 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11274 | train 2.994621 | val 3.355799 | tok/s 2625994.52 | norm 0.3900 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11275 | train 2.945916 | val 3.355799 | tok/s 2629940.98 | norm 0.3849 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11276 | train 3.053081 | val 3.355799 | tok/s 2622863.87 | norm 0.4024 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11277 | train 3.044071 | val 3.355799 | tok/s 2632479.59 | norm 0.3991 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11278 | train 2.944685 | val 3.355799 | tok/s 2625088.86 | norm 0.4084 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11279 | train 3.008032 | val 3.355799 | tok/s 2628587.33 | norm 0.3956 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11280 | train 3.021501 | val 3.355799 | tok/s 2628405.20 | norm 0.4074 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11281 | train 3.076894 | val 3.355799 | tok/s 2623988.48 | norm 0.4182 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11282 | train 3.165217 | val 3.355799 | tok/s 2628898.90 | norm 0.4192 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11283 | train 2.920752 | val 3.355799 | tok/s 2626471.35 | norm 0.4250 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11284 | train 3.010405 | val 3.355799 | tok/s 2625150.48 | norm 0.3735 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11285 | train 2.889425 | val 3.355799 | tok/s 2632874.72 | norm 0.4195 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11286 | train 2.982407 | val 3.355799 | tok/s 2631421.23 | norm 0.4122 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11287 | train 2.953843 | val 3.355799 | tok/s 2627769.77 | norm 0.3909 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11288 | train 2.967908 | val 3.355799 | tok/s 2624185.20 | norm 0.4028 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11289 | train 2.924592 | val 3.355799 | tok/s 2628014.23 | norm 0.3785 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11290 | train 2.941919 | val 3.355799 | tok/s 2627762.37 | norm 0.3804 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11291 | train 3.050222 | val 3.355799 | tok/s 2628658.99 | norm 0.3802 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11292 | train 2.905890 | val 3.355799 | tok/s 2628390.28 | norm 0.3660 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11293 | train 2.989540 | val 3.355799 | tok/s 2626511.44 | norm 0.3974 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11294 | train 3.035478 | val 3.355799 | tok/s 2628616.36 | norm 0.4080 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11295 | train 2.985349 | val 3.355799 | tok/s 2631295.21 | norm 0.3949 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11296 | train 3.037798 | val 3.355799 | tok/s 2622888.71 | norm 0.3892 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11297 | train 2.938910 | val 3.355799 | tok/s 2632188.99 | norm 0.4105 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11298 | train 3.095061 | val 3.355799 | tok/s 2627491.82 | norm 0.3861 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11299 | train 3.013371 | val 3.355799 | tok/s 2630062.94 | norm 0.4219 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11300 | train 2.908938 | val 3.355799 | tok/s 2489130.04 | norm 0.4363 | dt 210.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11301 | train 2.910187 | val 3.355799 | tok/s 2623920.06 | norm 0.3999 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11302 | train 3.083192 | val 3.355799 | tok/s 2512779.07 | norm 0.3907 | dt 208.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11303 | train 3.164715 | val 3.355799 | tok/s 2628406.43 | norm 0.3846 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11304 | train 3.007479 | val 3.355799 | tok/s 2628339.97 | norm 0.3937 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11305 | train 2.887314 | val 3.355799 | tok/s 2618286.55 | norm 0.4128 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11306 | train 2.975568 | val 3.355799 | tok/s 2624871.97 | norm 0.3689 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11307 | train 3.258437 | val 3.355799 | tok/s 2627495.13 | norm 0.4025 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11308 | train 2.932757 | val 3.355799 | tok/s 2626888.39 | norm 0.4121 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11309 | train 2.833755 | val 3.355799 | tok/s 2628852.00 | norm 0.3897 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11310 | train 2.988031 | val 3.355799 | tok/s 2617354.15 | norm 0.4493 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11311 | train 2.942761 | val 3.355799 | tok/s 2621576.56 | norm 0.3959 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11312 | train 3.038379 | val 3.355799 | tok/s 2629289.52 | norm 0.4194 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11313 | train 2.920310 | val 3.355799 | tok/s 2629687.77 | norm 0.4581 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11314 | train 2.949070 | val 3.355799 | tok/s 2630886.98 | norm 0.4311 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11315 | train 3.060569 | val 3.355799 | tok/s 2622396.47 | norm 0.4022 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11316 | train 3.000061 | val 3.355799 | tok/s 2627425.34 | norm 0.4053 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11317 | train 2.940734 | val 3.355799 | tok/s 2630305.30 | norm 0.3644 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11318 | train 3.104940 | val 3.355799 | tok/s 2628186.92 | norm 0.3562 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11319 | train 3.046422 | val 3.355799 | tok/s 2627011.68 | norm 0.3543 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11320 | train 2.937774 | val 3.355799 | tok/s 2624898.67 | norm 0.3359 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11321 | train 2.994698 | val 3.355799 | tok/s 2627871.47 | norm 0.3969 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11322 | train 2.808950 | val 3.355799 | tok/s 2632740.23 | norm 0.4186 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11323 | train 2.940573 | val 3.355799 | tok/s 2628934.78 | norm 0.3947 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11324 | train 3.075201 | val 3.355799 | tok/s 2624566.74 | norm 0.4287 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11325 | train 2.979357 | val 3.355799 | tok/s 2622859.26 | norm 0.4253 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11326 | train 2.985564 | val 3.355799 | tok/s 2628354.12 | norm 0.3681 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11327 | train 3.059792 | val 3.355799 | tok/s 2627512.18 | norm 0.3735 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11328 | train 3.200356 | val 3.355799 | tok/s 2624919.61 | norm 0.4249 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11329 | train 3.038917 | val 3.355799 | tok/s 2629156.88 | norm 0.3740 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11330 | train 3.027264 | val 3.355799 | tok/s 2624423.08 | norm 0.3936 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11331 | train 2.941302 | val 3.355799 | tok/s 2622484.18 | norm 0.4181 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11332 | train 2.902018 | val 3.355799 | tok/s 2629957.51 | norm 0.3949 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11333 | train 2.907441 | val 3.355799 | tok/s 2630478.19 | norm 0.3763 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11334 | train 2.938779 | val 3.355799 | tok/s 2627252.24 | norm 0.3269 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11335 | train 2.990525 | val 3.355799 | tok/s 2628090.84 | norm 0.3424 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11336 | train 3.132607 | val 3.355799 | tok/s 2629636.18 | norm 0.4383 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11337 | train 2.997727 | val 3.355799 | tok/s 2633719.64 | norm 0.3956 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11338 | train 2.968275 | val 3.355799 | tok/s 2626004.17 | norm 0.3695 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11339 | train 2.853097 | val 3.355799 | tok/s 2628688.83 | norm 0.3900 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11340 | train 2.997641 | val 3.355799 | tok/s 2629018.65 | norm 0.3804 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11341 | train 2.982329 | val 3.355799 | tok/s 2630235.59 | norm 0.4654 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11342 | train 2.921721 | val 3.355799 | tok/s 2630967.99 | norm 0.3967 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11343 | train 2.928061 | val 3.355799 | tok/s 2626213.68 | norm 0.3655 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11344 | train 2.932416 | val 3.355799 | tok/s 2629803.41 | norm 0.3818 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11345 | train 3.026978 | val 3.355799 | tok/s 2627993.76 | norm 0.3964 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11346 | train 3.016240 | val 3.355799 | tok/s 2632412.00 | norm 0.3837 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11347 | train 2.993541 | val 3.355799 | tok/s 2603624.12 | norm 0.3690 | dt 201.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11348 | train 2.945189 | val 3.355799 | tok/s 2570658.25 | norm 0.3717 | dt 203.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11349 | train 2.837111 | val 3.355799 | tok/s 2632300.09 | norm 0.3940 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11350 | train 3.005167 | val 3.355799 | tok/s 2634281.21 | norm 0.4284 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11351 | train 2.900415 | val 3.355799 | tok/s 2631220.59 | norm 0.3666 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11352 | train 2.951507 | val 3.355799 | tok/s 2624308.46 | norm 0.4083 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11353 | train 3.085638 | val 3.355799 | tok/s 2621268.75 | norm 0.4274 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11354 | train 2.985653 | val 3.355799 | tok/s 2475074.51 | norm 0.4381 | dt 211.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11355 | train 2.982178 | val 3.355799 | tok/s 2627528.44 | norm 0.4538 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11356 | train 3.020580 | val 3.355799 | tok/s 2629564.86 | norm 0.4227 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11357 | train 2.959694 | val 3.355799 | tok/s 2621525.58 | norm 0.4022 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11358 | train 3.052310 | val 3.355799 | tok/s 2624263.30 | norm 0.4313 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11359 | train 2.996460 | val 3.355799 | tok/s 2631959.03 | norm 0.4556 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11360 | train 2.878643 | val 3.355799 | tok/s 2624870.52 | norm 0.4041 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11361 | train 3.178400 | val 3.355799 | tok/s 2627951.03 | norm 0.4012 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11362 | train 3.108306 | val 3.355799 | tok/s 2613668.41 | norm 0.3946 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11363 | train 2.943457 | val 3.355799 | tok/s 2622502.31 | norm 0.3902 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11364 | train 2.986883 | val 3.355799 | tok/s 2633290.44 | norm 0.3856 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11365 | train 3.082667 | val 3.355799 | tok/s 2630609.74 | norm 0.3793 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11366 | train 2.939619 | val 3.355799 | tok/s 2629981.96 | norm 0.3651 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11367 | train 2.969871 | val 3.355799 | tok/s 2621037.20 | norm 0.3693 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11368 | train 3.127813 | val 3.355799 | tok/s 2627127.02 | norm 0.3889 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11369 | train 3.073350 | val 3.355799 | tok/s 2633577.25 | norm 0.4014 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11370 | train 3.095170 | val 3.355799 | tok/s 2626884.35 | norm 0.3750 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11371 | train 3.034650 | val 3.355799 | tok/s 2630784.24 | norm 0.3521 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11372 | train 2.868812 | val 3.355799 | tok/s 2627315.26 | norm 0.3955 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11373 | train 2.886778 | val 3.355799 | tok/s 2571245.43 | norm 0.3824 | dt 203.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11374 | train 2.948812 | val 3.355799 | tok/s 2628171.17 | norm 0.3734 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11375 | train 2.927848 | val 3.355799 | tok/s 2631979.33 | norm 0.3822 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11376 | train 3.020038 | val 3.355799 | tok/s 2631443.12 | norm 0.3556 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11377 | train 3.003150 | val 3.355799 | tok/s 2627586.56 | norm 0.3914 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11378 | train 2.943094 | val 3.355799 | tok/s 2627427.26 | norm 0.4146 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11379 | train 3.014492 | val 3.355799 | tok/s 2630036.35 | norm 0.3845 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11380 | train 3.009010 | val 3.355799 | tok/s 2626751.44 | norm 0.4068 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11381 | train 3.028697 | val 3.355799 | tok/s 2629912.82 | norm 0.4274 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11382 | train 2.981979 | val 3.355799 | tok/s 2623089.16 | norm 0.4327 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11383 | train 3.036026 | val 3.355799 | tok/s 2627492.15 | norm 0.4453 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11384 | train 3.036884 | val 3.355799 | tok/s 2630415.11 | norm 0.4342 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11385 | train 2.883563 | val 3.355799 | tok/s 2628860.84 | norm 0.3950 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11386 | train 2.938920 | val 3.355799 | tok/s 2620842.34 | norm 0.4121 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11387 | train 2.905866 | val 3.355799 | tok/s 2624132.74 | norm 0.4177 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11388 | train 2.935094 | val 3.355799 | tok/s 2628524.28 | norm 0.3966 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11389 | train 2.971850 | val 3.355799 | tok/s 2628342.34 | norm 0.4015 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11390 | train 3.077548 | val 3.355799 | tok/s 2623749.77 | norm 0.3677 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11391 | train 2.976863 | val 3.355799 | tok/s 2628120.60 | norm 0.3533 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11392 | train 3.123662 | val 3.355799 | tok/s 2630284.23 | norm 0.3501 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11393 | train 2.959728 | val 3.355799 | tok/s 2627050.28 | norm 0.3631 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11394 | train 2.994268 | val 3.355799 | tok/s 2630881.36 | norm 0.3497 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11395 | train 2.951325 | val 3.355799 | tok/s 2630080.52 | norm 0.3653 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11396 | train 3.016793 | val 3.355799 | tok/s 2629268.21 | norm 0.3956 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11397 | train 2.988343 | val 3.355799 | tok/s 2626509.33 | norm 0.3933 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11398 | train 3.005755 | val 3.355799 | tok/s 2628172.58 | norm 0.4023 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11399 | train 3.097827 | val 3.355799 | tok/s 2625930.83 | norm 0.4155 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11400 | train 3.031944 | val 3.355799 | tok/s 2624956.00 | norm 0.3894 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11401 | train 2.992952 | val 3.355799 | tok/s 2624165.02 | norm 0.3890 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11402 | train 3.002390 | val 3.355799 | tok/s 2625064.95 | norm 0.3999 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11403 | train 2.890655 | val 3.355799 | tok/s 2628613.97 | norm 0.4355 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11404 | train 2.856106 | val 3.355799 | tok/s 2624628.39 | norm 0.4177 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11405 | train 2.992317 | val 3.355799 | tok/s 2622926.04 | norm 0.4120 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11406 | train 2.985917 | val 3.355799 | tok/s 2629700.15 | norm 0.4183 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11407 | train 2.990659 | val 3.355799 | tok/s 2629504.41 | norm 0.4389 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11408 | train 2.923847 | val 3.355799 | tok/s 2631189.15 | norm 0.3947 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11409 | train 1.519675 | val 3.355799 | tok/s 2629577.06 | norm 3.6134 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11410 | train 2.531901 | val 3.355799 | tok/s 2627441.80 | norm 1.1995 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11411 | train 2.892715 | val 3.355799 | tok/s 2479882.38 | norm 0.3009 | dt 211.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11412 | train 3.017710 | val 3.355799 | tok/s 2543047.47 | norm 0.4012 | dt 206.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11413 | train 2.898435 | val 3.355799 | tok/s 2627298.98 | norm 0.3507 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11414 | train 3.050084 | val 3.355799 | tok/s 2620951.34 | norm 0.4069 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11415 | train 2.920014 | val 3.355799 | tok/s 2624489.74 | norm 0.4100 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11416 | train 2.960153 | val 3.355799 | tok/s 2630072.65 | norm 0.3967 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11417 | train 3.076848 | val 3.355799 | tok/s 2628464.26 | norm 0.4567 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11418 | train 2.987377 | val 3.355799 | tok/s 2631527.91 | norm 0.3782 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11419 | train 2.797552 | val 3.355799 | tok/s 2631123.81 | norm 0.4189 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11420 | train 2.990641 | val 3.355799 | tok/s 2626409.88 | norm 0.4314 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11421 | train 3.011591 | val 3.355799 | tok/s 2632097.45 | norm 0.3693 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11422 | train 3.095456 | val 3.355799 | tok/s 2626888.89 | norm 0.3948 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11423 | train 2.920997 | val 3.355799 | tok/s 2604888.82 | norm 0.3990 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11424 | train 3.012939 | val 3.355799 | tok/s 2628268.87 | norm 0.3727 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11425 | train 2.905464 | val 3.355799 | tok/s 2628433.74 | norm 0.3907 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11426 | train 3.296447 | val 3.355799 | tok/s 2632078.62 | norm 0.6000 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11427 | train 2.987675 | val 3.355799 | tok/s 2628988.55 | norm 0.3956 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11428 | train 3.008311 | val 3.355799 | tok/s 2629471.89 | norm 0.3671 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11429 | train 2.998794 | val 3.355799 | tok/s 2626810.28 | norm 0.4025 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11430 | train 2.974671 | val 3.355799 | tok/s 2623629.62 | norm 0.3628 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11431 | train 3.060155 | val 3.355799 | tok/s 2630207.34 | norm 0.3833 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11432 | train 3.112682 | val 3.355799 | tok/s 2627259.13 | norm 0.3777 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11433 | train 3.500683 | val 3.355799 | tok/s 2628054.35 | norm 4.2519 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11434 | train 3.055421 | val 3.355799 | tok/s 2626588.68 | norm 0.5557 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11435 | train 3.051165 | val 3.355799 | tok/s 2620805.71 | norm 0.3610 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11436 | train 2.843328 | val 3.355799 | tok/s 2631750.08 | norm 0.3388 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11437 | train 3.215214 | val 3.355799 | tok/s 2629800.42 | norm 0.5497 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11438 | train 2.911147 | val 3.355799 | tok/s 2630359.56 | norm 0.4218 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11439 | train 2.953479 | val 3.355799 | tok/s 2629045.61 | norm 0.3845 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11440 | train 2.986656 | val 3.355799 | tok/s 2630559.08 | norm 0.4501 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11441 | train 2.980408 | val 3.355799 | tok/s 2629888.86 | norm 0.4423 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11442 | train 3.050525 | val 3.355799 | tok/s 2626634.45 | norm 0.4043 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11443 | train 2.992466 | val 3.355799 | tok/s 2632467.74 | norm 0.3922 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11444 | train 2.965369 | val 3.355799 | tok/s 2629682.15 | norm 0.3779 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11445 | train 2.980182 | val 3.355799 | tok/s 2631766.33 | norm 0.3990 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11446 | train 2.905617 | val 3.355799 | tok/s 2628682.74 | norm 0.3425 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11447 | train 3.017886 | val 3.355799 | tok/s 2623155.19 | norm 0.3842 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11448 | train 2.867073 | val 3.355799 | tok/s 2627837.15 | norm 0.4183 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11449 | train 3.064316 | val 3.355799 | tok/s 2626225.24 | norm 0.4065 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11450 | train 2.957850 | val 3.355799 | tok/s 2631081.20 | norm 0.4152 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11451 | train 2.910794 | val 3.355799 | tok/s 2627096.66 | norm 0.3762 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11452 | train 3.123891 | val 3.355799 | tok/s 2624855.91 | norm 0.3684 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11453 | train 3.027108 | val 3.355799 | tok/s 2552176.13 | norm 0.3999 | dt 205.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11454 | train 2.954947 | val 3.355799 | tok/s 2625910.45 | norm 0.3794 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11455 | train 3.072169 | val 3.355799 | tok/s 2628146.77 | norm 0.4043 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11456 | train 3.026680 | val 3.355799 | tok/s 2632191.81 | norm 0.4034 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11457 | train 3.046759 | val 3.355799 | tok/s 2626878.41 | norm 0.3507 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11458 | train 3.029176 | val 3.355799 | tok/s 2633089.70 | norm 0.3953 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11459 | train 2.938448 | val 3.355799 | tok/s 2628179.76 | norm 0.3944 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11460 | train 2.962872 | val 3.355799 | tok/s 2627548.05 | norm 0.3871 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11461 | train 2.836742 | val 3.355799 | tok/s 2463936.51 | norm 0.3744 | dt 212.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11462 | train 3.043344 | val 3.355799 | tok/s 2627019.49 | norm 0.3620 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11463 | train 3.003816 | val 3.355799 | tok/s 2512834.45 | norm 0.3673 | dt 208.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11464 | train 2.982770 | val 3.355799 | tok/s 2632117.75 | norm 0.3699 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11465 | train 2.956407 | val 3.355799 | tok/s 2629593.91 | norm 0.4221 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11466 | train 3.128748 | val 3.355799 | tok/s 2624446.54 | norm 0.4511 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11467 | train 3.034332 | val 3.355799 | tok/s 2630093.81 | norm 0.4158 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11468 | train 2.971214 | val 3.355799 | tok/s 2630452.23 | norm 0.3620 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11469 | train 3.017452 | val 3.355799 | tok/s 2627875.01 | norm 0.3652 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11470 | train 3.016468 | val 3.355799 | tok/s 2630343.85 | norm 0.4379 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11471 | train 2.844572 | val 3.355799 | tok/s 2609827.09 | norm 0.4141 | dt 200.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11472 | train 3.234619 | val 3.355799 | tok/s 2626406.88 | norm 0.3734 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11473 | train 3.046361 | val 3.355799 | tok/s 2631467.87 | norm 0.3893 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11474 | train 3.063055 | val 3.355799 | tok/s 2629507.01 | norm 0.4016 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11475 | train 3.002336 | val 3.355799 | tok/s 2632633.05 | norm 0.4063 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11476 | train 2.914525 | val 3.355799 | tok/s 2622240.16 | norm 0.3888 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11477 | train 3.032226 | val 3.355799 | tok/s 2577662.94 | norm 0.3961 | dt 203.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11478 | train 3.032768 | val 3.355799 | tok/s 2632984.43 | norm 0.4113 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11479 | train 3.057842 | val 3.355799 | tok/s 2629943.36 | norm 0.4555 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11480 | train 3.009454 | val 3.355799 | tok/s 2633742.55 | norm 0.4630 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11481 | train 3.160515 | val 3.355799 | tok/s 2628135.29 | norm 0.4458 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11482 | train 2.988536 | val 3.355799 | tok/s 2626589.88 | norm 0.3861 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11483 | train 2.949969 | val 3.355799 | tok/s 2635621.09 | norm 0.3699 | dt 198.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11484 | train 2.955864 | val 3.355799 | tok/s 2628810.03 | norm 0.4033 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11485 | train 2.987133 | val 3.355799 | tok/s 2631481.17 | norm 0.3746 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11486 | train 3.035730 | val 3.355799 | tok/s 2628566.47 | norm 0.3598 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11487 | train 3.047338 | val 3.355799 | tok/s 2606663.17 | norm 0.3783 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11488 | train 3.034609 | val 3.355799 | tok/s 2631664.60 | norm 0.3912 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11489 | train 3.056885 | val 3.355799 | tok/s 2629457.65 | norm 0.3752 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11490 | train 3.109760 | val 3.355799 | tok/s 2624049.30 | norm 0.3946 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11491 | train 2.923004 | val 3.355799 | tok/s 2626129.01 | norm 0.3614 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11492 | train 2.954501 | val 3.355799 | tok/s 2630374.15 | norm 0.3625 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11493 | train 2.926017 | val 3.355799 | tok/s 2627654.77 | norm 0.3788 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11494 | train 3.013602 | val 3.355799 | tok/s 2626646.26 | norm 0.4081 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11495 | train 3.012839 | val 3.355799 | tok/s 2628200.24 | norm 0.3870 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11496 | train 3.110466 | val 3.355799 | tok/s 2627321.19 | norm 0.5096 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11497 | train 2.903103 | val 3.355799 | tok/s 2626511.10 | norm 0.4232 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11498 | train 2.919832 | val 3.355799 | tok/s 2633997.54 | norm 0.3506 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11499 | train 3.041010 | val 3.355799 | tok/s 2623094.36 | norm 0.3592 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 11500 | val_loss 3.356704 step 11500 | train 3.009493 | val 3.356704 | tok/s 2615953.51 | norm 0.3780 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11501 | train 3.012397 | val 3.356704 | tok/s 2625227.11 | norm 0.3347 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11502 | train 3.086751 | val 3.356704 | tok/s 2625806.18 | norm 0.3263 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11503 | train 3.008882 | val 3.356704 | tok/s 2624708.01 | norm 0.3390 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11504 | train 2.886801 | val 3.356704 | tok/s 2629108.83 | norm 0.3593 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11505 | train 3.015756 | val 3.356704 | tok/s 2628297.56 | norm 0.3597 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11506 | train 3.019467 | val 3.356704 | tok/s 2629237.79 | norm 0.3930 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11507 | train 3.016859 | val 3.356704 | tok/s 2629859.32 | norm 0.4013 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11508 | train 3.021310 | val 3.356704 | tok/s 2628966.74 | norm 0.4183 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11509 | train 3.050470 | val 3.356704 | tok/s 2634445.77 | norm 0.4531 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11510 | train 3.013604 | val 3.356704 | tok/s 2626362.29 | norm 0.4082 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11511 | train 3.079133 | val 3.356704 | tok/s 2629728.54 | norm 0.4001 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11512 | train 3.017970 | val 3.356704 | tok/s 2627126.53 | norm 0.4091 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11513 | train 2.921160 | val 3.356704 | tok/s 2630024.35 | norm 0.4157 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11514 | train 3.143988 | val 3.356704 | tok/s 2632075.22 | norm 0.4194 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11515 | train 3.035080 | val 3.356704 | tok/s 2629910.60 | norm 0.4255 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11516 | train 2.984321 | val 3.356704 | tok/s 2632068.05 | norm 0.4463 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11517 | train 3.013346 | val 3.356704 | tok/s 2628260.38 | norm 0.4506 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11518 | train 2.973031 | val 3.356704 | tok/s 2632328.40 | norm 0.4260 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11519 | train 2.962039 | val 3.356704 | tok/s 2627269.96 | norm 0.4427 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11520 | train 2.889453 | val 3.356704 | tok/s 2620645.03 | norm 0.4336 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11521 | train 2.929302 | val 3.356704 | tok/s 2628913.66 | norm 0.4285 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11522 | train 2.929969 | val 3.356704 | tok/s 2623531.97 | norm 0.4398 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11523 | train 3.073536 | val 3.356704 | tok/s 2626413.65 | norm 0.4541 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11524 | train 2.982163 | val 3.356704 | tok/s 2629171.42 | norm 0.4034 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11525 | train 3.135462 | val 3.356704 | tok/s 2629118.68 | norm 0.4197 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11526 | train 2.969883 | val 3.356704 | tok/s 2630657.06 | norm 0.4271 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11527 | train 2.875869 | val 3.356704 | tok/s 2628437.02 | norm 0.3832 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11528 | train 3.073046 | val 3.356704 | tok/s 2630933.57 | norm 0.4071 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11529 | train 2.958847 | val 3.356704 | tok/s 2627895.66 | norm 0.3758 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11530 | train 3.003570 | val 3.356704 | tok/s 2629152.26 | norm 0.3498 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11531 | train 2.964481 | val 3.356704 | tok/s 2632234.68 | norm 0.3529 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11532 | train 2.943691 | val 3.356704 | tok/s 2480424.27 | norm 0.3649 | dt 211.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11533 | train 3.024528 | val 3.356704 | tok/s 2562367.49 | norm 0.3654 | dt 204.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11534 | train 2.882778 | val 3.356704 | tok/s 2625841.46 | norm 0.3763 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11535 | train 2.950877 | val 3.356704 | tok/s 2608446.84 | norm 0.3917 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11536 | train 2.987194 | val 3.356704 | tok/s 2633009.62 | norm 0.4352 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11537 | train 2.952611 | val 3.356704 | tok/s 2629639.22 | norm 0.4391 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11538 | train 2.861713 | val 3.356704 | tok/s 2630993.47 | norm 0.4393 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11539 | train 3.014278 | val 3.356704 | tok/s 2622411.50 | norm 0.4182 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11540 | train 2.914103 | val 3.356704 | tok/s 2624271.37 | norm 0.3910 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11541 | train 2.941984 | val 3.356704 | tok/s 2627727.96 | norm 0.3965 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11542 | train 2.982589 | val 3.356704 | tok/s 2627875.42 | norm 0.4166 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11543 | train 2.993310 | val 3.356704 | tok/s 2630457.34 | norm 0.3952 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11544 | train 3.076025 | val 3.356704 | tok/s 2622267.78 | norm 0.3493 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11545 | train 2.972867 | val 3.356704 | tok/s 2626400.90 | norm 0.3498 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11546 | train 2.870852 | val 3.356704 | tok/s 2634338.02 | norm 0.3971 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11547 | train 2.979689 | val 3.356704 | tok/s 2630355.35 | norm 0.4668 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11548 | train 2.983716 | val 3.356704 | tok/s 2630104.04 | norm 0.3508 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11549 | train 2.903173 | val 3.356704 | tok/s 2622309.43 | norm 0.3570 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11550 | train 3.064901 | val 3.356704 | tok/s 2626690.16 | norm 0.3676 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11551 | train 2.920038 | val 3.356704 | tok/s 2631747.57 | norm 0.4094 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11552 | train 2.886748 | val 3.356704 | tok/s 2622711.20 | norm 0.3651 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11553 | train 3.004678 | val 3.356704 | tok/s 2632594.94 | norm 0.3594 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11554 | train 3.202931 | val 3.356704 | tok/s 2627917.90 | norm 0.3781 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11555 | train 2.931224 | val 3.356704 | tok/s 2628253.58 | norm 0.4527 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11556 | train 2.980030 | val 3.356704 | tok/s 2630049.19 | norm 0.3961 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11557 | train 2.825037 | val 3.356704 | tok/s 2625486.90 | norm 0.3883 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11558 | train 2.946546 | val 3.356704 | tok/s 2630349.69 | norm 0.4198 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11559 | train 2.961770 | val 3.356704 | tok/s 2631473.95 | norm 0.4172 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11560 | train 3.059716 | val 3.356704 | tok/s 2625296.35 | norm 0.3976 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11561 | train 3.002906 | val 3.356704 | tok/s 2631574.22 | norm 0.4053 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11562 | train 2.980565 | val 3.356704 | tok/s 2572799.14 | norm 0.3954 | dt 203.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11563 | train 2.941790 | val 3.356704 | tok/s 2633738.85 | norm 0.3844 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11564 | train 3.112861 | val 3.356704 | tok/s 2621421.77 | norm 0.4151 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11565 | train 2.973834 | val 3.356704 | tok/s 2625134.76 | norm 0.3902 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11566 | train 2.849363 | val 3.356704 | tok/s 2622315.07 | norm 0.3536 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11567 | train 3.108822 | val 3.356704 | tok/s 2621959.83 | norm 0.4131 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11568 | train 2.831380 | val 3.356704 | tok/s 2626220.73 | norm 0.3771 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11569 | train 3.045987 | val 3.356704 | tok/s 2626545.72 | norm 0.3734 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11570 | train 3.028488 | val 3.356704 | tok/s 2630483.47 | norm 0.3881 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11571 | train 2.959725 | val 3.356704 | tok/s 2624088.99 | norm 0.3915 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11572 | train 3.039509 | val 3.356704 | tok/s 2626429.82 | norm 0.4056 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11573 | train 2.936650 | val 3.356704 | tok/s 2627021.71 | norm 0.3796 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11574 | train 2.934605 | val 3.356704 | tok/s 2521396.76 | norm 0.3847 | dt 207.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11575 | train 2.807120 | val 3.356704 | tok/s 2614211.64 | norm 0.3839 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11576 | train 3.150220 | val 3.356704 | tok/s 2625917.70 | norm 0.3808 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11577 | train 3.011535 | val 3.356704 | tok/s 2627998.91 | norm 0.3973 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11578 | train 3.309305 | val 3.356704 | tok/s 2631402.04 | norm 0.4618 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11579 | train 3.031878 | val 3.356704 | tok/s 2626773.64 | norm 0.4108 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11580 | train 2.977718 | val 3.356704 | tok/s 2630120.75 | norm 0.3530 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11581 | train 2.906678 | val 3.356704 | tok/s 2514227.43 | norm 0.3828 | dt 208.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11582 | train 2.950387 | val 3.356704 | tok/s 2626913.20 | norm 0.3922 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11583 | train 2.964001 | val 3.356704 | tok/s 2634436.35 | norm 0.3804 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11584 | train 2.830624 | val 3.356704 | tok/s 2633164.03 | norm 0.3952 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11585 | train 3.062712 | val 3.356704 | tok/s 2633036.45 | norm 0.3710 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11586 | train 2.930892 | val 3.356704 | tok/s 2625093.95 | norm 0.3653 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11587 | train 3.019140 | val 3.356704 | tok/s 2624767.22 | norm 0.4032 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11588 | train 3.071488 | val 3.356704 | tok/s 2626854.04 | norm 0.4328 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11589 | train 2.962103 | val 3.356704 | tok/s 2628369.34 | norm 0.3899 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11590 | train 2.843805 | val 3.356704 | tok/s 2628853.64 | norm 0.4197 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11591 | train 2.891080 | val 3.356704 | tok/s 2624925.43 | norm 0.4401 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11592 | train 2.990953 | val 3.356704 | tok/s 2627360.13 | norm 0.3733 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11593 | train 3.007596 | val 3.356704 | tok/s 2633861.99 | norm 0.4131 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11594 | train 2.992940 | val 3.356704 | tok/s 2628025.32 | norm 0.4014 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11595 | train 2.994671 | val 3.356704 | tok/s 2629164.09 | norm 0.3834 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11596 | train 3.107224 | val 3.356704 | tok/s 2628926.55 | norm 0.4087 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11597 | train 2.851160 | val 3.356704 | tok/s 2629498.26 | norm 0.3892 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11598 | train 2.840734 | val 3.356704 | tok/s 2630638.57 | norm 0.4088 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11599 | train 3.053124 | val 3.356704 | tok/s 2624067.01 | norm 0.4120 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11600 | train 3.057343 | val 3.356704 | tok/s 2621775.63 | norm 0.4228 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11601 | train 2.884075 | val 3.356704 | tok/s 2595572.64 | norm 0.3853 | dt 201.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11602 | train 3.065658 | val 3.356704 | tok/s 2628998.30 | norm 0.3558 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11603 | train 2.946413 | val 3.356704 | tok/s 2631363.58 | norm 0.3742 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11604 | train 2.946121 | val 3.356704 | tok/s 2589610.97 | norm 0.3950 | dt 202.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11605 | train 2.995718 | val 3.356704 | tok/s 2630295.27 | norm 0.3732 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11606 | train 3.075083 | val 3.356704 | tok/s 2627253.34 | norm 0.3793 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11607 | train 3.122741 | val 3.356704 | tok/s 2629706.47 | norm 0.4239 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11608 | train 2.966317 | val 3.356704 | tok/s 2630315.82 | norm 0.4365 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11609 | train 2.941700 | val 3.356704 | tok/s 2626361.84 | norm 0.4034 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11610 | train 2.968662 | val 3.356704 | tok/s 2629521.84 | norm 0.4090 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11611 | train 2.924662 | val 3.356704 | tok/s 2593535.68 | norm 0.4180 | dt 202.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11612 | train 2.981193 | val 3.356704 | tok/s 2628321.99 | norm 0.4102 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11613 | train 2.973515 | val 3.356704 | tok/s 2620890.28 | norm 0.3665 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11614 | train 2.932268 | val 3.356704 | tok/s 2620689.00 | norm 0.3463 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11615 | train 2.984774 | val 3.356704 | tok/s 2633047.84 | norm 0.3864 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11616 | train 2.841266 | val 3.356704 | tok/s 2628041.88 | norm 0.4215 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11617 | train 2.963916 | val 3.356704 | tok/s 2625100.35 | norm 0.3923 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11618 | train 2.997584 | val 3.356704 | tok/s 2626382.06 | norm 0.3839 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11619 | train 3.080783 | val 3.356704 | tok/s 2626768.90 | norm 0.3871 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11620 | train 3.035291 | val 3.356704 | tok/s 2601810.38 | norm 0.3579 | dt 201.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11621 | train 2.988672 | val 3.356704 | tok/s 2626715.06 | norm 0.3672 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11622 | train 2.828021 | val 3.356704 | tok/s 2533476.84 | norm 0.4042 | dt 206.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11623 | train 2.931019 | val 3.356704 | tok/s 2614632.09 | norm 0.3654 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11624 | train 3.011020 | val 3.356704 | tok/s 2570781.30 | norm 0.3868 | dt 203.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11625 | train 3.052216 | val 3.356704 | tok/s 2629256.09 | norm 0.3999 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11626 | train 3.044219 | val 3.356704 | tok/s 2623758.01 | norm 0.3829 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11627 | train 3.084867 | val 3.356704 | tok/s 2632629.97 | norm 0.3837 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11628 | train 2.948208 | val 3.356704 | tok/s 2623380.28 | norm 0.3936 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11629 | train 2.935080 | val 3.356704 | tok/s 2567683.35 | norm 0.3695 | dt 204.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11630 | train 2.958892 | val 3.356704 | tok/s 2628836.14 | norm 0.3437 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11631 | train 2.920214 | val 3.356704 | tok/s 2534869.19 | norm 0.3766 | dt 206.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11632 | train 2.998603 | val 3.356704 | tok/s 2628544.40 | norm 0.4008 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11633 | train 2.896228 | val 3.356704 | tok/s 2624060.62 | norm 0.3987 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11634 | train 3.005135 | val 3.356704 | tok/s 2627899.02 | norm 0.3875 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11635 | train 2.863088 | val 3.356704 | tok/s 2627277.42 | norm 0.3826 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11636 | train 3.042585 | val 3.356704 | tok/s 2626891.09 | norm 0.4084 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11637 | train 2.960192 | val 3.356704 | tok/s 2630193.88 | norm 0.4126 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11638 | train 3.076591 | val 3.356704 | tok/s 2626662.08 | norm 0.3735 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11639 | train 2.958210 | val 3.356704 | tok/s 2628551.60 | norm 0.3950 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11640 | train 3.194837 | val 3.356704 | tok/s 2631698.84 | norm 0.3955 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11641 | train 3.150933 | val 3.356704 | tok/s 2628711.32 | norm 0.3983 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11642 | train 2.920692 | val 3.356704 | tok/s 2633244.08 | norm 0.3885 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11643 | train 3.066342 | val 3.356704 | tok/s 2531674.03 | norm 0.4233 | dt 207.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11644 | train 2.932612 | val 3.356704 | tok/s 2625554.41 | norm 0.4444 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11645 | train 3.147324 | val 3.356704 | tok/s 2628157.27 | norm 0.3893 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11646 | train 3.013504 | val 3.356704 | tok/s 2628156.17 | norm 0.3832 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11647 | train 3.004508 | val 3.356704 | tok/s 2630063.20 | norm 0.4180 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11648 | train 2.954228 | val 3.356704 | tok/s 2626851.39 | norm 0.4271 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11649 | train 2.944461 | val 3.356704 | tok/s 2483967.69 | norm 0.3805 | dt 211.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11650 | train 2.948967 | val 3.356704 | tok/s 2560240.66 | norm 0.3620 | dt 204.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11651 | train 3.058763 | val 3.356704 | tok/s 2620059.60 | norm 0.4253 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11652 | train 3.010780 | val 3.356704 | tok/s 2627876.87 | norm 0.3645 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11653 | train 3.036363 | val 3.356704 | tok/s 2630363.98 | norm 0.3803 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11654 | train 2.947585 | val 3.356704 | tok/s 2628615.44 | norm 0.4152 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11655 | train 2.989881 | val 3.356704 | tok/s 2629312.08 | norm 0.4150 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11656 | train 3.110979 | val 3.356704 | tok/s 2628446.56 | norm 0.4098 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11657 | train 2.932219 | val 3.356704 | tok/s 2628607.65 | norm 0.4303 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11658 | train 2.950841 | val 3.356704 | tok/s 2626108.54 | norm 0.4145 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11659 | train 2.922351 | val 3.356704 | tok/s 2626190.54 | norm 0.4105 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11660 | train 3.075883 | val 3.356704 | tok/s 2625894.48 | norm 0.4102 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11661 | train 2.921969 | val 3.356704 | tok/s 2625792.12 | norm 0.3804 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11662 | train 3.139313 | val 3.356704 | tok/s 2631810.87 | norm 0.4251 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11663 | train 2.901046 | val 3.356704 | tok/s 2614405.11 | norm 0.3990 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11664 | train 3.162104 | val 3.356704 | tok/s 2623230.26 | norm 0.4001 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11665 | train 3.022491 | val 3.356704 | tok/s 2623304.16 | norm 0.3988 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11666 | train 3.109501 | val 3.356704 | tok/s 2627076.71 | norm 0.3790 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11667 | train 2.955175 | val 3.356704 | tok/s 2628883.67 | norm 0.3995 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11668 | train 2.946575 | val 3.356704 | tok/s 2616352.03 | norm 0.4065 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11669 | train 2.959975 | val 3.356704 | tok/s 2626131.36 | norm 0.3770 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11670 | train 2.991900 | val 3.356704 | tok/s 2627117.53 | norm 0.3755 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11671 | train 3.037110 | val 3.356704 | tok/s 2626930.84 | norm 0.3906 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11672 | train 2.898406 | val 3.356704 | tok/s 2621369.83 | norm 0.3769 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11673 | train 2.957711 | val 3.356704 | tok/s 2617768.21 | norm 0.6068 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11674 | train 2.995310 | val 3.356704 | tok/s 2624658.61 | norm 0.3736 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11675 | train 3.032647 | val 3.356704 | tok/s 2624225.96 | norm 0.3217 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11676 | train 3.034411 | val 3.356704 | tok/s 2628814.22 | norm 0.3703 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11677 | train 2.884911 | val 3.356704 | tok/s 2615150.12 | norm 0.4432 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11678 | train 2.903214 | val 3.356704 | tok/s 2621029.90 | norm 0.4358 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11679 | train 2.976791 | val 3.356704 | tok/s 2628132.72 | norm 0.4054 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11680 | train 2.980766 | val 3.356704 | tok/s 2623364.74 | norm 0.4000 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11681 | train 3.039207 | val 3.356704 | tok/s 2626902.70 | norm 0.3921 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11682 | train 2.976437 | val 3.356704 | tok/s 2624982.70 | norm 0.3744 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11683 | train 2.902941 | val 3.356704 | tok/s 2619058.70 | norm 0.3866 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11684 | train 3.010767 | val 3.356704 | tok/s 2628985.23 | norm 0.4148 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11685 | train 3.069728 | val 3.356704 | tok/s 2625573.06 | norm 0.4249 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11686 | train 2.996212 | val 3.356704 | tok/s 2626259.95 | norm 0.3878 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11687 | train 2.964636 | val 3.356704 | tok/s 2623810.43 | norm 0.4034 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11688 | train 3.030792 | val 3.356704 | tok/s 2623687.48 | norm 0.3766 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11689 | train 3.034066 | val 3.356704 | tok/s 2629776.75 | norm 0.4393 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11690 | train 2.995890 | val 3.356704 | tok/s 2627653.26 | norm 0.3614 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11691 | train 2.899787 | val 3.356704 | tok/s 2629857.16 | norm 0.3454 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11692 | train 2.925363 | val 3.356704 | tok/s 2626208.30 | norm 0.4002 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11693 | train 2.868659 | val 3.356704 | tok/s 2624474.81 | norm 0.3838 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11694 | train 2.951281 | val 3.356704 | tok/s 2624626.15 | norm 0.3784 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11695 | train 2.992671 | val 3.356704 | tok/s 2629335.53 | norm 0.3986 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11696 | train 3.011246 | val 3.356704 | tok/s 2624469.75 | norm 0.3998 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11697 | train 3.130872 | val 3.356704 | tok/s 2623631.48 | norm 0.4189 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11698 | train 2.971705 | val 3.356704 | tok/s 2618955.68 | norm 0.4062 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11699 | train 2.961159 | val 3.356704 | tok/s 2609340.28 | norm 0.3898 | dt 200.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11700 | train 3.020660 | val 3.356704 | tok/s 2605312.90 | norm 0.4213 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11701 | train 2.976049 | val 3.356704 | tok/s 2626516.07 | norm 0.3974 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11702 | train 2.947119 | val 3.356704 | tok/s 2624859.27 | norm 0.3723 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11703 | train 2.862156 | val 3.356704 | tok/s 2624670.79 | norm 0.3914 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11704 | train 2.892110 | val 3.356704 | tok/s 2623645.32 | norm 0.5115 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11705 | train 2.985239 | val 3.356704 | tok/s 2532345.27 | norm 0.4081 | dt 207.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11706 | train 2.939435 | val 3.356704 | tok/s 2604985.19 | norm 0.3651 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11707 | train 2.924605 | val 3.356704 | tok/s 2630090.71 | norm 0.4009 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11708 | train 2.965503 | val 3.356704 | tok/s 2629873.61 | norm 0.4006 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11709 | train 2.877074 | val 3.356704 | tok/s 2624888.12 | norm 0.3931 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11710 | train 3.188282 | val 3.356704 | tok/s 2626608.36 | norm 0.4660 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11711 | train 3.076955 | val 3.356704 | tok/s 2629476.75 | norm 0.4023 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11712 | train 3.034753 | val 3.356704 | tok/s 2580275.07 | norm 0.3741 | dt 203.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11713 | train 2.981184 | val 3.356704 | tok/s 2601114.42 | norm 0.3613 | dt 201.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11714 | train 2.986608 | val 3.356704 | tok/s 2629039.98 | norm 0.3641 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11715 | train 2.973478 | val 3.356704 | tok/s 2630738.87 | norm 0.3772 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11716 | train 3.078899 | val 3.356704 | tok/s 2628474.81 | norm 0.3328 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11717 | train 2.908973 | val 3.356704 | tok/s 2628529.30 | norm 0.3611 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11718 | train 3.027389 | val 3.356704 | tok/s 2628508.10 | norm 0.4153 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11719 | train 3.021415 | val 3.356704 | tok/s 2629612.10 | norm 0.4018 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11720 | train 2.925647 | val 3.356704 | tok/s 2629002.07 | norm 0.3814 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11721 | train 2.983713 | val 3.356704 | tok/s 2632620.17 | norm 0.4296 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11722 | train 2.859977 | val 3.356704 | tok/s 2628752.08 | norm 0.4214 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11723 | train 2.966489 | val 3.356704 | tok/s 2629059.50 | norm 0.3302 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11724 | train 2.913632 | val 3.356704 | tok/s 2624839.78 | norm 0.4127 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11725 | train 3.041085 | val 3.356704 | tok/s 2630179.98 | norm 0.3957 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11726 | train 2.964513 | val 3.356704 | tok/s 2632778.59 | norm 0.4129 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11727 | train 2.963120 | val 3.356704 | tok/s 2629525.40 | norm 0.4257 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11728 | train 2.990434 | val 3.356704 | tok/s 2634338.31 | norm 0.3980 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11729 | train 3.023481 | val 3.356704 | tok/s 2626009.15 | norm 0.3918 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11730 | train 2.902003 | val 3.356704 | tok/s 2628970.29 | norm 0.4498 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11731 | train 2.994095 | val 3.356704 | tok/s 2626115.92 | norm 0.4062 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11732 | train 3.078165 | val 3.356704 | tok/s 2626491.39 | norm 0.3709 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11733 | train 3.135951 | val 3.356704 | tok/s 2631131.29 | norm 0.3708 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11734 | train 2.898545 | val 3.356704 | tok/s 2627496.77 | norm 0.3774 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11735 | train 2.939495 | val 3.356704 | tok/s 2628026.76 | norm 0.3872 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11736 | train 2.998186 | val 3.356704 | tok/s 2628293.92 | norm 0.3823 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11737 | train 2.981164 | val 3.356704 | tok/s 2626919.29 | norm 0.4191 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11738 | train 3.061093 | val 3.356704 | tok/s 2632757.99 | norm 0.4710 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11739 | train 2.936943 | val 3.356704 | tok/s 2629552.22 | norm 0.3720 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11740 | train 2.990942 | val 3.356704 | tok/s 2628672.48 | norm 0.3484 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11741 | train 2.991601 | val 3.356704 | tok/s 2632191.83 | norm 0.3663 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11742 | train 2.879569 | val 3.356704 | tok/s 2623968.57 | norm 0.3485 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11743 | train 2.984167 | val 3.356704 | tok/s 2537953.17 | norm 0.3519 | dt 206.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11744 | train 2.981990 | val 3.356704 | tok/s 2626999.37 | norm 0.3815 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11745 | train 2.907512 | val 3.356704 | tok/s 2629521.94 | norm 0.3923 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11746 | train 3.005978 | val 3.356704 | tok/s 2619385.54 | norm 0.3926 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11747 | train 2.985627 | val 3.356704 | tok/s 2628148.67 | norm 0.4101 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11748 | train 3.014867 | val 3.356704 | tok/s 2626304.63 | norm 0.3907 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11749 | train 2.901577 | val 3.356704 | tok/s 2625670.26 | norm 0.4306 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 11750 | val_loss 3.355107 step 11750 | train 2.969028 | val 3.355107 | tok/s 2614545.48 | norm 0.4529 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11751 | train 2.987264 | val 3.355107 | tok/s 2630536.61 | norm 0.3761 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11752 | train 2.899452 | val 3.355107 | tok/s 2630123.77 | norm 0.4146 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11753 | train 2.990316 | val 3.355107 | tok/s 2627611.76 | norm 0.4046 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11754 | train 2.897764 | val 3.355107 | tok/s 2629233.94 | norm 0.3799 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11755 | train 2.985174 | val 3.355107 | tok/s 2627404.05 | norm 0.3912 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11756 | train 3.109252 | val 3.355107 | tok/s 2628371.57 | norm 0.3608 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11757 | train 2.976033 | val 3.355107 | tok/s 2624522.90 | norm 0.3485 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11758 | train 2.990920 | val 3.355107 | tok/s 2624637.54 | norm 0.3481 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11759 | train 2.843657 | val 3.355107 | tok/s 2626393.27 | norm 0.3416 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11760 | train 3.221957 | val 3.355107 | tok/s 2628843.68 | norm 0.4452 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11761 | train 3.033657 | val 3.355107 | tok/s 2629681.55 | norm 0.4173 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11762 | train 3.195863 | val 3.355107 | tok/s 2626794.47 | norm 0.3912 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11763 | train 3.173222 | val 3.355107 | tok/s 2627640.51 | norm 0.3694 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11764 | train 3.051602 | val 3.355107 | tok/s 2633496.11 | norm 0.4072 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11765 | train 2.955467 | val 3.355107 | tok/s 2628221.22 | norm 0.3783 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11766 | train 2.914333 | val 3.355107 | tok/s 2630043.56 | norm 0.3709 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11767 | train 2.937919 | val 3.355107 | tok/s 2620052.00 | norm 0.5221 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11768 | train 3.017858 | val 3.355107 | tok/s 2627515.43 | norm 0.3956 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11769 | train 2.950409 | val 3.355107 | tok/s 2632890.72 | norm 0.3354 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11770 | train 2.827888 | val 3.355107 | tok/s 2631529.64 | norm 0.4239 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11771 | train 3.017617 | val 3.355107 | tok/s 2622324.67 | norm 0.4433 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11772 | train 2.972040 | val 3.355107 | tok/s 2620191.01 | norm 0.4082 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11773 | train 2.986251 | val 3.355107 | tok/s 2619256.17 | norm 0.3643 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11774 | train 3.028867 | val 3.355107 | tok/s 2630798.20 | norm 0.3736 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11775 | train 3.034380 | val 3.355107 | tok/s 2627999.68 | norm 0.3625 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11776 | train 3.012521 | val 3.355107 | tok/s 2621061.18 | norm 0.3471 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11777 | train 3.056078 | val 3.355107 | tok/s 2626774.23 | norm 0.3964 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11778 | train 3.003717 | val 3.355107 | tok/s 2626762.71 | norm 0.4070 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11779 | train 2.833387 | val 3.355107 | tok/s 2630732.02 | norm 0.3657 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11780 | train 3.111094 | val 3.355107 | tok/s 2632650.34 | norm 0.3856 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11781 | train 2.828677 | val 3.355107 | tok/s 2626698.69 | norm 0.3737 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11782 | train 2.863722 | val 3.355107 | tok/s 2623649.43 | norm 0.3998 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11783 | train 3.005023 | val 3.355107 | tok/s 2628586.76 | norm 0.4037 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11784 | train 3.006610 | val 3.355107 | tok/s 2627821.68 | norm 0.3953 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11785 | train 2.940682 | val 3.355107 | tok/s 2629189.89 | norm 0.3905 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11786 | train 3.158764 | val 3.355107 | tok/s 2626980.18 | norm 0.4035 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11787 | train 2.888041 | val 3.355107 | tok/s 2628438.27 | norm 0.4039 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11788 | train 2.873123 | val 3.355107 | tok/s 2629771.12 | norm 0.3975 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11789 | train 2.931324 | val 3.355107 | tok/s 2633630.63 | norm 0.4129 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11790 | train 3.001518 | val 3.355107 | tok/s 2620677.28 | norm 0.3975 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11791 | train 3.044079 | val 3.355107 | tok/s 2623017.09 | norm 0.3663 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11792 | train 3.001345 | val 3.355107 | tok/s 2627825.38 | norm 0.3794 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11793 | train 3.014933 | val 3.355107 | tok/s 2626681.90 | norm 0.3732 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11794 | train 2.950286 | val 3.355107 | tok/s 2629954.02 | norm 0.3714 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11795 | train 3.004844 | val 3.355107 | tok/s 2552915.80 | norm 0.3773 | dt 205.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11796 | train 3.056025 | val 3.355107 | tok/s 2629038.45 | norm 0.3696 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11797 | train 3.067499 | val 3.355107 | tok/s 2613906.37 | norm 0.3820 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11798 | train 3.058500 | val 3.355107 | tok/s 2627340.70 | norm 0.4176 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11799 | train 2.919504 | val 3.355107 | tok/s 2621892.17 | norm 0.3512 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11800 | train 2.957067 | val 3.355107 | tok/s 2623477.22 | norm 0.3674 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11801 | train 2.914620 | val 3.355107 | tok/s 2633408.72 | norm 0.4133 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11802 | train 2.903258 | val 3.355107 | tok/s 2629605.25 | norm 0.4069 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11803 | train 3.062097 | val 3.355107 | tok/s 2631300.80 | norm 0.3924 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11804 | train 2.959060 | val 3.355107 | tok/s 2629355.32 | norm 0.3600 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11805 | train 3.127154 | val 3.355107 | tok/s 2631173.68 | norm 0.3340 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11806 | train 3.000939 | val 3.355107 | tok/s 2629517.84 | norm 0.3396 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11807 | train 3.262701 | val 3.355107 | tok/s 2627305.96 | norm 0.4106 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11808 | train 2.971463 | val 3.355107 | tok/s 2632085.68 | norm 0.3911 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11809 | train 2.948810 | val 3.355107 | tok/s 2628518.05 | norm 0.3808 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11810 | train 3.085694 | val 3.355107 | tok/s 2626479.31 | norm 0.3917 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11811 | train 2.934396 | val 3.355107 | tok/s 2610062.70 | norm 0.3828 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11812 | train 2.843893 | val 3.355107 | tok/s 2627095.19 | norm 0.3675 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11813 | train 3.029302 | val 3.355107 | tok/s 2630686.69 | norm 0.3914 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11814 | train 2.900910 | val 3.355107 | tok/s 2632589.50 | norm 0.3729 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11815 | train 2.906225 | val 3.355107 | tok/s 2627954.91 | norm 0.3594 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11816 | train 2.988477 | val 3.355107 | tok/s 2629332.76 | norm 0.4199 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11817 | train 3.136097 | val 3.355107 | tok/s 2631434.45 | norm 0.4127 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11818 | train 2.932620 | val 3.355107 | tok/s 2630007.00 | norm 0.3693 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11819 | train 2.971433 | val 3.355107 | tok/s 2631839.77 | norm 0.3518 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11820 | train 3.014201 | val 3.355107 | tok/s 2629753.49 | norm 0.3260 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11821 | train 2.985864 | val 3.355107 | tok/s 2491721.12 | norm 0.3673 | dt 210.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11822 | train 2.932891 | val 3.355107 | tok/s 2628283.04 | norm 0.4328 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11823 | train 2.999565 | val 3.355107 | tok/s 2551583.02 | norm 0.4033 | dt 205.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11824 | train 3.001625 | val 3.355107 | tok/s 2624054.45 | norm 0.4356 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11825 | train 2.932982 | val 3.355107 | tok/s 2624699.50 | norm 0.4077 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11826 | train 3.013312 | val 3.355107 | tok/s 2630454.33 | norm 0.3717 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11827 | train 2.900081 | val 3.355107 | tok/s 2629114.37 | norm 0.3822 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11828 | train 2.862509 | val 3.355107 | tok/s 2628313.30 | norm 0.4234 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11829 | train 3.022420 | val 3.355107 | tok/s 2618506.74 | norm 0.4154 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11830 | train 3.044313 | val 3.355107 | tok/s 2625233.54 | norm 0.4159 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11831 | train 3.061834 | val 3.355107 | tok/s 2630244.43 | norm 0.3755 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11832 | train 3.024780 | val 3.355107 | tok/s 2627539.97 | norm 0.3717 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11833 | train 3.027334 | val 3.355107 | tok/s 2626567.14 | norm 0.4013 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11834 | train 2.909399 | val 3.355107 | tok/s 2623872.87 | norm 0.3607 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11835 | train 2.938567 | val 3.355107 | tok/s 2628211.64 | norm 0.4057 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11836 | train 3.072347 | val 3.355107 | tok/s 2629872.84 | norm 0.3956 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11837 | train 2.963025 | val 3.355107 | tok/s 2627346.24 | norm 0.3938 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11838 | train 2.885025 | val 3.355107 | tok/s 2630972.84 | norm 0.3945 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11839 | train 3.111727 | val 3.355107 | tok/s 2628227.77 | norm 0.4081 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11840 | train 3.001610 | val 3.355107 | tok/s 2628330.48 | norm 0.3899 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11841 | train 2.996120 | val 3.355107 | tok/s 2627855.74 | norm 0.3767 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11842 | train 3.090751 | val 3.355107 | tok/s 2628151.58 | norm 0.3861 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11843 | train 2.981627 | val 3.355107 | tok/s 2625101.06 | norm 0.4318 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11844 | train 3.108397 | val 3.355107 | tok/s 2624279.75 | norm 0.3932 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11845 | train 3.054863 | val 3.355107 | tok/s 2630589.23 | norm 0.3763 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11846 | train 2.976892 | val 3.355107 | tok/s 2630577.90 | norm 0.3839 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11847 | train 3.161664 | val 3.355107 | tok/s 2627653.31 | norm 0.4479 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11848 | train 2.932267 | val 3.355107 | tok/s 2627312.33 | norm 0.4195 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11849 | train 3.054287 | val 3.355107 | tok/s 2624933.80 | norm 0.3712 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11850 | train 2.888930 | val 3.355107 | tok/s 2629779.90 | norm 0.3684 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11851 | train 2.935133 | val 3.355107 | tok/s 2621872.03 | norm 0.3988 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11852 | train 3.099276 | val 3.355107 | tok/s 2621973.18 | norm 0.3860 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11853 | train 3.045071 | val 3.355107 | tok/s 2630736.95 | norm 0.3372 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11854 | train 2.954581 | val 3.355107 | tok/s 2629314.35 | norm 0.3563 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11855 | train 2.920493 | val 3.355107 | tok/s 2628904.37 | norm 0.3815 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11856 | train 2.964803 | val 3.355107 | tok/s 2617647.53 | norm 0.4124 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11857 | train 3.168801 | val 3.355107 | tok/s 2621293.19 | norm 0.3771 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11858 | train 2.847233 | val 3.355107 | tok/s 2628775.37 | norm 0.3595 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11859 | train 2.838789 | val 3.355107 | tok/s 2627059.17 | norm 0.3615 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11860 | train 3.018243 | val 3.355107 | tok/s 2629177.20 | norm 0.3744 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11861 | train 3.001076 | val 3.355107 | tok/s 2624308.70 | norm 0.4078 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11862 | train 3.022004 | val 3.355107 | tok/s 2625991.80 | norm 0.3887 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11863 | train 2.946740 | val 3.355107 | tok/s 2629361.75 | norm 0.3691 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11864 | train 3.057982 | val 3.355107 | tok/s 2623989.95 | norm 0.3832 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11865 | train 2.929265 | val 3.355107 | tok/s 2633360.10 | norm 0.3946 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11866 | train 2.983249 | val 3.355107 | tok/s 2626111.71 | norm 0.4507 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11867 | train 3.054916 | val 3.355107 | tok/s 2628656.58 | norm 0.4574 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11868 | train 2.888886 | val 3.355107 | tok/s 2631932.41 | norm 0.4199 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11869 | train 2.965827 | val 3.355107 | tok/s 2626002.67 | norm 0.4140 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11870 | train 2.915044 | val 3.355107 | tok/s 2629602.58 | norm 0.3885 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11871 | train 2.966082 | val 3.355107 | tok/s 2624261.41 | norm 0.3792 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11872 | train 2.871296 | val 3.355107 | tok/s 2488254.65 | norm 0.4063 | dt 210.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11873 | train 3.045578 | val 3.355107 | tok/s 2630398.66 | norm 0.4038 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11874 | train 2.881431 | val 3.355107 | tok/s 2635530.03 | norm 0.4173 | dt 198.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11875 | train 3.051695 | val 3.355107 | tok/s 2629941.88 | norm 0.4262 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11876 | train 2.984356 | val 3.355107 | tok/s 2622871.73 | norm 0.3725 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11877 | train 3.000354 | val 3.355107 | tok/s 2624683.57 | norm 0.3823 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11878 | train 2.889552 | val 3.355107 | tok/s 2631243.96 | norm 0.3700 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11879 | train 2.898036 | val 3.355107 | tok/s 2623242.85 | norm 0.3832 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11880 | train 3.019315 | val 3.355107 | tok/s 2628703.24 | norm 0.4314 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11881 | train 3.080291 | val 3.355107 | tok/s 2624638.91 | norm 0.3806 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11882 | train 2.893423 | val 3.355107 | tok/s 2625191.87 | norm 0.3717 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11883 | train 3.037298 | val 3.355107 | tok/s 2631366.46 | norm 0.4081 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11884 | train 3.034734 | val 3.355107 | tok/s 2627471.27 | norm 0.3995 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11885 | train 3.000450 | val 3.355107 | tok/s 2630604.02 | norm 0.4103 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11886 | train 2.985613 | val 3.355107 | tok/s 2627528.05 | norm 0.3908 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11887 | train 3.179976 | val 3.355107 | tok/s 2629180.95 | norm 0.4155 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11888 | train 2.956890 | val 3.355107 | tok/s 2630491.57 | norm 0.4034 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11889 | train 3.092815 | val 3.355107 | tok/s 2624895.12 | norm 0.3654 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11890 | train 3.100827 | val 3.355107 | tok/s 2629266.29 | norm 0.4005 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11891 | train 2.976454 | val 3.355107 | tok/s 2623264.28 | norm 0.4217 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11892 | train 3.031203 | val 3.355107 | tok/s 2627744.80 | norm 0.3905 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11893 | train 2.948374 | val 3.355107 | tok/s 2628412.53 | norm 0.3914 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11894 | train 2.897358 | val 3.355107 | tok/s 2626224.90 | norm 0.4041 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11895 | train 2.956690 | val 3.355107 | tok/s 2629167.69 | norm 0.4126 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11896 | train 3.028785 | val 3.355107 | tok/s 2631511.03 | norm 0.3851 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11897 | train 3.028942 | val 3.355107 | tok/s 2627735.16 | norm 0.4136 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11898 | train 2.910035 | val 3.355107 | tok/s 2631054.78 | norm 0.4069 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11899 | train 3.385523 | val 3.355107 | tok/s 2620586.57 | norm 0.5316 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11900 | train 2.980462 | val 3.355107 | tok/s 2622535.45 | norm 0.3901 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11901 | train 3.051957 | val 3.355107 | tok/s 2622909.01 | norm 0.3459 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11902 | train 2.878646 | val 3.355107 | tok/s 2625098.61 | norm 0.3506 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11903 | train 3.038491 | val 3.355107 | tok/s 2621677.01 | norm 0.3672 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11904 | train 2.970284 | val 3.355107 | tok/s 2621984.81 | norm 0.3789 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11905 | train 3.055888 | val 3.355107 | tok/s 2626777.77 | norm 0.3390 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11906 | train 2.891011 | val 3.355107 | tok/s 2631026.49 | norm 0.3607 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11907 | train 2.924538 | val 3.355107 | tok/s 2628859.73 | norm 0.3929 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11908 | train 3.050339 | val 3.355107 | tok/s 2624504.82 | norm 0.4105 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11909 | train 2.985813 | val 3.355107 | tok/s 2621090.11 | norm 0.4322 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11910 | train 2.942708 | val 3.355107 | tok/s 2630309.09 | norm 0.3881 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11911 | train 3.056441 | val 3.355107 | tok/s 2628198.91 | norm 0.3816 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11912 | train 3.048411 | val 3.355107 | tok/s 2628653.23 | norm 0.3999 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11913 | train 3.037908 | val 3.355107 | tok/s 2629084.64 | norm 0.4174 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11914 | train 3.078935 | val 3.355107 | tok/s 2628631.79 | norm 0.4468 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11915 | train 3.018310 | val 3.355107 | tok/s 2631058.44 | norm 0.3964 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11916 | train 2.917774 | val 3.355107 | tok/s 2631880.43 | norm 0.3634 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11917 | train 3.008440 | val 3.355107 | tok/s 2630113.97 | norm 0.4125 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11918 | train 3.018124 | val 3.355107 | tok/s 2620704.79 | norm 0.3896 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11919 | train 3.113636 | val 3.355107 | tok/s 2628348.75 | norm 0.4711 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11920 | train 3.088508 | val 3.355107 | tok/s 2629415.64 | norm 0.3894 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11921 | train 2.933751 | val 3.355107 | tok/s 2628454.75 | norm 0.3656 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11922 | train 3.012191 | val 3.355107 | tok/s 2629053.23 | norm 0.4367 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11923 | train 3.039328 | val 3.355107 | tok/s 2625165.79 | norm 0.4542 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11924 | train 2.992865 | val 3.355107 | tok/s 2626888.42 | norm 0.4169 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11925 | train 2.954690 | val 3.355107 | tok/s 2623424.04 | norm 0.4123 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11926 | train 2.995955 | val 3.355107 | tok/s 2629211.46 | norm 0.3905 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11927 | train 2.928779 | val 3.355107 | tok/s 2634910.45 | norm 0.3846 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11928 | train 2.941187 | val 3.355107 | tok/s 2631791.44 | norm 0.3847 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11929 | train 3.092624 | val 3.355107 | tok/s 2632519.87 | norm 0.4111 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11930 | train 2.958052 | val 3.355107 | tok/s 2632470.30 | norm 0.3725 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11931 | train 2.985819 | val 3.355107 | tok/s 2630623.16 | norm 0.3760 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11932 | train 3.119725 | val 3.355107 | tok/s 2520169.56 | norm 0.4274 | dt 208.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11933 | train 3.045372 | val 3.355107 | tok/s 2466547.63 | norm 0.3776 | dt 212.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11934 | train 2.927274 | val 3.355107 | tok/s 2626421.19 | norm 0.3631 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11935 | train 3.075993 | val 3.355107 | tok/s 2629350.49 | norm 0.3926 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11936 | train 3.088188 | val 3.355107 | tok/s 2631160.92 | norm 0.3554 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11937 | train 3.008251 | val 3.355107 | tok/s 2633081.97 | norm 0.3816 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11938 | train 2.986044 | val 3.355107 | tok/s 2629858.52 | norm 0.3922 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11939 | train 3.036964 | val 3.355107 | tok/s 2588293.95 | norm 0.4253 | dt 202.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11940 | train 3.114995 | val 3.355107 | tok/s 2584045.69 | norm 0.4163 | dt 202.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11941 | train 2.991268 | val 3.355107 | tok/s 2633944.63 | norm 0.3620 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11942 | train 2.915282 | val 3.355107 | tok/s 2628685.47 | norm 0.3936 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11943 | train 2.935165 | val 3.355107 | tok/s 2629578.91 | norm 0.4118 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11944 | train 3.039382 | val 3.355107 | tok/s 2631652.39 | norm 0.3667 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11945 | train 3.050038 | val 3.355107 | tok/s 2630757.31 | norm 0.3869 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11946 | train 3.145301 | val 3.355107 | tok/s 2618814.59 | norm 0.4427 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11947 | train 2.992609 | val 3.355107 | tok/s 2628533.41 | norm 0.4104 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11948 | train 2.957006 | val 3.355107 | tok/s 2626142.08 | norm 0.3895 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11949 | train 2.856892 | val 3.355107 | tok/s 2625155.63 | norm 0.4270 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11950 | train 3.013522 | val 3.355107 | tok/s 2629900.61 | norm 0.6053 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11951 | train 3.053962 | val 3.355107 | tok/s 2629824.63 | norm 0.4437 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11952 | train 3.048487 | val 3.355107 | tok/s 2632266.83 | norm 0.3758 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11953 | train 3.047648 | val 3.355107 | tok/s 2634834.76 | norm 0.3257 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11954 | train 2.934550 | val 3.355107 | tok/s 2624246.84 | norm 0.3367 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11955 | train 3.034208 | val 3.355107 | tok/s 2631757.37 | norm 0.3732 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11956 | train 2.938195 | val 3.355107 | tok/s 2628659.89 | norm 0.3796 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11957 | train 2.934041 | val 3.355107 | tok/s 2631350.51 | norm 0.3650 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11958 | train 2.847575 | val 3.355107 | tok/s 2629812.30 | norm 0.4337 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11959 | train 2.977396 | val 3.355107 | tok/s 2631965.96 | norm 0.4165 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11960 | train 3.035228 | val 3.355107 | tok/s 2634614.44 | norm 0.3581 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11961 | train 2.922841 | val 3.355107 | tok/s 2628561.35 | norm 0.3580 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11962 | train 3.084096 | val 3.355107 | tok/s 2630859.52 | norm 0.3951 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11963 | train 3.135012 | val 3.355107 | tok/s 2627065.68 | norm 0.4129 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11964 | train 3.025152 | val 3.355107 | tok/s 2629948.42 | norm 0.3883 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11965 | train 2.994410 | val 3.355107 | tok/s 2632282.23 | norm 0.3493 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11966 | train 2.954482 | val 3.355107 | tok/s 2630498.49 | norm 0.3534 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11967 | train 2.964835 | val 3.355107 | tok/s 2632639.35 | norm 0.3645 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11968 | train 3.037002 | val 3.355107 | tok/s 2633231.11 | norm 0.3738 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11969 | train 3.054282 | val 3.355107 | tok/s 2633998.40 | norm 0.3993 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11970 | train 3.016058 | val 3.355107 | tok/s 2630999.88 | norm 0.3935 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11971 | train 2.950243 | val 3.355107 | tok/s 2630840.85 | norm 0.3784 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11972 | train 2.979788 | val 3.355107 | tok/s 2630586.73 | norm 0.3830 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11973 | train 2.932058 | val 3.355107 | tok/s 2630266.28 | norm 0.4217 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11974 | train 2.856718 | val 3.355107 | tok/s 2628278.40 | norm 0.4158 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11975 | train 2.838023 | val 3.355107 | tok/s 2625823.29 | norm 0.4109 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11976 | train 2.991511 | val 3.355107 | tok/s 2631570.96 | norm 0.3880 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11977 | train 3.056279 | val 3.355107 | tok/s 2628306.54 | norm 0.3736 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11978 | train 2.978746 | val 3.355107 | tok/s 2629134.57 | norm 0.3957 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11979 | train 3.121597 | val 3.355107 | tok/s 2629245.63 | norm 0.3894 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11980 | train 2.976261 | val 3.355107 | tok/s 2627259.33 | norm 0.3994 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11981 | train 3.085943 | val 3.355107 | tok/s 2630263.14 | norm 0.4202 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11982 | train 3.041946 | val 3.355107 | tok/s 2594691.20 | norm 0.4021 | dt 202.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11983 | train 2.983302 | val 3.355107 | tok/s 2630411.87 | norm 0.3975 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11984 | train 3.100961 | val 3.355107 | tok/s 2633507.18 | norm 0.3902 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11985 | train 2.993622 | val 3.355107 | tok/s 2629742.02 | norm 0.4110 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11986 | train 3.078173 | val 3.355107 | tok/s 2632324.85 | norm 0.4014 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11987 | train 2.950750 | val 3.355107 | tok/s 2628400.98 | norm 0.3896 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11988 | train 2.965683 | val 3.355107 | tok/s 2568682.81 | norm 0.4216 | dt 204.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11989 | train 3.053823 | val 3.355107 | tok/s 2469001.47 | norm 0.4379 | dt 212.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11990 | train 2.898639 | val 3.355107 | tok/s 2627988.04 | norm 0.3911 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11991 | train 3.161895 | val 3.355107 | tok/s 2631240.24 | norm 0.4074 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11992 | train 3.002028 | val 3.355107 | tok/s 2629561.81 | norm 0.4136 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11993 | train 3.057898 | val 3.355107 | tok/s 2631474.62 | norm 0.3754 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11994 | train 2.996780 | val 3.355107 | tok/s 2619724.87 | norm 0.3666 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11995 | train 2.949217 | val 3.355107 | tok/s 2630367.10 | norm 0.3851 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11996 | train 3.012723 | val 3.355107 | tok/s 2630602.44 | norm 0.3610 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11997 | train 3.001751 | val 3.355107 | tok/s 2626333.08 | norm 0.3572 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11998 | train 3.114604 | val 3.355107 | tok/s 2627572.78 | norm 0.3762 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 11999 | train 3.039318 | val 3.355107 | tok/s 2626245.12 | norm 0.3624 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 12000 | val_loss 3.346371 step 12000 | train 3.009441 | val 3.346371 | tok/s 2605016.54 | norm 0.3706 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12001 | train 3.057727 | val 3.346371 | tok/s 2629524.62 | norm 0.3920 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12002 | train 2.851798 | val 3.346371 | tok/s 2629215.00 | norm 0.4045 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12003 | train 3.030174 | val 3.346371 | tok/s 2627239.41 | norm 0.4180 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12004 | train 2.931068 | val 3.346371 | tok/s 2625520.49 | norm 0.4365 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12005 | train 3.002241 | val 3.346371 | tok/s 2629414.91 | norm 0.4311 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12006 | train 3.148357 | val 3.346371 | tok/s 2628364.14 | norm 0.3887 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12007 | train 2.942931 | val 3.346371 | tok/s 2625099.13 | norm 0.4132 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12008 | train 2.941791 | val 3.346371 | tok/s 2627753.78 | norm 0.3996 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12009 | train 2.974002 | val 3.346371 | tok/s 2632344.86 | norm 0.4008 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12010 | train 2.978953 | val 3.346371 | tok/s 2616012.31 | norm 0.3897 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12011 | train 2.927189 | val 3.346371 | tok/s 2554432.78 | norm 0.3647 | dt 205.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12012 | train 3.011889 | val 3.346371 | tok/s 2599110.79 | norm 0.3597 | dt 201.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12013 | train 3.069828 | val 3.346371 | tok/s 2629916.39 | norm 0.3872 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12014 | train 3.023482 | val 3.346371 | tok/s 2631786.45 | norm 0.3635 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12015 | train 2.921275 | val 3.346371 | tok/s 2631701.37 | norm 0.3559 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12016 | train 2.955756 | val 3.346371 | tok/s 2632549.13 | norm 0.3714 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12017 | train 3.053956 | val 3.346371 | tok/s 2629502.94 | norm 0.3742 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12018 | train 3.353673 | val 3.346371 | tok/s 2628640.08 | norm 0.4683 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12019 | train 2.992881 | val 3.346371 | tok/s 2629853.23 | norm 0.3701 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12020 | train 3.088897 | val 3.346371 | tok/s 2632960.66 | norm 0.3607 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12021 | train 2.840613 | val 3.346371 | tok/s 2626975.21 | norm 0.3778 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12022 | train 3.026546 | val 3.346371 | tok/s 2630065.71 | norm 0.3779 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12023 | train 2.921668 | val 3.346371 | tok/s 2630191.46 | norm 0.3864 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12024 | train 2.944340 | val 3.346371 | tok/s 2610392.97 | norm 0.3745 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12025 | train 2.928807 | val 3.346371 | tok/s 2626582.14 | norm 0.3701 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12026 | train 3.052960 | val 3.346371 | tok/s 2626236.03 | norm 0.4030 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12027 | train 2.950421 | val 3.346371 | tok/s 2542605.31 | norm 0.3849 | dt 206.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12028 | train 2.969833 | val 3.346371 | tok/s 2604588.56 | norm 0.4058 | dt 201.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12029 | train 2.808730 | val 3.346371 | tok/s 2631198.77 | norm 0.4822 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12030 | train 2.910185 | val 3.346371 | tok/s 2632715.94 | norm 0.4187 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12031 | train 2.961594 | val 3.346371 | tok/s 2632540.88 | norm 0.3654 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12032 | train 2.916522 | val 3.346371 | tok/s 2627590.84 | norm 0.3971 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12033 | train 2.976140 | val 3.346371 | tok/s 2628517.52 | norm 0.3878 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12034 | train 2.990558 | val 3.346371 | tok/s 2628953.02 | norm 0.3618 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12035 | train 3.035328 | val 3.346371 | tok/s 2631879.65 | norm 0.4457 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12036 | train 3.295441 | val 3.346371 | tok/s 2630208.81 | norm 0.4607 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12037 | train 2.996112 | val 3.346371 | tok/s 2631979.09 | norm 0.3632 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12038 | train 2.932983 | val 3.346371 | tok/s 2630824.37 | norm 0.3128 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12039 | train 2.947529 | val 3.346371 | tok/s 2628769.49 | norm 0.3329 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12040 | train 2.945771 | val 3.346371 | tok/s 2629272.60 | norm 0.3778 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12041 | train 3.085625 | val 3.346371 | tok/s 2626381.02 | norm 0.4110 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12042 | train 2.899474 | val 3.346371 | tok/s 2631715.22 | norm 0.3758 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12043 | train 2.970112 | val 3.346371 | tok/s 2623628.07 | norm 0.3690 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12044 | train 2.940648 | val 3.346371 | tok/s 2628079.55 | norm 0.4282 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12045 | train 3.041753 | val 3.346371 | tok/s 2632056.24 | norm 0.4903 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12046 | train 2.908012 | val 3.346371 | tok/s 2632780.66 | norm 0.3801 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12047 | train 2.984559 | val 3.346371 | tok/s 2631203.37 | norm 0.3845 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12048 | train 2.928385 | val 3.346371 | tok/s 2627728.64 | norm 0.3950 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12049 | train 2.970057 | val 3.346371 | tok/s 2599857.39 | norm 0.3818 | dt 201.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12050 | train 3.027516 | val 3.346371 | tok/s 2633610.89 | norm 0.3657 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12051 | train 2.964109 | val 3.346371 | tok/s 2628013.98 | norm 0.3565 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12052 | train 3.088280 | val 3.346371 | tok/s 2628452.22 | norm 0.4196 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12053 | train 2.985881 | val 3.346371 | tok/s 2630937.78 | norm 0.4189 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12054 | train 2.988428 | val 3.346371 | tok/s 2626913.57 | norm 0.4275 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12055 | train 3.001960 | val 3.346371 | tok/s 2629041.64 | norm 0.3841 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12056 | train 2.911796 | val 3.346371 | tok/s 2623036.06 | norm 0.3608 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12057 | train 2.981308 | val 3.346371 | tok/s 2627863.05 | norm 0.3781 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12058 | train 3.024552 | val 3.346371 | tok/s 2629964.08 | norm 0.3933 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12059 | train 3.009124 | val 3.346371 | tok/s 2628621.68 | norm 0.3912 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12060 | train 2.897399 | val 3.346371 | tok/s 2627670.68 | norm 0.3850 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12061 | train 2.966898 | val 3.346371 | tok/s 2628705.51 | norm 0.3546 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12062 | train 2.998567 | val 3.346371 | tok/s 2628660.70 | norm 0.3560 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12063 | train 2.882846 | val 3.346371 | tok/s 2626182.25 | norm 0.3777 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12064 | train 3.196176 | val 3.346371 | tok/s 2628732.20 | norm 0.4112 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12065 | train 2.903818 | val 3.346371 | tok/s 2629869.22 | norm 0.4025 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12066 | train 3.036003 | val 3.346371 | tok/s 2630357.35 | norm 0.3603 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12067 | train 2.924958 | val 3.346371 | tok/s 2631876.15 | norm 0.3806 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12068 | train 2.877956 | val 3.346371 | tok/s 2627926.66 | norm 0.4145 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12069 | train 2.971303 | val 3.346371 | tok/s 2630740.83 | norm 0.3836 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12070 | train 3.105537 | val 3.346371 | tok/s 2633905.42 | norm 0.3986 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12071 | train 2.861379 | val 3.346371 | tok/s 2629798.25 | norm 0.4180 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12072 | train 2.934834 | val 3.346371 | tok/s 2631221.24 | norm 0.3983 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12073 | train 2.955776 | val 3.346371 | tok/s 2628643.57 | norm 0.4900 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12074 | train 2.943021 | val 3.346371 | tok/s 2632660.40 | norm 0.3726 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12075 | train 2.980645 | val 3.346371 | tok/s 2631221.25 | norm 0.3794 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12076 | train 3.027076 | val 3.346371 | tok/s 2627760.30 | norm 0.4085 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12077 | train 2.908312 | val 3.346371 | tok/s 2628514.77 | norm 0.3896 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12078 | train 2.893368 | val 3.346371 | tok/s 2632249.98 | norm 0.3829 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12079 | train 2.892914 | val 3.346371 | tok/s 2630702.77 | norm 0.4049 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12080 | train 3.056453 | val 3.346371 | tok/s 2628940.49 | norm 0.3804 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12081 | train 2.990526 | val 3.346371 | tok/s 2631775.03 | norm 0.3974 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12082 | train 2.879185 | val 3.346371 | tok/s 2632302.05 | norm 0.4122 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12083 | train 2.880495 | val 3.346371 | tok/s 2629890.03 | norm 0.4295 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12084 | train 3.041171 | val 3.346371 | tok/s 2634082.12 | norm 0.4161 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12085 | train 2.985921 | val 3.346371 | tok/s 2634827.21 | norm 0.4272 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12086 | train 3.041188 | val 3.346371 | tok/s 2636535.80 | norm 0.4117 | dt 198.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12087 | train 3.066095 | val 3.346371 | tok/s 2626513.91 | norm 0.3756 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12088 | train 3.024165 | val 3.346371 | tok/s 2633936.33 | norm 0.3825 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12089 | train 2.975780 | val 3.346371 | tok/s 2630622.73 | norm 0.3540 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12090 | train 2.990616 | val 3.346371 | tok/s 2630961.75 | norm 0.3223 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12091 | train 3.223370 | val 3.346371 | tok/s 2626998.41 | norm 0.3285 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12092 | train 3.035200 | val 3.346371 | tok/s 2626969.47 | norm 0.3659 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12093 | train 3.019206 | val 3.346371 | tok/s 2630154.12 | norm 0.3783 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12094 | train 2.874804 | val 3.346371 | tok/s 2626632.60 | norm 0.3709 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12095 | train 3.011476 | val 3.346371 | tok/s 2627077.81 | norm 0.4405 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12096 | train 2.889866 | val 3.346371 | tok/s 2627246.94 | norm 0.3900 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12097 | train 2.940811 | val 3.346371 | tok/s 2631085.10 | norm 0.3779 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12098 | train 2.985546 | val 3.346371 | tok/s 2633736.07 | norm 0.4152 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12099 | train 2.949087 | val 3.346371 | tok/s 2624337.32 | norm 0.3679 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12100 | train 3.053728 | val 3.346371 | tok/s 2630316.71 | norm 0.3768 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12101 | train 2.998997 | val 3.346371 | tok/s 2630778.40 | norm 0.3863 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12102 | train 2.947645 | val 3.346371 | tok/s 2630445.38 | norm 0.3623 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12103 | train 3.038479 | val 3.346371 | tok/s 2632466.32 | norm 0.3607 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12104 | train 3.001487 | val 3.346371 | tok/s 2632764.47 | norm 0.3865 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12105 | train 3.059968 | val 3.346371 | tok/s 2493221.16 | norm 0.3719 | dt 210.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12106 | train 2.921526 | val 3.346371 | tok/s 2484080.38 | norm 0.3658 | dt 211.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12107 | train 2.916779 | val 3.346371 | tok/s 2628753.84 | norm 0.3672 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12108 | train 2.973917 | val 3.346371 | tok/s 2629936.88 | norm 0.3923 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12109 | train 3.017141 | val 3.346371 | tok/s 2631154.79 | norm 0.3855 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12110 | train 2.977174 | val 3.346371 | tok/s 2614951.09 | norm 0.3947 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12111 | train 3.059196 | val 3.346371 | tok/s 2631114.04 | norm 0.4115 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12112 | train 2.923851 | val 3.346371 | tok/s 2633003.37 | norm 0.3868 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12113 | train 2.966748 | val 3.346371 | tok/s 2629239.60 | norm 0.3852 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12114 | train 2.910371 | val 3.346371 | tok/s 2629625.25 | norm 0.4055 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12115 | train 2.950542 | val 3.346371 | tok/s 2631253.43 | norm 0.3840 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12116 | train 2.971561 | val 3.346371 | tok/s 2628692.62 | norm 0.3711 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12117 | train 2.894942 | val 3.346371 | tok/s 2634934.41 | norm 0.3743 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12118 | train 3.000094 | val 3.346371 | tok/s 2629460.85 | norm 0.3963 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12119 | train 3.014493 | val 3.346371 | tok/s 2632688.31 | norm 0.4039 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12120 | train 3.028675 | val 3.346371 | tok/s 2628824.18 | norm 0.4111 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12121 | train 2.936392 | val 3.346371 | tok/s 2633847.30 | norm 0.4218 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12122 | train 3.120423 | val 3.346371 | tok/s 2623681.90 | norm 0.4009 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12123 | train 3.023382 | val 3.346371 | tok/s 2630644.22 | norm 0.4099 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12124 | train 3.038001 | val 3.346371 | tok/s 2631253.33 | norm 0.3693 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12125 | train 3.016712 | val 3.346371 | tok/s 2631259.48 | norm 0.3578 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12126 | train 2.967777 | val 3.346371 | tok/s 2628215.09 | norm 0.3811 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12127 | train 3.173295 | val 3.346371 | tok/s 2631779.27 | norm 0.4660 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12128 | train 2.941224 | val 3.346371 | tok/s 2630411.89 | norm 0.3415 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12129 | train 2.936629 | val 3.346371 | tok/s 2633775.95 | norm 0.3365 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12130 | train 2.926849 | val 3.346371 | tok/s 2631656.01 | norm 0.3973 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12131 | train 2.999327 | val 3.346371 | tok/s 2632056.55 | norm 0.4102 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12132 | train 2.960469 | val 3.346371 | tok/s 2625593.31 | norm 0.4060 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12133 | train 2.989343 | val 3.346371 | tok/s 2630305.77 | norm 0.3830 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12134 | train 3.003379 | val 3.346371 | tok/s 2631718.98 | norm 0.4107 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12135 | train 2.872981 | val 3.346371 | tok/s 2629466.63 | norm 0.4091 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12136 | train 2.939991 | val 3.346371 | tok/s 2629070.72 | norm 0.4055 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12137 | train 2.969046 | val 3.346371 | tok/s 2628719.63 | norm 0.4110 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12138 | train 3.113751 | val 3.346371 | tok/s 2631945.40 | norm 0.3850 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12139 | train 3.014248 | val 3.346371 | tok/s 2630722.76 | norm 0.4058 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12140 | train 3.001999 | val 3.346371 | tok/s 2628495.03 | norm 0.4095 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12141 | train 3.012506 | val 3.346371 | tok/s 2633682.99 | norm 0.3684 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12142 | train 3.002235 | val 3.346371 | tok/s 2630861.19 | norm 0.3602 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12143 | train 2.903282 | val 3.346371 | tok/s 2628438.49 | norm 0.3727 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12144 | train 2.887478 | val 3.346371 | tok/s 2632164.61 | norm 0.3756 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12145 | train 2.956474 | val 3.346371 | tok/s 2633136.11 | norm 0.3796 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12146 | train 3.112367 | val 3.346371 | tok/s 2631463.84 | norm 0.4212 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12147 | train 2.950603 | val 3.346371 | tok/s 2632283.12 | norm 0.3808 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12148 | train 3.120898 | val 3.346371 | tok/s 2631590.45 | norm 0.3406 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12149 | train 3.248076 | val 3.346371 | tok/s 2631339.61 | norm 0.5553 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12150 | train 2.886662 | val 3.346371 | tok/s 2627787.20 | norm 0.4969 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12151 | train 2.948446 | val 3.346371 | tok/s 2628157.06 | norm 0.3693 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12152 | train 3.024713 | val 3.346371 | tok/s 2634244.07 | norm 0.4088 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12153 | train 2.924804 | val 3.346371 | tok/s 2631411.62 | norm 0.4158 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12154 | train 2.820738 | val 3.346371 | tok/s 2632099.99 | norm 0.4626 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12155 | train 3.073861 | val 3.346371 | tok/s 2632572.94 | norm 0.4135 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12156 | train 2.908928 | val 3.346371 | tok/s 2633105.67 | norm 0.3915 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12157 | train 2.929439 | val 3.346371 | tok/s 2631171.45 | norm 0.4160 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12158 | train 3.060208 | val 3.346371 | tok/s 2631730.30 | norm 0.3957 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12159 | train 3.053350 | val 3.346371 | tok/s 2631210.00 | norm 0.3720 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12160 | train 3.016162 | val 3.346371 | tok/s 2631629.31 | norm 0.3591 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12161 | train 2.944980 | val 3.346371 | tok/s 2632076.99 | norm 0.3710 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12162 | train 3.090233 | val 3.346371 | tok/s 2633388.90 | norm 0.3779 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12163 | train 2.830620 | val 3.346371 | tok/s 2632495.39 | norm 0.3564 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12164 | train 3.110843 | val 3.346371 | tok/s 2626430.14 | norm 0.3741 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12165 | train 3.105360 | val 3.346371 | tok/s 2629941.63 | norm 0.4153 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12166 | train 2.993714 | val 3.346371 | tok/s 2631903.57 | norm 0.4274 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12167 | train 2.957100 | val 3.346371 | tok/s 2633837.35 | norm 0.3987 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12168 | train 3.047226 | val 3.346371 | tok/s 2635794.68 | norm 0.3859 | dt 198.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12169 | train 2.974859 | val 3.346371 | tok/s 2629381.84 | norm 0.3598 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12170 | train 3.033569 | val 3.346371 | tok/s 2633164.44 | norm 0.3554 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12171 | train 2.924877 | val 3.346371 | tok/s 2630028.94 | norm 0.3604 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12172 | train 2.963019 | val 3.346371 | tok/s 2633966.94 | norm 0.3880 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12173 | train 3.049766 | val 3.346371 | tok/s 2632944.03 | norm 0.3891 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12174 | train 2.988103 | val 3.346371 | tok/s 2632984.20 | norm 0.3797 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12175 | train 2.807364 | val 3.346371 | tok/s 2519269.88 | norm 0.3886 | dt 208.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12176 | train 2.957735 | val 3.346371 | tok/s 2619284.46 | norm 0.3880 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12177 | train 2.997182 | val 3.346371 | tok/s 2629203.74 | norm 0.3857 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12178 | train 2.982637 | val 3.346371 | tok/s 2634215.40 | norm 0.4277 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12179 | train 2.951644 | val 3.346371 | tok/s 2628916.24 | norm 0.4465 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12180 | train 2.913603 | val 3.346371 | tok/s 2631853.34 | norm 0.4064 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12181 | train 2.947212 | val 3.346371 | tok/s 2630911.89 | norm 0.3870 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12182 | train 3.041984 | val 3.346371 | tok/s 2631871.59 | norm 0.3886 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12183 | train 2.979749 | val 3.346371 | tok/s 2627728.30 | norm 0.3952 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12184 | train 3.047632 | val 3.346371 | tok/s 2625927.57 | norm 0.4091 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12185 | train 3.109162 | val 3.346371 | tok/s 2628065.06 | norm 0.4088 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12186 | train 3.005991 | val 3.346371 | tok/s 2630578.87 | norm 0.3770 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12187 | train 2.917787 | val 3.346371 | tok/s 2632928.71 | norm 0.4126 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12188 | train 2.970711 | val 3.346371 | tok/s 2628353.34 | norm 0.4007 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12189 | train 2.906642 | val 3.346371 | tok/s 2631557.40 | norm 0.4073 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12190 | train 3.037020 | val 3.346371 | tok/s 2625306.79 | norm 0.4305 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12191 | train 2.998949 | val 3.346371 | tok/s 2633823.16 | norm 0.4302 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12192 | train 2.940945 | val 3.346371 | tok/s 2625814.63 | norm 0.3905 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12193 | train 3.072294 | val 3.346371 | tok/s 2630474.89 | norm 0.3640 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12194 | train 2.942110 | val 3.346371 | tok/s 2630158.20 | norm 0.3761 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12195 | train 2.922352 | val 3.346371 | tok/s 2631461.46 | norm 0.3649 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12196 | train 3.001644 | val 3.346371 | tok/s 2630430.39 | norm 0.4041 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12197 | train 2.862940 | val 3.346371 | tok/s 2631242.96 | norm 0.4144 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12198 | train 3.006541 | val 3.346371 | tok/s 2627907.41 | norm 0.4693 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12199 | train 2.897271 | val 3.346371 | tok/s 2625114.98 | norm 0.4341 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12200 | train 3.230161 | val 3.346371 | tok/s 2630992.33 | norm 0.3984 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12201 | train 2.941156 | val 3.346371 | tok/s 2629700.68 | norm 0.3858 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12202 | train 3.209305 | val 3.346371 | tok/s 2631090.97 | norm 0.4047 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12203 | train 3.010714 | val 3.346371 | tok/s 2631721.50 | norm 0.4169 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12204 | train 3.011557 | val 3.346371 | tok/s 2632110.25 | norm 0.4067 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12205 | train 3.023624 | val 3.346371 | tok/s 2631010.68 | norm 0.3855 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12206 | train 2.923342 | val 3.346371 | tok/s 2627911.72 | norm 0.3968 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12207 | train 3.121095 | val 3.346371 | tok/s 2627426.13 | norm 0.3900 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12208 | train 3.046219 | val 3.346371 | tok/s 2630833.34 | norm 0.3861 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12209 | train 2.994201 | val 3.346371 | tok/s 2627237.18 | norm 0.3777 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12210 | train 3.026868 | val 3.346371 | tok/s 2629977.18 | norm 0.3747 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12211 | train 2.864453 | val 3.346371 | tok/s 2625041.83 | norm 0.4009 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12212 | train 3.111196 | val 3.346371 | tok/s 2630713.79 | norm 0.3852 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12213 | train 2.885569 | val 3.346371 | tok/s 2630625.31 | norm 0.3730 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12214 | train 3.089501 | val 3.346371 | tok/s 2628121.88 | norm 0.4384 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12215 | train 2.837367 | val 3.346371 | tok/s 2634103.66 | norm 0.3715 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12216 | train 2.954076 | val 3.346371 | tok/s 2619093.99 | norm 0.3306 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12217 | train 3.014818 | val 3.346371 | tok/s 2634319.04 | norm 0.3284 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12218 | train 2.882536 | val 3.346371 | tok/s 2627151.32 | norm 0.3389 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12219 | train 2.921819 | val 3.346371 | tok/s 2631225.96 | norm 0.3639 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12220 | train 3.020088 | val 3.346371 | tok/s 2629838.51 | norm 0.3400 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12221 | train 2.954901 | val 3.346371 | tok/s 2631465.95 | norm 0.3521 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12222 | train 3.222723 | val 3.346371 | tok/s 2632633.81 | norm 0.4178 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12223 | train 3.095592 | val 3.346371 | tok/s 2633244.16 | norm 0.4715 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12224 | train 3.024236 | val 3.346371 | tok/s 2589781.65 | norm 0.3732 | dt 202.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12225 | train 3.085796 | val 3.346371 | tok/s 2633149.48 | norm 0.3634 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12226 | train 3.052291 | val 3.346371 | tok/s 2627951.50 | norm 0.3353 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12227 | train 3.053035 | val 3.346371 | tok/s 2633193.95 | norm 0.3959 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12228 | train 3.027779 | val 3.346371 | tok/s 2631120.59 | norm 0.4412 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12229 | train 3.002908 | val 3.346371 | tok/s 2634273.44 | norm 0.3816 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12230 | train 2.947874 | val 3.346371 | tok/s 2522021.97 | norm 0.3830 | dt 207.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12231 | train 2.984430 | val 3.346371 | tok/s 2587388.71 | norm 0.3880 | dt 202.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12232 | train 2.995270 | val 3.346371 | tok/s 2628805.68 | norm 0.3569 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12233 | train 3.008258 | val 3.346371 | tok/s 2633768.12 | norm 0.3662 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12234 | train 3.037116 | val 3.346371 | tok/s 2632719.36 | norm 0.3656 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12235 | train 2.928551 | val 3.346371 | tok/s 2632642.64 | norm 0.3644 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12236 | train 2.936488 | val 3.346371 | tok/s 2632388.74 | norm 0.3729 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12237 | train 3.049038 | val 3.346371 | tok/s 2628322.97 | norm 0.3724 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12238 | train 3.023214 | val 3.346371 | tok/s 2631166.50 | norm 0.4514 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12239 | train 2.952237 | val 3.346371 | tok/s 2630791.97 | norm 0.3959 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12240 | train 2.910305 | val 3.346371 | tok/s 2627484.84 | norm 0.3622 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12241 | train 2.949643 | val 3.346371 | tok/s 2628371.20 | norm 0.3829 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12242 | train 2.928353 | val 3.346371 | tok/s 2633402.51 | norm 0.3709 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12243 | train 2.951679 | val 3.346371 | tok/s 2633703.45 | norm 0.3856 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12244 | train 3.109343 | val 3.346371 | tok/s 2627712.01 | norm 0.3877 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12245 | train 2.976596 | val 3.346371 | tok/s 2630653.85 | norm 0.3466 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12246 | train 2.979949 | val 3.346371 | tok/s 2632796.46 | norm 0.3726 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12247 | train 3.054869 | val 3.346371 | tok/s 2632253.77 | norm 0.4399 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12248 | train 2.941966 | val 3.346371 | tok/s 2620212.00 | norm 0.4161 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12249 | train 3.008651 | val 3.346371 | tok/s 2625928.38 | norm 0.4004 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 12250 | val_loss 3.348749 step 12250 | train 3.050533 | val 3.348749 | tok/s 2610611.00 | norm 0.4028 | dt 200.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12251 | train 2.933107 | val 3.348749 | tok/s 2626355.55 | norm 0.3831 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12252 | train 3.114576 | val 3.348749 | tok/s 2627520.58 | norm 0.3984 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12253 | train 3.029645 | val 3.348749 | tok/s 2630183.16 | norm 0.3592 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12254 | train 3.011465 | val 3.348749 | tok/s 2630279.09 | norm 0.3469 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12255 | train 2.974880 | val 3.348749 | tok/s 2628757.48 | norm 0.3727 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12256 | train 3.018820 | val 3.348749 | tok/s 2630869.09 | norm 0.3802 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12257 | train 3.081408 | val 3.348749 | tok/s 2631594.45 | norm 0.4475 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12258 | train 2.946105 | val 3.348749 | tok/s 2632402.97 | norm 0.4300 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12259 | train 2.973745 | val 3.348749 | tok/s 2630560.85 | norm 0.3868 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12260 | train 3.053417 | val 3.348749 | tok/s 2628437.10 | norm 0.3916 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12261 | train 2.995425 | val 3.348749 | tok/s 2629761.49 | norm 0.4051 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12262 | train 3.036541 | val 3.348749 | tok/s 2629705.61 | norm 0.4030 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12263 | train 2.866098 | val 3.348749 | tok/s 2630845.54 | norm 0.3769 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12264 | train 2.904379 | val 3.348749 | tok/s 2632792.97 | norm 0.3191 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12265 | train 3.071659 | val 3.348749 | tok/s 2627600.73 | norm 0.3685 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12266 | train 2.856651 | val 3.348749 | tok/s 2624097.50 | norm 0.4266 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12267 | train 2.969134 | val 3.348749 | tok/s 2631351.62 | norm 0.4257 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12268 | train 3.079282 | val 3.348749 | tok/s 2626330.48 | norm 0.4321 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12269 | train 2.939090 | val 3.348749 | tok/s 2625816.41 | norm 0.4057 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12270 | train 3.030960 | val 3.348749 | tok/s 2631027.19 | norm 0.4008 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12271 | train 3.017049 | val 3.348749 | tok/s 2631223.55 | norm 0.4426 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12272 | train 2.992229 | val 3.348749 | tok/s 2629154.48 | norm 0.4190 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12273 | train 3.038645 | val 3.348749 | tok/s 2629838.68 | norm 0.3742 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12274 | train 2.817357 | val 3.348749 | tok/s 2630087.99 | norm 0.3729 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12275 | train 2.922497 | val 3.348749 | tok/s 2630127.61 | norm 0.3998 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12276 | train 3.012358 | val 3.348749 | tok/s 2626916.29 | norm 0.4193 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12277 | train 2.860733 | val 3.348749 | tok/s 2631239.92 | norm 0.3512 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12278 | train 2.901346 | val 3.348749 | tok/s 2632760.55 | norm 0.3737 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12279 | train 2.873059 | val 3.348749 | tok/s 2629662.38 | norm 0.3734 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12280 | train 2.964285 | val 3.348749 | tok/s 2627745.04 | norm 0.3714 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12281 | train 3.000757 | val 3.348749 | tok/s 2632180.02 | norm 0.3471 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12282 | train 3.126860 | val 3.348749 | tok/s 2630991.41 | norm 0.3576 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12283 | train 2.961017 | val 3.348749 | tok/s 2631406.95 | norm 0.3571 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12284 | train 3.103263 | val 3.348749 | tok/s 2633448.43 | norm 0.4330 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12285 | train 2.953297 | val 3.348749 | tok/s 2634374.57 | norm 0.4369 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12286 | train 2.894667 | val 3.348749 | tok/s 2630891.13 | norm 0.4354 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12287 | train 2.949887 | val 3.348749 | tok/s 2628610.75 | norm 0.4202 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12288 | train 3.034549 | val 3.348749 | tok/s 2632901.77 | norm 0.3900 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12289 | train 3.010574 | val 3.348749 | tok/s 2628624.65 | norm 0.3824 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12290 | train 2.936237 | val 3.348749 | tok/s 2630682.65 | norm 0.3875 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12291 | train 3.009401 | val 3.348749 | tok/s 2631154.96 | norm 0.3864 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12292 | train 3.000561 | val 3.348749 | tok/s 2630957.92 | norm 0.3508 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12293 | train 2.993330 | val 3.348749 | tok/s 2629646.79 | norm 0.3934 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12294 | train 2.911804 | val 3.348749 | tok/s 2630087.89 | norm 0.4462 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12295 | train 2.978520 | val 3.348749 | tok/s 2628660.77 | norm 0.4156 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12296 | train 3.094209 | val 3.348749 | tok/s 2626409.75 | norm 0.3959 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12297 | train 2.966492 | val 3.348749 | tok/s 2628443.00 | norm 0.3976 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12298 | train 3.136413 | val 3.348749 | tok/s 2610498.47 | norm 0.4241 | dt 200.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12299 | train 3.097954 | val 3.348749 | tok/s 2628643.32 | norm 0.4630 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12300 | train 3.031238 | val 3.348749 | tok/s 2630699.18 | norm 0.4398 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12301 | train 3.144020 | val 3.348749 | tok/s 2628437.90 | norm 0.4212 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12302 | train 2.997210 | val 3.348749 | tok/s 2629832.27 | norm 0.4319 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12303 | train 2.784742 | val 3.348749 | tok/s 2631815.37 | norm 0.3596 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12304 | train 2.882516 | val 3.348749 | tok/s 2630764.55 | norm 0.3926 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12305 | train 2.881503 | val 3.348749 | tok/s 2631146.95 | norm 0.4208 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12306 | train 2.990637 | val 3.348749 | tok/s 2629758.44 | norm 0.3816 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12307 | train 3.065637 | val 3.348749 | tok/s 2634618.93 | norm 0.3716 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12308 | train 2.971769 | val 3.348749 | tok/s 2615108.44 | norm 0.4125 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12309 | train 3.143054 | val 3.348749 | tok/s 2624408.53 | norm 0.3855 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12310 | train 2.953184 | val 3.348749 | tok/s 2626940.33 | norm 0.3691 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12311 | train 3.077283 | val 3.348749 | tok/s 2624402.83 | norm 0.3860 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12312 | train 2.914726 | val 3.348749 | tok/s 2624768.01 | norm 0.3969 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12313 | train 3.026573 | val 3.348749 | tok/s 2627929.16 | norm 0.3709 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12314 | train 2.987474 | val 3.348749 | tok/s 2628680.40 | norm 0.3509 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12315 | train 2.944619 | val 3.348749 | tok/s 2625787.79 | norm 0.3581 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12316 | train 2.980861 | val 3.348749 | tok/s 2629191.58 | norm 0.3989 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12317 | train 3.074041 | val 3.348749 | tok/s 2630117.93 | norm 0.3868 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12318 | train 2.948390 | val 3.348749 | tok/s 2632165.65 | norm 0.3846 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12319 | train 3.068225 | val 3.348749 | tok/s 2632107.38 | norm 0.4249 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12320 | train 3.082099 | val 3.348749 | tok/s 2629869.97 | norm 0.3695 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12321 | train 3.007020 | val 3.348749 | tok/s 2630209.83 | norm 0.3824 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12322 | train 3.006185 | val 3.348749 | tok/s 2580407.95 | norm 0.4025 | dt 203.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12323 | train 3.005018 | val 3.348749 | tok/s 2622144.73 | norm 0.4007 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12324 | train 3.050571 | val 3.348749 | tok/s 2630410.52 | norm 0.4365 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12325 | train 3.143461 | val 3.348749 | tok/s 2630031.17 | norm 0.3893 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12326 | train 2.904532 | val 3.348749 | tok/s 2631907.54 | norm 0.4343 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12327 | train 2.936547 | val 3.348749 | tok/s 2631729.64 | norm 0.3622 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12328 | train 2.970177 | val 3.348749 | tok/s 2634480.31 | norm 0.4036 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12329 | train 3.005423 | val 3.348749 | tok/s 2631900.90 | norm 0.4348 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12330 | train 2.931659 | val 3.348749 | tok/s 2629621.61 | norm 0.3936 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12331 | train 3.406695 | val 3.348749 | tok/s 2631785.56 | norm 0.5759 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12332 | train 2.922268 | val 3.348749 | tok/s 2629087.91 | norm 0.3700 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12333 | train 2.956042 | val 3.348749 | tok/s 2632674.57 | norm 0.3314 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12334 | train 2.957499 | val 3.348749 | tok/s 2626419.19 | norm 0.4033 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12335 | train 3.130321 | val 3.348749 | tok/s 2631764.04 | norm 0.3865 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12336 | train 2.922299 | val 3.348749 | tok/s 2633219.99 | norm 0.4058 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12337 | train 3.037983 | val 3.348749 | tok/s 2631631.60 | norm 0.3577 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12338 | train 3.029858 | val 3.348749 | tok/s 2629296.43 | norm 0.3528 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12339 | train 2.880185 | val 3.348749 | tok/s 2630220.44 | norm 0.3925 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12340 | train 2.926239 | val 3.348749 | tok/s 2632278.00 | norm 0.4069 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12341 | train 2.957725 | val 3.348749 | tok/s 2634132.74 | norm 0.4007 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12342 | train 2.924265 | val 3.348749 | tok/s 2631644.13 | norm 0.4105 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12343 | train 2.895622 | val 3.348749 | tok/s 2632477.17 | norm 0.3933 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12344 | train 3.019955 | val 3.348749 | tok/s 2629534.61 | norm 0.4029 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12345 | train 3.083651 | val 3.348749 | tok/s 2631340.26 | norm 1.4629 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12346 | train 2.977204 | val 3.348749 | tok/s 2632366.61 | norm 0.4815 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12347 | train 3.042445 | val 3.348749 | tok/s 2630685.25 | norm 0.3643 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12348 | train 2.970111 | val 3.348749 | tok/s 2631566.19 | norm 0.3244 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12349 | train 2.991194 | val 3.348749 | tok/s 2633135.86 | norm 0.4050 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12350 | train 2.971632 | val 3.348749 | tok/s 2634870.49 | norm 0.3909 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12351 | train 2.983768 | val 3.348749 | tok/s 2632907.51 | norm 0.3503 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12352 | train 2.998756 | val 3.348749 | tok/s 2630488.42 | norm 0.3680 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12353 | train 2.920547 | val 3.348749 | tok/s 2626847.62 | norm 0.3495 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12354 | train 2.947506 | val 3.348749 | tok/s 2630252.91 | norm 0.3975 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12355 | train 2.936639 | val 3.348749 | tok/s 2632131.89 | norm 0.3521 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12356 | train 2.983685 | val 3.348749 | tok/s 2633021.05 | norm 0.3578 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12357 | train 2.910631 | val 3.348749 | tok/s 2630540.96 | norm 0.4110 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12358 | train 2.953724 | val 3.348749 | tok/s 2631724.36 | norm 0.4007 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12359 | train 3.068689 | val 3.348749 | tok/s 2630318.75 | norm 0.3704 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12360 | train 3.120876 | val 3.348749 | tok/s 2629530.25 | norm 0.4103 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12361 | train 3.017203 | val 3.348749 | tok/s 2629897.65 | norm 0.3831 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12362 | train 2.992461 | val 3.348749 | tok/s 2626760.58 | norm 0.3781 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12363 | train 3.108342 | val 3.348749 | tok/s 2632023.60 | norm 0.3935 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12364 | train 3.158418 | val 3.348749 | tok/s 2585930.61 | norm 0.3839 | dt 202.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12365 | train 2.970809 | val 3.348749 | tok/s 2626185.64 | norm 0.3861 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12366 | train 2.975633 | val 3.348749 | tok/s 2562631.33 | norm 0.3976 | dt 204.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12367 | train 2.954988 | val 3.348749 | tok/s 2629501.63 | norm 0.3767 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12368 | train 3.099522 | val 3.348749 | tok/s 2629757.26 | norm 0.4038 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12369 | train 3.031617 | val 3.348749 | tok/s 2626321.70 | norm 0.3795 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12370 | train 2.921139 | val 3.348749 | tok/s 2631502.26 | norm 0.3764 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12371 | train 3.014251 | val 3.348749 | tok/s 2635293.67 | norm 0.3781 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12372 | train 3.022403 | val 3.348749 | tok/s 2630979.58 | norm 0.3719 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12373 | train 3.031805 | val 3.348749 | tok/s 2631032.10 | norm 0.4016 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12374 | train 2.943425 | val 3.348749 | tok/s 2630604.28 | norm 0.3775 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12375 | train 2.980367 | val 3.348749 | tok/s 2631047.64 | norm 0.3622 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12376 | train 3.075277 | val 3.348749 | tok/s 2633631.07 | norm 0.3688 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12377 | train 2.804626 | val 3.348749 | tok/s 2630358.92 | norm 0.3477 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12378 | train 3.123888 | val 3.348749 | tok/s 2631394.21 | norm 0.3568 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12379 | train 2.962238 | val 3.348749 | tok/s 2629293.49 | norm 0.3865 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12380 | train 2.983095 | val 3.348749 | tok/s 2633618.03 | norm 0.3912 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12381 | train 2.984818 | val 3.348749 | tok/s 2632913.67 | norm 0.3865 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12382 | train 3.184630 | val 3.348749 | tok/s 2554565.34 | norm 0.6687 | dt 205.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12383 | train 3.015357 | val 3.348749 | tok/s 2630750.21 | norm 0.3751 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12384 | train 3.040652 | val 3.348749 | tok/s 2627480.89 | norm 0.3827 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12385 | train 3.012418 | val 3.348749 | tok/s 2630394.49 | norm 0.4065 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12386 | train 2.962838 | val 3.348749 | tok/s 2632245.74 | norm 0.3804 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12387 | train 3.010406 | val 3.348749 | tok/s 2629139.16 | norm 0.3619 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12388 | train 2.986052 | val 3.348749 | tok/s 2632294.01 | norm 0.3896 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12389 | train 2.975671 | val 3.348749 | tok/s 2633533.20 | norm 0.3751 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12390 | train 2.953055 | val 3.348749 | tok/s 2623050.17 | norm 0.3666 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12391 | train 3.143253 | val 3.348749 | tok/s 2627406.92 | norm 0.4803 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12392 | train 2.989513 | val 3.348749 | tok/s 2634252.21 | norm 0.3549 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12393 | train 3.024902 | val 3.348749 | tok/s 2634771.84 | norm 0.3626 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12394 | train 3.101809 | val 3.348749 | tok/s 2622045.73 | norm 0.4009 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12395 | train 3.148328 | val 3.348749 | tok/s 2628819.35 | norm 0.3863 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12396 | train 2.952436 | val 3.348749 | tok/s 2630394.44 | norm 0.3652 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12397 | train 3.003711 | val 3.348749 | tok/s 2631495.54 | norm 0.3930 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12398 | train 2.986833 | val 3.348749 | tok/s 2626151.32 | norm 0.3995 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12399 | train 2.978644 | val 3.348749 | tok/s 2628102.74 | norm 0.3818 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12400 | train 3.047923 | val 3.348749 | tok/s 2627758.27 | norm 0.4193 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12401 | train 2.966371 | val 3.348749 | tok/s 2632318.57 | norm 0.3856 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12402 | train 2.999103 | val 3.348749 | tok/s 2631776.74 | norm 0.3339 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12403 | train 3.006599 | val 3.348749 | tok/s 2630819.33 | norm 0.3671 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12404 | train 3.083140 | val 3.348749 | tok/s 2630449.80 | norm 0.3864 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12405 | train 2.971765 | val 3.348749 | tok/s 2632287.75 | norm 0.3746 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12406 | train 3.013557 | val 3.348749 | tok/s 2633486.09 | norm 0.3978 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12407 | train 2.995126 | val 3.348749 | tok/s 2634032.80 | norm 0.3969 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12408 | train 3.088437 | val 3.348749 | tok/s 2631564.85 | norm 0.3945 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12409 | train 3.097367 | val 3.348749 | tok/s 2630864.92 | norm 0.4274 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12410 | train 3.115534 | val 3.348749 | tok/s 2600213.53 | norm 0.4720 | dt 201.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12411 | train 3.045558 | val 3.348749 | tok/s 2628527.69 | norm 0.3971 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12412 | train 2.942610 | val 3.348749 | tok/s 2633048.65 | norm 0.3529 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12413 | train 3.033888 | val 3.348749 | tok/s 2632246.94 | norm 0.3937 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12414 | train 3.049862 | val 3.348749 | tok/s 2617284.58 | norm 0.3955 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12415 | train 2.986359 | val 3.348749 | tok/s 2632250.69 | norm 0.3671 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12416 | train 3.036792 | val 3.348749 | tok/s 2632774.07 | norm 0.4176 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12417 | train 2.923851 | val 3.348749 | tok/s 2632146.49 | norm 0.4239 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12418 | train 3.020590 | val 3.348749 | tok/s 2631394.61 | norm 0.3900 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12419 | train 2.987716 | val 3.348749 | tok/s 2612843.77 | norm 0.3951 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12420 | train 2.975628 | val 3.348749 | tok/s 2631656.06 | norm 0.3905 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12421 | train 2.848243 | val 3.348749 | tok/s 2630475.10 | norm 0.3814 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12422 | train 2.942949 | val 3.348749 | tok/s 2631091.41 | norm 0.3862 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12423 | train 2.941973 | val 3.348749 | tok/s 2632136.94 | norm 0.3799 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12424 | train 2.945744 | val 3.348749 | tok/s 2633143.28 | norm 0.4057 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12425 | train 2.944717 | val 3.348749 | tok/s 2631088.49 | norm 0.4212 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12426 | train 2.929490 | val 3.348749 | tok/s 2631316.28 | norm 0.3950 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12427 | train 2.973405 | val 3.348749 | tok/s 2633166.45 | norm 0.3551 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12428 | train 2.989164 | val 3.348749 | tok/s 2632138.68 | norm 0.3594 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12429 | train 2.968201 | val 3.348749 | tok/s 2632762.26 | norm 0.3702 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12430 | train 2.921957 | val 3.348749 | tok/s 2630496.88 | norm 0.3810 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12431 | train 2.982046 | val 3.348749 | tok/s 2634500.40 | norm 0.3909 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12432 | train 2.949253 | val 3.348749 | tok/s 2631352.79 | norm 0.3767 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12433 | train 2.875365 | val 3.348749 | tok/s 2486600.55 | norm 0.3849 | dt 210.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12434 | train 3.033887 | val 3.348749 | tok/s 2515234.32 | norm 0.3975 | dt 208.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12435 | train 2.927443 | val 3.348749 | tok/s 2625208.16 | norm 0.3660 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12436 | train 2.938933 | val 3.348749 | tok/s 2627707.75 | norm 0.3672 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12437 | train 3.086628 | val 3.348749 | tok/s 2629311.20 | norm 0.3981 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12438 | train 2.883183 | val 3.348749 | tok/s 2634158.51 | norm 0.3542 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12439 | train 3.063139 | val 3.348749 | tok/s 2630289.00 | norm 0.3691 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12440 | train 2.958121 | val 3.348749 | tok/s 2625674.14 | norm 0.3677 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12441 | train 2.996149 | val 3.348749 | tok/s 2626380.59 | norm 0.3812 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12442 | train 2.931926 | val 3.348749 | tok/s 2632077.59 | norm 0.4325 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12443 | train 2.913310 | val 3.348749 | tok/s 2614131.70 | norm 0.3950 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12444 | train 2.935247 | val 3.348749 | tok/s 2629646.79 | norm 0.3937 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12445 | train 2.957430 | val 3.348749 | tok/s 2623149.40 | norm 0.4106 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12446 | train 3.078939 | val 3.348749 | tok/s 2628354.44 | norm 0.4138 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12447 | train 2.965408 | val 3.348749 | tok/s 2629245.92 | norm 0.4067 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12448 | train 2.994741 | val 3.348749 | tok/s 2628430.22 | norm 0.3787 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12449 | train 2.986413 | val 3.348749 | tok/s 2631918.99 | norm 0.3778 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12450 | train 2.978741 | val 3.348749 | tok/s 2630570.70 | norm 0.3865 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12451 | train 3.097536 | val 3.348749 | tok/s 2621810.34 | norm 0.4149 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12452 | train 2.932762 | val 3.348749 | tok/s 2631300.94 | norm 0.4040 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12453 | train 2.977352 | val 3.348749 | tok/s 2628798.74 | norm 0.3929 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12454 | train 2.929386 | val 3.348749 | tok/s 2630690.30 | norm 0.4084 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12455 | train 2.926194 | val 3.348749 | tok/s 2632143.81 | norm 0.3833 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12456 | train 2.847876 | val 3.348749 | tok/s 2632658.60 | norm 0.4128 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12457 | train 2.860019 | val 3.348749 | tok/s 2630553.92 | norm 0.4033 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12458 | train 3.167014 | val 3.348749 | tok/s 2631656.05 | norm 0.4969 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12459 | train 2.893604 | val 3.348749 | tok/s 2632327.23 | norm 0.3942 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12460 | train 2.864458 | val 3.348749 | tok/s 2633147.01 | norm 0.3899 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12461 | train 3.078689 | val 3.348749 | tok/s 2629463.57 | norm 0.4053 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12462 | train 3.001935 | val 3.348749 | tok/s 2630172.87 | norm 0.3791 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12463 | train 3.135537 | val 3.348749 | tok/s 2630667.95 | norm 0.4810 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12464 | train 2.895526 | val 3.348749 | tok/s 2628101.67 | norm 0.4030 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12465 | train 2.929554 | val 3.348749 | tok/s 2632929.08 | norm 0.3180 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12466 | train 3.025456 | val 3.348749 | tok/s 2628434.26 | norm 0.4030 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12467 | train 3.029337 | val 3.348749 | tok/s 2628696.35 | norm 0.3982 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12468 | train 2.903650 | val 3.348749 | tok/s 2629624.77 | norm 0.3664 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12469 | train 2.895121 | val 3.348749 | tok/s 2626939.55 | norm 0.3991 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12470 | train 3.061516 | val 3.348749 | tok/s 2629194.59 | norm 0.3943 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12471 | train 3.068196 | val 3.348749 | tok/s 2630181.10 | norm 0.4111 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12472 | train 2.906155 | val 3.348749 | tok/s 2632230.87 | norm 0.3850 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12473 | train 3.096501 | val 3.348749 | tok/s 2628439.17 | norm 0.3824 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12474 | train 3.065407 | val 3.348749 | tok/s 2631935.12 | norm 0.3689 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12475 | train 2.924407 | val 3.348749 | tok/s 2628228.00 | norm 0.3666 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12476 | train 2.949053 | val 3.348749 | tok/s 2628988.36 | norm 0.3948 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12477 | train 2.916428 | val 3.348749 | tok/s 2630739.86 | norm 0.4169 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12478 | train 3.031416 | val 3.348749 | tok/s 2630193.32 | norm 0.3896 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12479 | train 2.937382 | val 3.348749 | tok/s 2629929.60 | norm 0.3625 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12480 | train 3.001020 | val 3.348749 | tok/s 2629822.15 | norm 0.4146 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12481 | train 3.015925 | val 3.348749 | tok/s 2629950.57 | norm 0.3759 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12482 | train 2.990881 | val 3.348749 | tok/s 2631209.64 | norm 0.3430 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12483 | train 2.913992 | val 3.348749 | tok/s 2629344.86 | norm 0.3601 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12484 | train 2.919827 | val 3.348749 | tok/s 2630731.30 | norm 0.3715 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12485 | train 2.971069 | val 3.348749 | tok/s 2633894.46 | norm 0.4068 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12486 | train 2.942775 | val 3.348749 | tok/s 2632070.26 | norm 0.4084 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12487 | train 2.937151 | val 3.348749 | tok/s 2628766.33 | norm 0.3853 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12488 | train 2.995917 | val 3.348749 | tok/s 2633122.39 | norm 0.4082 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12489 | train 2.991820 | val 3.348749 | tok/s 2629404.49 | norm 0.4150 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12490 | train 3.046485 | val 3.348749 | tok/s 2635032.17 | norm 0.4195 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12491 | train 2.970884 | val 3.348749 | tok/s 2627941.13 | norm 0.3964 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12492 | train 3.004685 | val 3.348749 | tok/s 2628623.29 | norm 0.3665 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12493 | train 3.016038 | val 3.348749 | tok/s 2631471.50 | norm 0.3389 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12494 | train 2.857183 | val 3.348749 | tok/s 2620351.13 | norm 0.3307 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12495 | train 2.990375 | val 3.348749 | tok/s 2633392.90 | norm 0.3530 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12496 | train 2.759869 | val 3.348749 | tok/s 2633117.76 | norm 0.4588 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12497 | train 2.989198 | val 3.348749 | tok/s 2629027.56 | norm 0.3748 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12498 | train 2.938736 | val 3.348749 | tok/s 2630963.30 | norm 0.3928 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12499 | train 2.972543 | val 3.348749 | tok/s 2628697.99 | norm 0.3920 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 12500 | val_loss 3.348522 step 12500 | train 3.172888 | val 3.348522 | tok/s 2615824.39 | norm 0.4615 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12501 | train 2.928252 | val 3.348522 | tok/s 2630502.73 | norm 0.3707 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12502 | train 3.012457 | val 3.348522 | tok/s 2634705.94 | norm 0.3662 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12503 | train 2.954513 | val 3.348522 | tok/s 2634074.46 | norm 0.3669 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12504 | train 2.925792 | val 3.348522 | tok/s 2630729.65 | norm 0.3720 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12505 | train 3.078210 | val 3.348522 | tok/s 2630208.56 | norm 0.3993 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12506 | train 3.102154 | val 3.348522 | tok/s 2625953.48 | norm 0.4347 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12507 | train 3.046754 | val 3.348522 | tok/s 2630012.47 | norm 0.4401 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12508 | train 3.043411 | val 3.348522 | tok/s 2634346.56 | norm 0.3728 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12509 | train 3.041795 | val 3.348522 | tok/s 2498194.30 | norm 0.3849 | dt 209.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12510 | train 2.980715 | val 3.348522 | tok/s 2635862.74 | norm 0.3867 | dt 198.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12511 | train 2.948316 | val 3.348522 | tok/s 2630599.43 | norm 0.3769 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12512 | train 2.941746 | val 3.348522 | tok/s 2633544.01 | norm 0.3738 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12513 | train 3.026870 | val 3.348522 | tok/s 2627242.04 | norm 0.4176 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12514 | train 3.103909 | val 3.348522 | tok/s 2632289.04 | norm 0.4232 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12515 | train 3.009444 | val 3.348522 | tok/s 2629423.36 | norm 0.3528 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12516 | train 3.019651 | val 3.348522 | tok/s 2630986.09 | norm 0.3704 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12517 | train 2.902906 | val 3.348522 | tok/s 2629118.20 | norm 0.4298 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12518 | train 3.020062 | val 3.348522 | tok/s 2625976.69 | norm 0.4077 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12519 | train 3.085787 | val 3.348522 | tok/s 2630300.69 | norm 0.4153 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12520 | train 3.039379 | val 3.348522 | tok/s 2630725.84 | norm 0.4253 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12521 | train 2.911362 | val 3.348522 | tok/s 2631217.54 | norm 0.3956 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12522 | train 3.174176 | val 3.348522 | tok/s 2631133.19 | norm 0.3936 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12523 | train 2.964283 | val 3.348522 | tok/s 2626841.69 | norm 0.4190 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12524 | train 2.964152 | val 3.348522 | tok/s 2629835.78 | norm 0.3651 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12525 | train 2.833678 | val 3.348522 | tok/s 2630583.32 | norm 0.3608 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12526 | train 3.087641 | val 3.348522 | tok/s 2629987.83 | norm 0.4184 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12527 | train 2.977236 | val 3.348522 | tok/s 2629914.00 | norm 0.4239 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12528 | train 2.964564 | val 3.348522 | tok/s 2625878.69 | norm 0.3875 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12529 | train 2.967468 | val 3.348522 | tok/s 2624794.08 | norm 0.4018 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12530 | train 2.980209 | val 3.348522 | tok/s 2628460.95 | norm 0.4198 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12531 | train 3.041529 | val 3.348522 | tok/s 2629939.02 | norm 0.3912 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12532 | train 2.798216 | val 3.348522 | tok/s 2629857.81 | norm 0.3787 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12533 | train 3.011551 | val 3.348522 | tok/s 2627346.62 | norm 0.4107 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12534 | train 2.992156 | val 3.348522 | tok/s 2632598.74 | norm 0.3921 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12535 | train 2.974598 | val 3.348522 | tok/s 2632534.09 | norm 0.3900 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12536 | train 2.927582 | val 3.348522 | tok/s 2631777.62 | norm 0.3907 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12537 | train 3.025380 | val 3.348522 | tok/s 2630956.47 | norm 0.3837 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12538 | train 3.032228 | val 3.348522 | tok/s 2628841.63 | norm 0.3997 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12539 | train 2.921600 | val 3.348522 | tok/s 2632214.40 | norm 0.3958 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12540 | train 3.062656 | val 3.348522 | tok/s 2632676.96 | norm 0.3898 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12541 | train 3.043162 | val 3.348522 | tok/s 2628004.82 | norm 0.4070 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12542 | train 2.909592 | val 3.348522 | tok/s 2631939.60 | norm 0.3966 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12543 | train 2.998374 | val 3.348522 | tok/s 2629760.68 | norm 0.3704 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12544 | train 2.947492 | val 3.348522 | tok/s 2631884.09 | norm 0.3680 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12545 | train 3.082878 | val 3.348522 | tok/s 2633727.18 | norm 0.3875 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12546 | train 2.962858 | val 3.348522 | tok/s 2636080.14 | norm 0.3587 | dt 198.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12547 | train 2.938385 | val 3.348522 | tok/s 2634615.42 | norm 0.3269 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12548 | train 3.068406 | val 3.348522 | tok/s 2631836.56 | norm 0.3572 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12549 | train 2.957223 | val 3.348522 | tok/s 2631636.86 | norm 0.3945 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12550 | train 2.947488 | val 3.348522 | tok/s 2633322.47 | norm 0.3732 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12551 | train 2.826832 | val 3.348522 | tok/s 2633230.74 | norm 0.3982 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12552 | train 2.919625 | val 3.348522 | tok/s 2633087.07 | norm 0.3895 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12553 | train 2.994195 | val 3.348522 | tok/s 2632217.71 | norm 0.3927 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12554 | train 2.941892 | val 3.348522 | tok/s 2631203.08 | norm 0.3689 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12555 | train 2.993560 | val 3.348522 | tok/s 2629680.54 | norm 0.3605 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12556 | train 3.001258 | val 3.348522 | tok/s 2635046.34 | norm 0.3708 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12557 | train 3.085504 | val 3.348522 | tok/s 2632496.02 | norm 0.3965 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12558 | train 2.986219 | val 3.348522 | tok/s 2631398.32 | norm 0.4143 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12559 | train 2.926519 | val 3.348522 | tok/s 2628974.71 | norm 0.3749 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12560 | train 2.953721 | val 3.348522 | tok/s 2628989.20 | norm 0.3773 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12561 | train 3.080118 | val 3.348522 | tok/s 2631318.28 | norm 0.3877 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12562 | train 2.833979 | val 3.348522 | tok/s 2628183.62 | norm 0.4013 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12563 | train 2.855801 | val 3.348522 | tok/s 2628800.52 | norm 0.4001 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12564 | train 2.965876 | val 3.348522 | tok/s 2631461.38 | norm 0.3897 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12565 | train 2.917191 | val 3.348522 | tok/s 2632204.89 | norm 0.4110 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12566 | train 2.911784 | val 3.348522 | tok/s 2631922.34 | norm 0.3866 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12567 | train 2.881194 | val 3.348522 | tok/s 2633132.00 | norm 0.4030 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12568 | train 2.922890 | val 3.348522 | tok/s 2631628.47 | norm 0.4248 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12569 | train 3.167130 | val 3.348522 | tok/s 2632164.00 | norm 0.3680 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12570 | train 3.044384 | val 3.348522 | tok/s 2633762.19 | norm 0.3860 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12571 | train 2.901335 | val 3.348522 | tok/s 2631475.99 | norm 0.3770 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12572 | train 2.972611 | val 3.348522 | tok/s 2632346.23 | norm 0.3796 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12573 | train 3.069827 | val 3.348522 | tok/s 2633713.37 | norm 0.3813 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12574 | train 3.013061 | val 3.348522 | tok/s 2628349.97 | norm 0.3874 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12575 | train 3.007419 | val 3.348522 | tok/s 2627862.71 | norm 0.4366 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12576 | train 2.983634 | val 3.348522 | tok/s 2629457.77 | norm 0.4228 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12577 | train 3.090368 | val 3.348522 | tok/s 2632156.63 | norm 0.4339 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12578 | train 2.997127 | val 3.348522 | tok/s 2483727.60 | norm 0.4321 | dt 211.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12579 | train 2.941963 | val 3.348522 | tok/s 2634103.39 | norm 0.3908 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12580 | train 2.820246 | val 3.348522 | tok/s 2631370.27 | norm 0.3849 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12581 | train 2.862087 | val 3.348522 | tok/s 2629636.75 | norm 0.3820 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12582 | train 2.914551 | val 3.348522 | tok/s 2632663.04 | norm 0.3801 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12583 | train 2.916873 | val 3.348522 | tok/s 2628593.14 | norm 0.3808 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12584 | train 3.032826 | val 3.348522 | tok/s 2630473.06 | norm 0.4138 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12585 | train 3.004183 | val 3.348522 | tok/s 2630605.83 | norm 0.3523 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12586 | train 2.927128 | val 3.348522 | tok/s 2627340.55 | norm 0.3211 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12587 | train 2.814215 | val 3.348522 | tok/s 2632073.37 | norm 0.3885 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12588 | train 2.928262 | val 3.348522 | tok/s 2623457.01 | norm 0.4059 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12589 | train 2.984596 | val 3.348522 | tok/s 2631439.93 | norm 0.4010 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12590 | train 3.025439 | val 3.348522 | tok/s 2630590.22 | norm 0.3880 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12591 | train 3.041298 | val 3.348522 | tok/s 2629652.97 | norm 0.4019 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12592 | train 2.907845 | val 3.348522 | tok/s 2629574.09 | norm 0.3930 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12593 | train 2.888356 | val 3.348522 | tok/s 2628773.37 | norm 0.3923 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12594 | train 2.928829 | val 3.348522 | tok/s 2629234.43 | norm 0.3802 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12595 | train 2.944962 | val 3.348522 | tok/s 2632050.63 | norm 0.3877 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12596 | train 2.894847 | val 3.348522 | tok/s 2630287.76 | norm 0.3963 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12597 | train 3.023388 | val 3.348522 | tok/s 2634204.14 | norm 0.3919 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12598 | train 3.008661 | val 3.348522 | tok/s 2628508.44 | norm 0.3942 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12599 | train 2.997284 | val 3.348522 | tok/s 2627541.48 | norm 0.3847 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12600 | train 3.238400 | val 3.348522 | tok/s 2630512.29 | norm 0.4153 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12601 | train 2.826403 | val 3.348522 | tok/s 2629782.96 | norm 0.3992 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12602 | train 2.986230 | val 3.348522 | tok/s 2627958.85 | norm 0.4267 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12603 | train 2.999834 | val 3.348522 | tok/s 2629670.03 | norm 0.4690 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12604 | train 2.951755 | val 3.348522 | tok/s 2629837.31 | norm 0.3830 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12605 | train 3.244532 | val 3.348522 | tok/s 2633721.90 | norm 0.4074 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12606 | train 2.860971 | val 3.348522 | tok/s 2632652.49 | norm 0.3651 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12607 | train 3.085451 | val 3.348522 | tok/s 2617025.04 | norm 0.3943 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12608 | train 2.972501 | val 3.348522 | tok/s 2627616.53 | norm 0.4261 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12609 | train 2.897699 | val 3.348522 | tok/s 2629119.80 | norm 0.3864 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12610 | train 2.859306 | val 3.348522 | tok/s 2629239.52 | norm 0.3770 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12611 | train 2.951270 | val 3.348522 | tok/s 2630417.58 | norm 0.3937 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12612 | train 2.944299 | val 3.348522 | tok/s 2615657.89 | norm 0.4251 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12613 | train 2.897053 | val 3.348522 | tok/s 2633177.12 | norm 0.4229 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12614 | train 2.942087 | val 3.348522 | tok/s 2631767.48 | norm 0.4285 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12615 | train 2.895890 | val 3.348522 | tok/s 2630162.65 | norm 0.3988 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12616 | train 2.949475 | val 3.348522 | tok/s 2633163.97 | norm 0.3528 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12617 | train 2.972470 | val 3.348522 | tok/s 2633078.69 | norm 0.3445 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12618 | train 3.114847 | val 3.348522 | tok/s 2632402.22 | norm 0.3713 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12619 | train 2.928609 | val 3.348522 | tok/s 2632213.95 | norm 0.3624 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12620 | train 2.959339 | val 3.348522 | tok/s 2629709.74 | norm 0.3755 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12621 | train 3.104182 | val 3.348522 | tok/s 2628328.99 | norm 0.3808 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12622 | train 2.962673 | val 3.348522 | tok/s 2631574.11 | norm 0.3771 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12623 | train 3.040645 | val 3.348522 | tok/s 2631456.63 | norm 0.4139 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12624 | train 3.029039 | val 3.348522 | tok/s 2629767.33 | norm 0.3904 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12625 | train 2.926450 | val 3.348522 | tok/s 2629281.33 | norm 0.3494 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12626 | train 2.972750 | val 3.348522 | tok/s 2629418.81 | norm 0.3624 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12627 | train 2.921443 | val 3.348522 | tok/s 2633619.66 | norm 0.3948 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12628 | train 3.020526 | val 3.348522 | tok/s 2631667.45 | norm 0.3954 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12629 | train 3.106925 | val 3.348522 | tok/s 2630404.01 | norm 0.4282 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12630 | train 2.960139 | val 3.348522 | tok/s 2632962.69 | norm 0.4121 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12631 | train 2.865435 | val 3.348522 | tok/s 2631748.51 | norm 0.3921 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12632 | train 2.934277 | val 3.348522 | tok/s 2628444.17 | norm 0.4077 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12633 | train 3.066964 | val 3.348522 | tok/s 2630848.60 | norm 0.3872 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12634 | train 2.963022 | val 3.348522 | tok/s 2632663.69 | norm 0.4115 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12635 | train 2.959686 | val 3.348522 | tok/s 2631154.37 | norm 0.3991 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12636 | train 3.033079 | val 3.348522 | tok/s 2630586.81 | norm 0.3866 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12637 | train 3.065180 | val 3.348522 | tok/s 2630642.42 | norm 0.4195 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12638 | train 2.881553 | val 3.348522 | tok/s 2625973.68 | norm 0.4103 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12639 | train 2.908081 | val 3.348522 | tok/s 2628088.41 | norm 0.4019 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12640 | train 2.961760 | val 3.348522 | tok/s 2541897.31 | norm 0.4184 | dt 206.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12641 | train 2.942828 | val 3.348522 | tok/s 2629632.50 | norm 0.4131 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12642 | train 2.911496 | val 3.348522 | tok/s 2628210.35 | norm 0.3867 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12643 | train 2.975158 | val 3.348522 | tok/s 2631009.63 | norm 0.3780 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12644 | train 2.889292 | val 3.348522 | tok/s 2629024.95 | norm 0.3613 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12645 | train 2.900612 | val 3.348522 | tok/s 2629641.52 | norm 0.3673 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12646 | train 2.888559 | val 3.348522 | tok/s 2481571.11 | norm 0.3850 | dt 211.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12647 | train 2.925615 | val 3.348522 | tok/s 2628232.64 | norm 0.3612 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12648 | train 3.057096 | val 3.348522 | tok/s 2633386.82 | norm 0.3706 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12649 | train 3.020494 | val 3.348522 | tok/s 2631925.04 | norm 0.3957 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12650 | train 2.951389 | val 3.348522 | tok/s 2629522.28 | norm 0.3825 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12651 | train 3.074150 | val 3.348522 | tok/s 2628490.03 | norm 0.3663 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12652 | train 2.933010 | val 3.348522 | tok/s 2628706.88 | norm 0.4003 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12653 | train 3.275323 | val 3.348522 | tok/s 2631064.91 | norm 0.7453 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12654 | train 2.945466 | val 3.348522 | tok/s 2622045.61 | norm 0.4009 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12655 | train 2.922133 | val 3.348522 | tok/s 2623016.39 | norm 0.3518 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12656 | train 3.011671 | val 3.348522 | tok/s 2627715.89 | norm 0.4091 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12657 | train 2.928603 | val 3.348522 | tok/s 2624652.17 | norm 0.3791 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12658 | train 2.957743 | val 3.348522 | tok/s 2629621.65 | norm 0.3916 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12659 | train 3.009221 | val 3.348522 | tok/s 2629067.23 | norm 0.4014 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12660 | train 3.017349 | val 3.348522 | tok/s 2630384.02 | norm 0.3670 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12661 | train 2.944641 | val 3.348522 | tok/s 2629344.92 | norm 0.3574 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12662 | train 2.923962 | val 3.348522 | tok/s 2626252.91 | norm 0.3530 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12663 | train 2.993431 | val 3.348522 | tok/s 2628350.26 | norm 0.3602 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12664 | train 2.967601 | val 3.348522 | tok/s 2625308.65 | norm 0.3812 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12665 | train 2.964960 | val 3.348522 | tok/s 2630963.18 | norm 0.3478 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12666 | train 2.966476 | val 3.348522 | tok/s 2631543.69 | norm 0.3658 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12667 | train 2.957746 | val 3.348522 | tok/s 2631028.60 | norm 0.4239 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12668 | train 2.917268 | val 3.348522 | tok/s 2633609.55 | norm 0.4035 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12669 | train 2.877008 | val 3.348522 | tok/s 2634419.15 | norm 0.3754 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12670 | train 2.952187 | val 3.348522 | tok/s 2633716.89 | norm 0.3620 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12671 | train 3.044132 | val 3.348522 | tok/s 2627225.40 | norm 0.4065 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12672 | train 2.906101 | val 3.348522 | tok/s 2628985.94 | norm 0.3785 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12673 | train 3.015316 | val 3.348522 | tok/s 2631348.02 | norm 0.3502 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12674 | train 3.079145 | val 3.348522 | tok/s 2630148.61 | norm 0.3957 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12675 | train 3.022995 | val 3.348522 | tok/s 2626911.44 | norm 0.4101 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12676 | train 2.919684 | val 3.348522 | tok/s 2627881.67 | norm 0.3716 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12677 | train 2.885727 | val 3.348522 | tok/s 2630212.36 | norm 0.3698 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12678 | train 2.901303 | val 3.348522 | tok/s 2628875.34 | norm 0.4025 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12679 | train 2.985233 | val 3.348522 | tok/s 2628777.50 | norm 0.4148 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12680 | train 2.934751 | val 3.348522 | tok/s 2630737.45 | norm 0.4065 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12681 | train 2.933419 | val 3.348522 | tok/s 2631892.96 | norm 0.3684 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12682 | train 2.968606 | val 3.348522 | tok/s 2630056.18 | norm 0.3796 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12683 | train 2.882049 | val 3.348522 | tok/s 2630709.44 | norm 0.4171 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12684 | train 2.880096 | val 3.348522 | tok/s 2630420.18 | norm 0.3810 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12685 | train 3.022559 | val 3.348522 | tok/s 2630529.27 | norm 0.3834 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12686 | train 2.959233 | val 3.348522 | tok/s 2629292.87 | norm 0.3906 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12687 | train 2.981396 | val 3.348522 | tok/s 2628024.04 | norm 0.3695 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12688 | train 2.948382 | val 3.348522 | tok/s 2633047.22 | norm 0.3860 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12689 | train 3.059642 | val 3.348522 | tok/s 2630354.40 | norm 0.3926 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12690 | train 2.930098 | val 3.348522 | tok/s 2630193.08 | norm 0.3847 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12691 | train 3.087824 | val 3.348522 | tok/s 2630648.26 | norm 0.3738 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12692 | train 2.971120 | val 3.348522 | tok/s 2629368.19 | norm 0.4176 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12693 | train 2.936086 | val 3.348522 | tok/s 2627506.22 | norm 0.3823 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12694 | train 3.017349 | val 3.348522 | tok/s 2629043.67 | norm 0.3984 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12695 | train 2.965614 | val 3.348522 | tok/s 2626434.86 | norm 0.4325 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12696 | train 3.012788 | val 3.348522 | tok/s 2627030.35 | norm 0.4244 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12697 | train 3.155229 | val 3.348522 | tok/s 2630727.07 | norm 0.3820 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12698 | train 2.981796 | val 3.348522 | tok/s 2630837.81 | norm 0.4173 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12699 | train 2.920110 | val 3.348522 | tok/s 2630704.87 | norm 0.4218 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12700 | train 2.989395 | val 3.348522 | tok/s 2626544.22 | norm 0.4047 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12701 | train 2.869524 | val 3.348522 | tok/s 2630982.19 | norm 0.3529 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12702 | train 3.000490 | val 3.348522 | tok/s 2628744.41 | norm 0.3370 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12703 | train 3.005351 | val 3.348522 | tok/s 2630675.20 | norm 0.3612 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12704 | train 2.923749 | val 3.348522 | tok/s 2629847.05 | norm 0.3942 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12705 | train 3.029951 | val 3.348522 | tok/s 2634062.16 | norm 0.4507 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12706 | train 3.040536 | val 3.348522 | tok/s 2628374.40 | norm 0.4364 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12707 | train 3.048134 | val 3.348522 | tok/s 2632631.61 | norm 0.3672 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12708 | train 2.959819 | val 3.348522 | tok/s 2633179.93 | norm 0.3823 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12709 | train 3.064901 | val 3.348522 | tok/s 2630430.93 | norm 0.4279 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12710 | train 2.991738 | val 3.348522 | tok/s 2632088.01 | norm 0.4086 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12711 | train 2.940126 | val 3.348522 | tok/s 2632361.05 | norm 0.4437 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12712 | train 2.967813 | val 3.348522 | tok/s 2630675.47 | norm 0.4331 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12713 | train 3.095487 | val 3.348522 | tok/s 2632668.38 | norm 0.4164 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12714 | train 2.893987 | val 3.348522 | tok/s 2630894.08 | norm 0.3604 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12715 | train 2.895473 | val 3.348522 | tok/s 2631702.97 | norm 0.3722 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12716 | train 2.858020 | val 3.348522 | tok/s 2628757.17 | norm 0.3920 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12717 | train 3.054631 | val 3.348522 | tok/s 2627260.89 | norm 0.3998 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12718 | train 3.060279 | val 3.348522 | tok/s 2628207.49 | norm 0.3634 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12719 | train 2.910281 | val 3.348522 | tok/s 2629766.13 | norm 0.3916 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12720 | train 2.932506 | val 3.348522 | tok/s 2631324.47 | norm 0.3616 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12721 | train 2.938393 | val 3.348522 | tok/s 2630982.63 | norm 0.3631 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12722 | train 2.897229 | val 3.348522 | tok/s 2628863.65 | norm 0.3975 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12723 | train 2.917753 | val 3.348522 | tok/s 2628783.31 | norm 0.3587 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12724 | train 3.045034 | val 3.348522 | tok/s 2629450.12 | norm 0.3720 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12725 | train 3.008875 | val 3.348522 | tok/s 2628760.93 | norm 0.3843 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12726 | train 3.077958 | val 3.348522 | tok/s 2630892.22 | norm 0.3781 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12727 | train 2.969324 | val 3.348522 | tok/s 2630052.58 | norm 0.3660 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12728 | train 3.021396 | val 3.348522 | tok/s 2625990.33 | norm 0.3606 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12729 | train 3.077510 | val 3.348522 | tok/s 2629056.30 | norm 0.3855 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12730 | train 3.006023 | val 3.348522 | tok/s 2624879.68 | norm 0.3728 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12731 | train 2.943440 | val 3.348522 | tok/s 2630951.53 | norm 0.3521 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12732 | train 3.027586 | val 3.348522 | tok/s 2630933.35 | norm 0.3572 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12733 | train 2.846812 | val 3.348522 | tok/s 2629790.85 | norm 0.3913 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12734 | train 2.942400 | val 3.348522 | tok/s 2631598.30 | norm 0.4455 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12735 | train 2.836874 | val 3.348522 | tok/s 2626149.74 | norm 0.4503 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12736 | train 2.985443 | val 3.348522 | tok/s 2631592.96 | norm 0.4074 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12737 | train 3.013335 | val 3.348522 | tok/s 2628921.77 | norm 0.3584 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12738 | train 2.981034 | val 3.348522 | tok/s 2629463.07 | norm 0.3651 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12739 | train 2.923485 | val 3.348522 | tok/s 2631990.62 | norm 0.3857 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12740 | train 2.872883 | val 3.348522 | tok/s 2630096.57 | norm 0.3605 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12741 | train 2.919908 | val 3.348522 | tok/s 2630337.43 | norm 0.3672 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12742 | train 3.052100 | val 3.348522 | tok/s 2624686.70 | norm 0.4000 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12743 | train 2.958592 | val 3.348522 | tok/s 2627467.34 | norm 0.3796 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12744 | train 2.954340 | val 3.348522 | tok/s 2632192.08 | norm 0.4172 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12745 | train 2.984828 | val 3.348522 | tok/s 2629748.77 | norm 0.4203 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12746 | train 2.888103 | val 3.348522 | tok/s 2630133.27 | norm 0.3688 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12747 | train 2.910018 | val 3.348522 | tok/s 2631436.45 | norm 0.5488 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12748 | train 2.880002 | val 3.348522 | tok/s 2630646.42 | norm 0.3626 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12749 | train 3.066456 | val 3.348522 | tok/s 2630010.38 | norm 0.3583 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 12750 | val_loss 3.348853 step 12750 | train 2.991518 | val 3.348853 | tok/s 2614836.37 | norm 0.3907 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12751 | train 3.123844 | val 3.348853 | tok/s 2629554.49 | norm 0.4140 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12752 | train 3.094374 | val 3.348853 | tok/s 2630022.39 | norm 0.3669 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12753 | train 3.002931 | val 3.348853 | tok/s 2631340.93 | norm 0.3382 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12754 | train 3.011622 | val 3.348853 | tok/s 2628216.21 | norm 0.4018 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12755 | train 2.989373 | val 3.348853 | tok/s 2631019.16 | norm 0.3812 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12756 | train 3.048480 | val 3.348853 | tok/s 2631316.65 | norm 0.3636 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12757 | train 2.968192 | val 3.348853 | tok/s 2630904.61 | norm 0.4027 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12758 | train 2.966394 | val 3.348853 | tok/s 2632325.24 | norm 0.3818 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12759 | train 3.164505 | val 3.348853 | tok/s 2630777.45 | norm 0.3803 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12760 | train 3.329490 | val 3.348853 | tok/s 2632833.94 | norm 0.4317 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12761 | train 3.009631 | val 3.348853 | tok/s 2635377.74 | norm 0.4030 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12762 | train 3.002857 | val 3.348853 | tok/s 2632832.41 | norm 0.3769 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12763 | train 2.978455 | val 3.348853 | tok/s 2630026.52 | norm 0.4113 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12764 | train 2.918377 | val 3.348853 | tok/s 2629879.30 | norm 0.3739 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12765 | train 2.914140 | val 3.348853 | tok/s 2630789.01 | norm 0.3458 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12766 | train 3.036657 | val 3.348853 | tok/s 2616708.99 | norm 0.3629 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12767 | train 2.995619 | val 3.348853 | tok/s 2623592.53 | norm 0.4162 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12768 | train 2.953240 | val 3.348853 | tok/s 2626818.19 | norm 0.3762 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12769 | train 2.998952 | val 3.348853 | tok/s 2630308.12 | norm 0.3533 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12770 | train 2.925179 | val 3.348853 | tok/s 2629120.29 | norm 0.4030 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12771 | train 3.005353 | val 3.348853 | tok/s 2631381.78 | norm 0.4008 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12772 | train 3.028422 | val 3.348853 | tok/s 2632574.02 | norm 0.4165 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12773 | train 3.010062 | val 3.348853 | tok/s 2628798.32 | norm 0.4047 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12774 | train 3.027335 | val 3.348853 | tok/s 2630894.84 | norm 0.3841 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12775 | train 2.861691 | val 3.348853 | tok/s 2633334.06 | norm 0.3950 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12776 | train 3.017909 | val 3.348853 | tok/s 2629964.99 | norm 0.3779 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12777 | train 2.991902 | val 3.348853 | tok/s 2626464.81 | norm 0.3487 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12778 | train 3.001273 | val 3.348853 | tok/s 2627213.27 | norm 0.3488 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12779 | train 2.995582 | val 3.348853 | tok/s 2630680.66 | norm 0.3737 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12780 | train 3.008974 | val 3.348853 | tok/s 2630230.24 | norm 0.4078 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12781 | train 2.958293 | val 3.348853 | tok/s 2632244.59 | norm 0.3752 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12782 | train 3.026326 | val 3.348853 | tok/s 2632996.25 | norm 0.3354 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12783 | train 2.978391 | val 3.348853 | tok/s 2629673.11 | norm 0.3476 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12784 | train 3.024450 | val 3.348853 | tok/s 2630346.72 | norm 0.3886 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12785 | train 2.971054 | val 3.348853 | tok/s 2628003.78 | norm 0.3647 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12786 | train 2.936010 | val 3.348853 | tok/s 2631204.51 | norm 0.3586 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12787 | train 3.059025 | val 3.348853 | tok/s 2629999.58 | norm 0.4104 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12788 | train 2.931117 | val 3.348853 | tok/s 2634566.85 | norm 0.3931 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12789 | train 2.967577 | val 3.348853 | tok/s 2634416.41 | norm 0.3833 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12790 | train 2.896106 | val 3.348853 | tok/s 2632397.55 | norm 0.4084 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12791 | train 2.909555 | val 3.348853 | tok/s 2627317.69 | norm 0.4426 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12792 | train 2.891078 | val 3.348853 | tok/s 2630895.81 | norm 0.4053 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12793 | train 2.931608 | val 3.348853 | tok/s 2627279.65 | norm 0.3878 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12794 | train 3.034607 | val 3.348853 | tok/s 2632504.26 | norm 0.3938 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12795 | train 3.042382 | val 3.348853 | tok/s 2632738.93 | norm 0.3934 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12796 | train 2.944741 | val 3.348853 | tok/s 2628426.07 | norm 0.3773 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12797 | train 3.005798 | val 3.348853 | tok/s 2631324.35 | norm 0.4077 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12798 | train 2.981445 | val 3.348853 | tok/s 2625155.01 | norm 0.4311 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12799 | train 3.013921 | val 3.348853 | tok/s 2629979.33 | norm 0.3880 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12800 | train 2.821530 | val 3.348853 | tok/s 2632027.99 | norm 0.4022 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12801 | train 3.016628 | val 3.348853 | tok/s 2627721.49 | norm 0.4128 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12802 | train 2.972114 | val 3.348853 | tok/s 2631733.21 | norm 0.3859 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12803 | train 2.934154 | val 3.348853 | tok/s 2627513.14 | norm 0.4314 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12804 | train 3.004280 | val 3.348853 | tok/s 2627307.34 | norm 0.4313 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12805 | train 2.883095 | val 3.348853 | tok/s 2633047.59 | norm 0.3959 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12806 | train 3.077653 | val 3.348853 | tok/s 2632464.15 | norm 0.4151 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12807 | train 3.016427 | val 3.348853 | tok/s 2631121.66 | norm 0.4117 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12808 | train 2.986763 | val 3.348853 | tok/s 2628421.39 | norm 0.3883 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12809 | train 2.809474 | val 3.348853 | tok/s 2632891.71 | norm 0.3805 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12810 | train 2.957169 | val 3.348853 | tok/s 2633560.95 | norm 0.3669 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12811 | train 2.978527 | val 3.348853 | tok/s 2629782.23 | norm 0.3771 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12812 | train 2.913446 | val 3.348853 | tok/s 2632480.12 | norm 0.3575 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12813 | train 3.006939 | val 3.348853 | tok/s 2627913.17 | norm 0.5176 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12814 | train 3.024062 | val 3.348853 | tok/s 2628680.52 | norm 0.4036 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12815 | train 3.060116 | val 3.348853 | tok/s 2629458.85 | norm 0.3555 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12816 | train 2.980037 | val 3.348853 | tok/s 2630436.37 | norm 0.3908 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12817 | train 3.083130 | val 3.348853 | tok/s 2630329.09 | norm 0.4130 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12818 | train 3.145678 | val 3.348853 | tok/s 2626735.53 | norm 0.3973 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12819 | train 3.052230 | val 3.348853 | tok/s 2632009.55 | norm 0.4731 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12820 | train 3.058637 | val 3.348853 | tok/s 2630937.53 | norm 0.3808 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12821 | train 3.047810 | val 3.348853 | tok/s 2630805.64 | norm 0.3739 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12822 | train 2.880633 | val 3.348853 | tok/s 2630583.95 | norm 0.3954 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12823 | train 3.059162 | val 3.348853 | tok/s 2632347.51 | norm 0.3738 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12824 | train 2.856578 | val 3.348853 | tok/s 2631587.60 | norm 0.3737 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12825 | train 2.978611 | val 3.348853 | tok/s 2632701.16 | norm 0.3701 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12826 | train 2.830484 | val 3.348853 | tok/s 2633052.54 | norm 0.3795 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12827 | train 2.861851 | val 3.348853 | tok/s 2629973.80 | norm 0.4030 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12828 | train 2.917945 | val 3.348853 | tok/s 2632049.22 | norm 0.3915 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12829 | train 2.977687 | val 3.348853 | tok/s 2632068.07 | norm 0.4006 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12830 | train 2.994650 | val 3.348853 | tok/s 2628330.50 | norm 0.3762 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12831 | train 3.012014 | val 3.348853 | tok/s 2619441.56 | norm 0.3583 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12832 | train 3.089351 | val 3.348853 | tok/s 2627120.29 | norm 0.3952 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12833 | train 2.912822 | val 3.348853 | tok/s 2627023.29 | norm 0.4115 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12834 | train 3.032560 | val 3.348853 | tok/s 2632602.52 | norm 0.3798 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12835 | train 3.097855 | val 3.348853 | tok/s 2627998.08 | norm 0.3624 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12836 | train 2.930035 | val 3.348853 | tok/s 2629883.32 | norm 0.4223 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12837 | train 3.039733 | val 3.348853 | tok/s 2631226.14 | norm 0.4227 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12838 | train 3.010643 | val 3.348853 | tok/s 2632316.90 | norm 0.4006 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12839 | train 2.941759 | val 3.348853 | tok/s 2622834.17 | norm 0.4152 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12840 | train 2.963819 | val 3.348853 | tok/s 2629926.29 | norm 0.3926 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12841 | train 3.137601 | val 3.348853 | tok/s 2631162.76 | norm 0.4128 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12842 | train 3.106526 | val 3.348853 | tok/s 2630792.22 | norm 0.3989 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12843 | train 3.035592 | val 3.348853 | tok/s 2632284.80 | norm 0.3680 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12844 | train 3.009995 | val 3.348853 | tok/s 2630033.23 | norm 0.3885 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12845 | train 2.984427 | val 3.348853 | tok/s 2631336.98 | norm 0.4164 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12846 | train 2.974884 | val 3.348853 | tok/s 2628774.70 | norm 0.3834 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12847 | train 2.888316 | val 3.348853 | tok/s 2631104.43 | norm 0.3847 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12848 | train 2.925544 | val 3.348853 | tok/s 2630752.42 | norm 0.3886 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12849 | train 3.106331 | val 3.348853 | tok/s 2630697.29 | norm 0.4400 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12850 | train 2.979361 | val 3.348853 | tok/s 2631466.65 | norm 0.3851 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12851 | train 3.041098 | val 3.348853 | tok/s 2633104.79 | norm 0.3519 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12852 | train 2.943024 | val 3.348853 | tok/s 2629301.11 | norm 0.3624 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12853 | train 2.963699 | val 3.348853 | tok/s 2627802.03 | norm 0.3995 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12854 | train 3.093448 | val 3.348853 | tok/s 2629253.89 | norm 0.3889 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12855 | train 2.969372 | val 3.348853 | tok/s 2631161.31 | norm 0.3950 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12856 | train 3.027375 | val 3.348853 | tok/s 2632343.18 | norm 0.4256 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12857 | train 2.911300 | val 3.348853 | tok/s 2628940.29 | norm 0.3754 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12858 | train 2.888920 | val 3.348853 | tok/s 2631255.89 | norm 0.3821 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12859 | train 2.979981 | val 3.348853 | tok/s 2631985.00 | norm 0.3720 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12860 | train 3.035098 | val 3.348853 | tok/s 2628169.14 | norm 0.3844 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12861 | train 3.014504 | val 3.348853 | tok/s 2630302.79 | norm 0.3718 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12862 | train 3.002383 | val 3.348853 | tok/s 2628246.70 | norm 0.3477 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12863 | train 2.936089 | val 3.348853 | tok/s 2628700.27 | norm 0.3634 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12864 | train 2.960404 | val 3.348853 | tok/s 2629309.43 | norm 0.3488 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12865 | train 2.946346 | val 3.348853 | tok/s 2628152.13 | norm 0.4206 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12866 | train 2.891962 | val 3.348853 | tok/s 2626937.00 | norm 0.4717 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12867 | train 2.995984 | val 3.348853 | tok/s 2628570.68 | norm 0.4155 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12868 | train 2.963651 | val 3.348853 | tok/s 2631475.26 | norm 0.4012 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12869 | train 2.941322 | val 3.348853 | tok/s 2631207.42 | norm 0.4127 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12870 | train 2.967869 | val 3.348853 | tok/s 2625845.74 | norm 0.3775 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12871 | train 2.855743 | val 3.348853 | tok/s 2629367.73 | norm 0.3832 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12872 | train 2.851032 | val 3.348853 | tok/s 2629714.87 | norm 0.3551 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12873 | train 3.094289 | val 3.348853 | tok/s 2629840.53 | norm 0.3779 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12874 | train 3.134846 | val 3.348853 | tok/s 2630799.79 | norm 0.4212 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12875 | train 2.918330 | val 3.348853 | tok/s 2632088.51 | norm 0.3908 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12876 | train 2.989544 | val 3.348853 | tok/s 2629893.27 | norm 0.3932 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12877 | train 3.068968 | val 3.348853 | tok/s 2626019.53 | norm 0.3984 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12878 | train 2.865259 | val 3.348853 | tok/s 2626413.28 | norm 0.4390 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12879 | train 2.988871 | val 3.348853 | tok/s 2629867.74 | norm 0.3823 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12880 | train 2.938123 | val 3.348853 | tok/s 2570288.05 | norm 0.3402 | dt 203.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12881 | train 3.027879 | val 3.348853 | tok/s 2590528.30 | norm 0.3412 | dt 202.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12882 | train 2.872724 | val 3.348853 | tok/s 2630735.64 | norm 0.3791 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12883 | train 2.800797 | val 3.348853 | tok/s 2629238.93 | norm 0.4073 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12884 | train 2.939138 | val 3.348853 | tok/s 2628940.93 | norm 0.3919 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12885 | train 2.997423 | val 3.348853 | tok/s 2615126.16 | norm 0.4132 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12886 | train 2.919404 | val 3.348853 | tok/s 2626959.37 | norm 0.4106 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12887 | train 3.088884 | val 3.348853 | tok/s 2631796.73 | norm 0.4085 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12888 | train 3.007234 | val 3.348853 | tok/s 2630963.07 | norm 0.3752 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12889 | train 3.078867 | val 3.348853 | tok/s 2631468.74 | norm 0.3989 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12890 | train 2.923965 | val 3.348853 | tok/s 2632192.23 | norm 0.4516 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12891 | train 2.906382 | val 3.348853 | tok/s 2632190.14 | norm 0.3848 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12892 | train 2.939853 | val 3.348853 | tok/s 2537453.86 | norm 0.3760 | dt 206.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12893 | train 2.941540 | val 3.348853 | tok/s 2630845.64 | norm 0.4302 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12894 | train 3.051990 | val 3.348853 | tok/s 2629794.36 | norm 0.4252 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12895 | train 2.971044 | val 3.348853 | tok/s 2633451.89 | norm 0.3804 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12896 | train 2.998521 | val 3.348853 | tok/s 2633832.23 | norm 0.3784 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12897 | train 3.000935 | val 3.348853 | tok/s 2634580.93 | norm 0.3767 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12898 | train 2.919964 | val 3.348853 | tok/s 2628362.40 | norm 0.4011 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12899 | train 3.022921 | val 3.348853 | tok/s 2626775.45 | norm 0.4129 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12900 | train 3.132696 | val 3.348853 | tok/s 2616708.58 | norm 0.3886 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12901 | train 2.926406 | val 3.348853 | tok/s 2629158.29 | norm 0.3434 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12902 | train 2.938644 | val 3.348853 | tok/s 2630408.01 | norm 0.3782 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12903 | train 2.938751 | val 3.348853 | tok/s 2630316.09 | norm 0.3823 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12904 | train 2.950924 | val 3.348853 | tok/s 2630316.46 | norm 0.3660 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12905 | train 2.996675 | val 3.348853 | tok/s 2631682.27 | norm 0.4253 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12906 | train 2.943278 | val 3.348853 | tok/s 2629197.37 | norm 0.4038 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12907 | train 2.967132 | val 3.348853 | tok/s 2627953.62 | norm 0.3715 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12908 | train 2.881407 | val 3.348853 | tok/s 2629134.79 | norm 0.3632 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12909 | train 3.031696 | val 3.348853 | tok/s 2633498.17 | norm 0.3812 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12910 | train 3.038245 | val 3.348853 | tok/s 2570746.78 | norm 0.4491 | dt 203.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12911 | train 3.034529 | val 3.348853 | tok/s 2567515.57 | norm 0.4554 | dt 204.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12912 | train 3.028268 | val 3.348853 | tok/s 2631323.86 | norm 0.4155 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12913 | train 2.957435 | val 3.348853 | tok/s 2628496.70 | norm 0.3839 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12914 | train 3.080340 | val 3.348853 | tok/s 2632678.71 | norm 0.3837 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12915 | train 3.030475 | val 3.348853 | tok/s 2632966.09 | norm 0.3825 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12916 | train 3.110084 | val 3.348853 | tok/s 2631080.36 | norm 0.4066 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12917 | train 3.017829 | val 3.348853 | tok/s 2630449.10 | norm 0.3808 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12918 | train 2.913624 | val 3.348853 | tok/s 2633623.66 | norm 0.3546 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12919 | train 3.040325 | val 3.348853 | tok/s 2630947.47 | norm 0.3729 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12920 | train 2.996852 | val 3.348853 | tok/s 2632909.08 | norm 0.3986 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12921 | train 2.944028 | val 3.348853 | tok/s 2629535.72 | norm 0.4070 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12922 | train 3.008068 | val 3.348853 | tok/s 2627391.00 | norm 0.4304 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12923 | train 2.970951 | val 3.348853 | tok/s 2622639.83 | norm 0.4023 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12924 | train 2.884188 | val 3.348853 | tok/s 2624630.58 | norm 0.4428 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12925 | train 2.951147 | val 3.348853 | tok/s 2630169.28 | norm 0.4502 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12926 | train 3.047175 | val 3.348853 | tok/s 2629199.79 | norm 0.3989 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12927 | train 3.038131 | val 3.348853 | tok/s 2628228.27 | norm 0.4015 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12928 | train 3.002315 | val 3.348853 | tok/s 2622662.74 | norm 0.4023 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12929 | train 3.142027 | val 3.348853 | tok/s 2630554.07 | norm 0.3965 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12930 | train 2.999234 | val 3.348853 | tok/s 2624232.84 | norm 0.4353 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12931 | train 2.992544 | val 3.348853 | tok/s 2626952.67 | norm 0.3741 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12932 | train 2.893127 | val 3.348853 | tok/s 2631308.04 | norm 0.3936 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12933 | train 3.148403 | val 3.348853 | tok/s 2629898.70 | norm 0.4319 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12934 | train 3.034451 | val 3.348853 | tok/s 2628435.98 | norm 0.4315 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12935 | train 2.871274 | val 3.348853 | tok/s 2625971.84 | norm 0.3407 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12936 | train 2.935032 | val 3.348853 | tok/s 2631405.08 | norm 0.3601 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12937 | train 3.007913 | val 3.348853 | tok/s 2634113.91 | norm 0.4255 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12938 | train 2.911037 | val 3.348853 | tok/s 2633112.63 | norm 0.3643 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12939 | train 2.973074 | val 3.348853 | tok/s 2623385.69 | norm 0.3698 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12940 | train 2.962558 | val 3.348853 | tok/s 2630615.98 | norm 0.4066 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12941 | train 3.107440 | val 3.348853 | tok/s 2629876.42 | norm 0.4282 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12942 | train 3.011020 | val 3.348853 | tok/s 2548396.07 | norm 0.4170 | dt 205.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12943 | train 2.883250 | val 3.348853 | tok/s 2573222.01 | norm 0.3565 | dt 203.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12944 | train 2.936484 | val 3.348853 | tok/s 2629285.64 | norm 0.3378 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12945 | train 3.049559 | val 3.348853 | tok/s 2635324.74 | norm 0.3565 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12946 | train 2.977091 | val 3.348853 | tok/s 2634036.75 | norm 0.4034 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12947 | train 2.950835 | val 3.348853 | tok/s 2635408.18 | norm 0.3931 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12948 | train 2.950081 | val 3.348853 | tok/s 2629580.21 | norm 0.4140 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12949 | train 2.983405 | val 3.348853 | tok/s 2631740.69 | norm 0.4182 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12950 | train 2.922825 | val 3.348853 | tok/s 2630593.03 | norm 0.4223 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12951 | train 3.041722 | val 3.348853 | tok/s 2628054.08 | norm 0.4005 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12952 | train 2.947947 | val 3.348853 | tok/s 2625440.21 | norm 0.3633 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12953 | train 2.993362 | val 3.348853 | tok/s 2629735.25 | norm 0.3665 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12954 | train 2.977281 | val 3.348853 | tok/s 2630444.50 | norm 0.4160 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12955 | train 2.970507 | val 3.348853 | tok/s 2630438.92 | norm 0.3906 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12956 | train 2.961260 | val 3.348853 | tok/s 2625640.12 | norm 0.4144 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12957 | train 3.043056 | val 3.348853 | tok/s 2623820.22 | norm 0.4063 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12958 | train 2.940050 | val 3.348853 | tok/s 2631389.35 | norm 0.3746 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12959 | train 3.054236 | val 3.348853 | tok/s 2630238.70 | norm 0.3918 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12960 | train 2.976312 | val 3.348853 | tok/s 2632242.29 | norm 0.4054 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12961 | train 2.994286 | val 3.348853 | tok/s 2633241.27 | norm 0.3921 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12962 | train 2.920360 | val 3.348853 | tok/s 2632217.20 | norm 0.3709 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12963 | train 3.111748 | val 3.348853 | tok/s 2624415.59 | norm 0.3636 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12964 | train 2.939318 | val 3.348853 | tok/s 2627567.07 | norm 0.3629 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12965 | train 3.049438 | val 3.348853 | tok/s 2628587.73 | norm 0.3841 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12966 | train 2.976298 | val 3.348853 | tok/s 2630666.88 | norm 0.4097 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12967 | train 2.991367 | val 3.348853 | tok/s 2627981.45 | norm 0.3622 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12968 | train 2.875458 | val 3.348853 | tok/s 2627651.22 | norm 0.3625 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12969 | train 3.061434 | val 3.348853 | tok/s 2633448.80 | norm 0.4114 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12970 | train 3.206496 | val 3.348853 | tok/s 2631988.72 | norm 0.4126 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12971 | train 2.945751 | val 3.348853 | tok/s 2633274.24 | norm 0.3979 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12972 | train 3.017540 | val 3.348853 | tok/s 2627728.05 | norm 0.3941 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12973 | train 2.990216 | val 3.348853 | tok/s 2629146.17 | norm 0.3830 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12974 | train 3.118674 | val 3.348853 | tok/s 2631953.81 | norm 0.3908 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12975 | train 3.015474 | val 3.348853 | tok/s 2631272.91 | norm 0.3915 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12976 | train 2.975012 | val 3.348853 | tok/s 2633577.65 | norm 0.4106 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12977 | train 2.945891 | val 3.348853 | tok/s 2634302.79 | norm 0.3800 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12978 | train 2.886814 | val 3.348853 | tok/s 2632526.54 | norm 0.3594 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12979 | train 3.036994 | val 3.348853 | tok/s 2631189.61 | norm 0.3876 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12980 | train 2.983392 | val 3.348853 | tok/s 2631510.75 | norm 0.3887 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12981 | train 2.931152 | val 3.348853 | tok/s 2632008.42 | norm 0.3945 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12982 | train 2.902478 | val 3.348853 | tok/s 2631341.20 | norm 0.3795 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12983 | train 2.973131 | val 3.348853 | tok/s 2627738.82 | norm 0.3486 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12984 | train 2.935912 | val 3.348853 | tok/s 2629120.05 | norm 0.3416 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12985 | train 2.948589 | val 3.348853 | tok/s 2630073.84 | norm 0.4053 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12986 | train 2.867601 | val 3.348853 | tok/s 2631796.49 | norm 0.4120 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12987 | train 3.029915 | val 3.348853 | tok/s 2628368.47 | norm 0.3868 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12988 | train 2.922044 | val 3.348853 | tok/s 2629307.19 | norm 0.3891 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12989 | train 2.855982 | val 3.348853 | tok/s 2632724.73 | norm 0.3985 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12990 | train 2.953355 | val 3.348853 | tok/s 2631314.82 | norm 0.3765 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12991 | train 3.045512 | val 3.348853 | tok/s 2631216.38 | norm 0.3862 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12992 | train 2.934361 | val 3.348853 | tok/s 2629690.67 | norm 0.4210 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12993 | train 3.002097 | val 3.348853 | tok/s 2630380.62 | norm 0.3956 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12994 | train 2.987318 | val 3.348853 | tok/s 2632456.07 | norm 0.4199 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12995 | train 3.204263 | val 3.348853 | tok/s 2630150.60 | norm 0.4724 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12996 | train 2.906976 | val 3.348853 | tok/s 2630272.44 | norm 0.3808 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12997 | train 3.045291 | val 3.348853 | tok/s 2632486.47 | norm 0.3496 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12998 | train 2.997578 | val 3.348853 | tok/s 2633324.73 | norm 0.3679 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 12999 | train 3.028339 | val 3.348853 | tok/s 2626313.19 | norm 0.3769 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 13000 | val_loss 3.340246 step 13000 | train 2.901407 | val 3.340246 | tok/s 2606870.69 | norm 0.3817 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13001 | train 2.936912 | val 3.340246 | tok/s 2628919.02 | norm 0.3695 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13002 | train 3.016647 | val 3.340246 | tok/s 2631569.66 | norm 0.3540 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13003 | train 2.988246 | val 3.340246 | tok/s 2633700.93 | norm 0.3372 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13004 | train 2.919265 | val 3.340246 | tok/s 2630444.88 | norm 0.3685 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13005 | train 2.996910 | val 3.340246 | tok/s 2628991.41 | norm 0.3634 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13006 | train 2.856536 | val 3.340246 | tok/s 2628621.97 | norm 0.4148 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13007 | train 2.977562 | val 3.340246 | tok/s 2550314.77 | norm 0.3846 | dt 205.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13008 | train 2.866535 | val 3.340246 | tok/s 2629321.79 | norm 0.3829 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13009 | train 3.046792 | val 3.340246 | tok/s 2615835.42 | norm 0.4042 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13010 | train 2.902138 | val 3.340246 | tok/s 2627208.84 | norm 0.3877 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13011 | train 2.998819 | val 3.340246 | tok/s 2629693.40 | norm 0.4074 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13012 | train 2.989094 | val 3.340246 | tok/s 2627434.99 | norm 0.3597 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13013 | train 2.937828 | val 3.340246 | tok/s 2627310.79 | norm 0.3515 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13014 | train 3.013741 | val 3.340246 | tok/s 2625614.41 | norm 0.3619 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13015 | train 3.003013 | val 3.340246 | tok/s 2629374.35 | norm 0.3958 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13016 | train 2.889005 | val 3.340246 | tok/s 2631795.29 | norm 0.3813 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13017 | train 2.997242 | val 3.340246 | tok/s 2629147.82 | norm 0.3714 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13018 | train 2.832264 | val 3.340246 | tok/s 2631443.55 | norm 0.6647 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13019 | train 2.957974 | val 3.340246 | tok/s 2631217.09 | norm 0.4462 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13020 | train 2.955301 | val 3.340246 | tok/s 2627903.67 | norm 0.3670 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13021 | train 3.045456 | val 3.340246 | tok/s 2630097.47 | norm 0.4018 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13022 | train 2.926203 | val 3.340246 | tok/s 2630774.67 | norm 0.3639 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13023 | train 2.995024 | val 3.340246 | tok/s 2480738.95 | norm 0.3793 | dt 211.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13024 | train 2.862359 | val 3.340246 | tok/s 2535334.19 | norm 0.3664 | dt 206.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13025 | train 3.000488 | val 3.340246 | tok/s 2630081.59 | norm 0.3437 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13026 | train 3.004544 | val 3.340246 | tok/s 2632438.63 | norm 0.3588 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13027 | train 3.025513 | val 3.340246 | tok/s 2631035.73 | norm 0.3978 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13028 | train 2.931509 | val 3.340246 | tok/s 2632157.06 | norm 0.3908 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13029 | train 2.969917 | val 3.340246 | tok/s 2634721.18 | norm 0.3808 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13030 | train 3.025576 | val 3.340246 | tok/s 2633643.84 | norm 0.3765 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13031 | train 2.974685 | val 3.340246 | tok/s 2630771.02 | norm 0.4087 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13032 | train 3.198799 | val 3.340246 | tok/s 2625781.45 | norm 0.4696 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13033 | train 2.881661 | val 3.340246 | tok/s 2611180.84 | norm 0.4376 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13034 | train 2.904870 | val 3.340246 | tok/s 2629310.50 | norm 0.3884 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13035 | train 3.046397 | val 3.340246 | tok/s 2630665.38 | norm 0.3753 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13036 | train 2.870741 | val 3.340246 | tok/s 2630239.19 | norm 0.3883 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13037 | train 3.033695 | val 3.340246 | tok/s 2629628.62 | norm 0.4180 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13038 | train 2.913502 | val 3.340246 | tok/s 2623859.66 | norm 0.3973 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13039 | train 2.938700 | val 3.340246 | tok/s 2628901.16 | norm 0.3759 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13040 | train 3.032274 | val 3.340246 | tok/s 2630502.21 | norm 0.4240 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13041 | train 2.919330 | val 3.340246 | tok/s 2629105.30 | norm 0.3821 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13042 | train 2.940753 | val 3.340246 | tok/s 2627481.57 | norm 0.3732 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13043 | train 2.900301 | val 3.340246 | tok/s 2630749.59 | norm 0.3819 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13044 | train 2.945025 | val 3.340246 | tok/s 2633346.96 | norm 0.3954 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13045 | train 2.918793 | val 3.340246 | tok/s 2628561.55 | norm 0.4043 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13046 | train 3.131334 | val 3.340246 | tok/s 2630765.97 | norm 0.4279 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13047 | train 2.933599 | val 3.340246 | tok/s 2627429.66 | norm 0.4053 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13048 | train 2.886100 | val 3.340246 | tok/s 2628992.27 | norm 0.3652 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13049 | train 2.905491 | val 3.340246 | tok/s 2627982.32 | norm 0.3474 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13050 | train 2.976423 | val 3.340246 | tok/s 2636615.67 | norm 0.3861 | dt 198.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13051 | train 2.983524 | val 3.340246 | tok/s 2634120.80 | norm 0.3826 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13052 | train 2.862837 | val 3.340246 | tok/s 2634299.70 | norm 0.3337 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13053 | train 3.136962 | val 3.340246 | tok/s 2634879.57 | norm 0.3598 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13054 | train 2.981202 | val 3.340246 | tok/s 2272670.16 | norm 0.3379 | dt 230.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13055 | train 2.942137 | val 3.340246 | tok/s 2632015.15 | norm 0.3567 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13056 | train 3.002110 | val 3.340246 | tok/s 2633890.92 | norm 0.3926 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13057 | train 2.982946 | val 3.340246 | tok/s 2627939.46 | norm 0.3958 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13058 | train 2.978622 | val 3.340246 | tok/s 2626584.47 | norm 0.4162 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13059 | train 2.969693 | val 3.340246 | tok/s 2627035.45 | norm 0.4042 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13060 | train 2.929351 | val 3.340246 | tok/s 2630021.68 | norm 0.3946 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13061 | train 3.036791 | val 3.340246 | tok/s 2629196.01 | norm 0.3996 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13062 | train 2.968971 | val 3.340246 | tok/s 2627317.94 | norm 0.4250 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13063 | train 2.946436 | val 3.340246 | tok/s 2628954.00 | norm 0.4346 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13064 | train 2.876857 | val 3.340246 | tok/s 2633454.23 | norm 0.3931 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13065 | train 2.977100 | val 3.340246 | tok/s 2632817.23 | norm 0.3951 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13066 | train 2.955775 | val 3.340246 | tok/s 2631419.98 | norm 0.4352 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13067 | train 3.023611 | val 3.340246 | tok/s 2634181.23 | norm 0.4076 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13068 | train 3.093776 | val 3.340246 | tok/s 2630043.70 | norm 0.4017 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13069 | train 3.015919 | val 3.340246 | tok/s 2626175.09 | norm 0.3792 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13070 | train 3.093192 | val 3.340246 | tok/s 2628405.81 | norm 0.3661 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13071 | train 2.941594 | val 3.340246 | tok/s 2627092.53 | norm 0.3764 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13072 | train 2.912692 | val 3.340246 | tok/s 2480139.20 | norm 0.3413 | dt 211.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13073 | train 3.107263 | val 3.340246 | tok/s 2629365.17 | norm 0.4204 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13074 | train 3.031220 | val 3.340246 | tok/s 2629746.14 | norm 0.4165 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13075 | train 2.973545 | val 3.340246 | tok/s 2629269.72 | norm 0.3678 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13076 | train 3.019190 | val 3.340246 | tok/s 2629643.61 | norm 0.3947 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13077 | train 3.033088 | val 3.340246 | tok/s 2630237.17 | norm 0.4040 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13078 | train 2.861529 | val 3.340246 | tok/s 2629429.65 | norm 0.3836 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13079 | train 2.978610 | val 3.340246 | tok/s 2625048.71 | norm 0.3806 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13080 | train 2.950809 | val 3.340246 | tok/s 2629794.77 | norm 0.3838 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13081 | train 2.933290 | val 3.340246 | tok/s 2631152.31 | norm 0.3818 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13082 | train 2.912369 | val 3.340246 | tok/s 2633200.45 | norm 0.3717 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13083 | train 2.948204 | val 3.340246 | tok/s 2625540.36 | norm 0.4020 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13084 | train 3.039711 | val 3.340246 | tok/s 2631732.36 | norm 0.3975 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13085 | train 3.042933 | val 3.340246 | tok/s 2626412.82 | norm 0.3657 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13086 | train 2.845851 | val 3.340246 | tok/s 2629957.43 | norm 0.4191 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13087 | train 3.013674 | val 3.340246 | tok/s 2631870.92 | norm 0.3500 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13088 | train 2.968258 | val 3.340246 | tok/s 2631541.84 | norm 0.5583 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13089 | train 2.956844 | val 3.340246 | tok/s 2629002.30 | norm 0.3657 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13090 | train 2.918696 | val 3.340246 | tok/s 2629696.59 | norm 0.3514 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13091 | train 2.991338 | val 3.340246 | tok/s 2616512.52 | norm 0.3632 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13092 | train 2.943832 | val 3.340246 | tok/s 2629962.87 | norm 0.3170 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13093 | train 2.936392 | val 3.340246 | tok/s 2628631.52 | norm 0.3535 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13094 | train 2.965755 | val 3.340246 | tok/s 2624854.46 | norm 0.3985 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13095 | train 2.881059 | val 3.340246 | tok/s 2634136.22 | norm 0.3713 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13096 | train 2.902560 | val 3.340246 | tok/s 2631436.75 | norm 0.3782 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13097 | train 2.968115 | val 3.340246 | tok/s 2632747.89 | norm 0.3767 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13098 | train 3.022221 | val 3.340246 | tok/s 2629820.57 | norm 0.3924 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13099 | train 3.034347 | val 3.340246 | tok/s 2627938.06 | norm 0.4101 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13100 | train 2.970749 | val 3.340246 | tok/s 2633238.53 | norm 0.4029 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13101 | train 3.019728 | val 3.340246 | tok/s 2625944.62 | norm 0.3895 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13102 | train 3.012262 | val 3.340246 | tok/s 2630205.53 | norm 0.4075 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13103 | train 2.950841 | val 3.340246 | tok/s 2632058.47 | norm 0.4029 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13104 | train 3.040473 | val 3.340246 | tok/s 2633339.32 | norm 0.3871 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13105 | train 2.968677 | val 3.340246 | tok/s 2630536.26 | norm 0.3690 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13106 | train 2.876962 | val 3.340246 | tok/s 2630359.80 | norm 0.3503 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13107 | train 3.022804 | val 3.340246 | tok/s 2628987.78 | norm 0.3694 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13108 | train 3.024105 | val 3.340246 | tok/s 2633835.81 | norm 0.3878 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13109 | train 3.077576 | val 3.340246 | tok/s 2631260.24 | norm 0.3947 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13110 | train 2.815536 | val 3.340246 | tok/s 2629931.81 | norm 0.3842 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13111 | train 2.941005 | val 3.340246 | tok/s 2629534.75 | norm 0.3632 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13112 | train 2.965986 | val 3.340246 | tok/s 2628777.35 | norm 0.4258 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13113 | train 3.017204 | val 3.340246 | tok/s 2630470.05 | norm 0.4198 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13114 | train 2.995418 | val 3.340246 | tok/s 2631369.76 | norm 0.3772 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13115 | train 3.095851 | val 3.340246 | tok/s 2627761.91 | norm 0.3484 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13116 | train 2.859950 | val 3.340246 | tok/s 2627673.35 | norm 0.3425 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13117 | train 3.201442 | val 3.340246 | tok/s 2630654.75 | norm 0.4199 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13118 | train 3.081021 | val 3.340246 | tok/s 2630918.90 | norm 0.4596 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13119 | train 2.942690 | val 3.340246 | tok/s 2628016.86 | norm 0.3829 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13120 | train 2.940323 | val 3.340246 | tok/s 2623699.23 | norm 0.3677 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13121 | train 2.931216 | val 3.340246 | tok/s 2621926.34 | norm 0.4320 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13122 | train 2.937754 | val 3.340246 | tok/s 2630147.66 | norm 0.4180 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13123 | train 2.944649 | val 3.340246 | tok/s 2631162.62 | norm 0.4036 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13124 | train 2.969362 | val 3.340246 | tok/s 2627715.83 | norm 0.4980 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13125 | train 2.914806 | val 3.340246 | tok/s 2631177.40 | norm 0.3908 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13126 | train 2.926259 | val 3.340246 | tok/s 2626684.62 | norm 0.3790 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13127 | train 2.977243 | val 3.340246 | tok/s 2631085.96 | norm 0.4066 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13128 | train 2.989267 | val 3.340246 | tok/s 2630446.49 | norm 0.3904 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13129 | train 2.916894 | val 3.340246 | tok/s 2631897.77 | norm 0.3612 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13130 | train 2.970205 | val 3.340246 | tok/s 2631382.11 | norm 0.3442 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13131 | train 2.982246 | val 3.340246 | tok/s 2631849.72 | norm 0.3715 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13132 | train 2.997672 | val 3.340246 | tok/s 2629429.89 | norm 0.3892 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13133 | train 2.925133 | val 3.340246 | tok/s 2630317.55 | norm 0.3675 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13134 | train 3.161222 | val 3.340246 | tok/s 2627886.45 | norm 0.3967 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13135 | train 2.908124 | val 3.340246 | tok/s 2630480.55 | norm 0.3969 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13136 | train 2.997642 | val 3.340246 | tok/s 2630045.51 | norm 0.3941 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13137 | train 2.988597 | val 3.340246 | tok/s 2627992.44 | norm 0.3918 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13138 | train 2.978654 | val 3.340246 | tok/s 2629921.66 | norm 0.3343 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13139 | train 2.949990 | val 3.340246 | tok/s 2632205.96 | norm 0.3649 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13140 | train 2.948494 | val 3.340246 | tok/s 2631053.81 | norm 0.3959 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13141 | train 2.969439 | val 3.340246 | tok/s 2631713.29 | norm 0.3632 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13142 | train 3.047599 | val 3.340246 | tok/s 2626932.38 | norm 0.3528 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13143 | train 2.962254 | val 3.340246 | tok/s 2630341.90 | norm 0.3684 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13144 | train 2.933546 | val 3.340246 | tok/s 2630040.44 | norm 0.3911 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13145 | train 2.818241 | val 3.340246 | tok/s 2631408.66 | norm 0.4168 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13146 | train 3.066993 | val 3.340246 | tok/s 2630391.91 | norm 0.4137 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13147 | train 2.945629 | val 3.340246 | tok/s 2629062.45 | norm 0.3827 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13148 | train 2.980827 | val 3.340246 | tok/s 2631628.24 | norm 0.3872 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13149 | train 2.905792 | val 3.340246 | tok/s 2628266.12 | norm 0.4057 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13150 | train 2.945700 | val 3.340246 | tok/s 2631167.81 | norm 0.3987 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13151 | train 2.914465 | val 3.340246 | tok/s 2626349.03 | norm 0.3850 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13152 | train 2.973424 | val 3.340246 | tok/s 2628276.62 | norm 0.3752 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13153 | train 2.923588 | val 3.340246 | tok/s 2628551.69 | norm 0.3559 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13154 | train 2.919406 | val 3.340246 | tok/s 2630679.63 | norm 0.3751 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13155 | train 3.088519 | val 3.340246 | tok/s 2631269.78 | norm 0.3893 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13156 | train 3.055525 | val 3.340246 | tok/s 2570115.74 | norm 0.3946 | dt 203.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13157 | train 2.843781 | val 3.340246 | tok/s 2632296.26 | norm 0.4074 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13158 | train 2.915549 | val 3.340246 | tok/s 2630912.01 | norm 0.3911 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13159 | train 2.994531 | val 3.340246 | tok/s 2629228.59 | norm 0.3844 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13160 | train 2.855479 | val 3.340246 | tok/s 2630481.89 | norm 0.4065 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13161 | train 3.075998 | val 3.340246 | tok/s 2630974.26 | norm 0.4011 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13162 | train 2.930295 | val 3.340246 | tok/s 2633011.36 | norm 0.3674 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13163 | train 2.974468 | val 3.340246 | tok/s 2633481.93 | norm 0.3737 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13164 | train 2.890768 | val 3.340246 | tok/s 2629661.27 | norm 0.3876 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13165 | train 2.941836 | val 3.340246 | tok/s 2630569.27 | norm 0.4065 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13166 | train 2.892495 | val 3.340246 | tok/s 2633709.93 | norm 0.3841 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13167 | train 2.993464 | val 3.340246 | tok/s 2634001.76 | norm 0.3582 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13168 | train 3.136026 | val 3.340246 | tok/s 2632823.66 | norm 0.3784 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13169 | train 3.027178 | val 3.340246 | tok/s 2629853.43 | norm 0.4191 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13170 | train 3.018524 | val 3.340246 | tok/s 2632641.57 | norm 0.4106 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13171 | train 2.906052 | val 3.340246 | tok/s 2627570.04 | norm 0.4023 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13172 | train 2.936208 | val 3.340246 | tok/s 2627430.05 | norm 0.3893 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13173 | train 2.918315 | val 3.340246 | tok/s 2628951.57 | norm 0.3778 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13174 | train 3.145304 | val 3.340246 | tok/s 2631707.63 | norm 0.4117 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13175 | train 2.984168 | val 3.340246 | tok/s 2632768.50 | norm 0.3757 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13176 | train 2.963427 | val 3.340246 | tok/s 2631697.67 | norm 0.3444 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13177 | train 3.006191 | val 3.340246 | tok/s 2628807.56 | norm 0.3880 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13178 | train 2.898242 | val 3.340246 | tok/s 2626145.67 | norm 0.4054 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13179 | train 2.947210 | val 3.340246 | tok/s 2626886.03 | norm 0.4146 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13180 | train 2.923937 | val 3.340246 | tok/s 2628538.26 | norm 0.4281 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13181 | train 2.929786 | val 3.340246 | tok/s 2627785.06 | norm 0.4052 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13182 | train 2.977820 | val 3.340246 | tok/s 2631120.17 | norm 0.3770 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13183 | train 3.093113 | val 3.340246 | tok/s 2628406.33 | norm 0.3774 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13184 | train 3.111916 | val 3.340246 | tok/s 2629307.10 | norm 0.4196 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13185 | train 2.997372 | val 3.340246 | tok/s 2629885.67 | norm 0.4294 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13186 | train 2.984675 | val 3.340246 | tok/s 2630579.99 | norm 0.3959 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13187 | train 2.987329 | val 3.340246 | tok/s 2629332.57 | norm 0.3922 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13188 | train 2.848990 | val 3.340246 | tok/s 2633248.09 | norm 0.3756 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13189 | train 2.999707 | val 3.340246 | tok/s 2628971.36 | norm 0.3999 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13190 | train 2.931686 | val 3.340246 | tok/s 2629439.44 | norm 0.4090 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13191 | train 3.095905 | val 3.340246 | tok/s 2626980.85 | norm 0.3768 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13192 | train 2.981462 | val 3.340246 | tok/s 2630441.59 | norm 0.3429 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13193 | train 3.308391 | val 3.340246 | tok/s 2627105.96 | norm 0.7279 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13194 | train 2.943611 | val 3.340246 | tok/s 2626674.52 | norm 0.3411 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13195 | train 2.976636 | val 3.340246 | tok/s 2628205.83 | norm 0.3153 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13196 | train 2.993650 | val 3.340246 | tok/s 2628479.51 | norm 0.3651 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13197 | train 2.921917 | val 3.340246 | tok/s 2628204.85 | norm 0.3648 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13198 | train 3.008105 | val 3.340246 | tok/s 2630608.59 | norm 0.3683 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13199 | train 3.317804 | val 3.340246 | tok/s 2624918.52 | norm 0.8502 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13200 | train 3.165251 | val 3.340246 | tok/s 2627723.72 | norm 0.4699 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13201 | train 2.986031 | val 3.340246 | tok/s 2626701.53 | norm 0.3331 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13202 | train 2.931989 | val 3.340246 | tok/s 2629961.55 | norm 0.3533 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13203 | train 2.977941 | val 3.340246 | tok/s 2631303.68 | norm 0.4036 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13204 | train 2.891263 | val 3.340246 | tok/s 2633950.26 | norm 0.3920 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13205 | train 2.879907 | val 3.340246 | tok/s 2631043.20 | norm 0.3696 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13206 | train 3.009151 | val 3.340246 | tok/s 2627998.29 | norm 0.3729 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13207 | train 2.972905 | val 3.340246 | tok/s 2630252.07 | norm 0.3622 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13208 | train 2.937720 | val 3.340246 | tok/s 2628160.44 | norm 0.3419 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13209 | train 3.036584 | val 3.340246 | tok/s 2606135.24 | norm 0.3680 | dt 201.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13210 | train 3.037635 | val 3.340246 | tok/s 2576338.34 | norm 0.3490 | dt 203.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13211 | train 2.908683 | val 3.340246 | tok/s 2629337.20 | norm 0.3492 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13212 | train 2.957804 | val 3.340246 | tok/s 2631373.62 | norm 0.3725 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13213 | train 3.001157 | val 3.340246 | tok/s 2630825.76 | norm 0.3899 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13214 | train 2.986348 | val 3.340246 | tok/s 2632651.90 | norm 0.3458 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13215 | train 3.013847 | val 3.340246 | tok/s 2633833.90 | norm 0.3432 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13216 | train 2.998302 | val 3.340246 | tok/s 2633233.08 | norm 0.3571 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13217 | train 3.001695 | val 3.340246 | tok/s 2627036.24 | norm 0.3810 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13218 | train 3.026153 | val 3.340246 | tok/s 2628701.34 | norm 0.4003 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13219 | train 2.844481 | val 3.340246 | tok/s 2630856.07 | norm 0.3774 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13220 | train 2.960930 | val 3.340246 | tok/s 2631720.50 | norm 0.3764 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13221 | train 2.907408 | val 3.340246 | tok/s 2631712.97 | norm 0.4102 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13222 | train 2.942568 | val 3.340246 | tok/s 2625243.02 | norm 0.4272 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13223 | train 2.896220 | val 3.340246 | tok/s 2630508.69 | norm 0.3891 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13224 | train 3.058252 | val 3.340246 | tok/s 2629124.83 | norm 0.4142 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13225 | train 2.942233 | val 3.340246 | tok/s 2630770.99 | norm 0.3511 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13226 | train 2.961344 | val 3.340246 | tok/s 2633371.64 | norm 0.3661 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13227 | train 3.116202 | val 3.340246 | tok/s 2629465.26 | norm 0.4363 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13228 | train 2.936590 | val 3.340246 | tok/s 2631011.69 | norm 0.3969 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13229 | train 2.837652 | val 3.340246 | tok/s 2630170.35 | norm 0.3667 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13230 | train 2.911905 | val 3.340246 | tok/s 2634388.80 | norm 0.4084 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13231 | train 2.989903 | val 3.340246 | tok/s 2628280.82 | norm 0.4485 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13232 | train 2.906960 | val 3.340246 | tok/s 2629274.91 | norm 0.3825 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13233 | train 2.930609 | val 3.340246 | tok/s 2628635.03 | norm 0.3851 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13234 | train 3.062005 | val 3.340246 | tok/s 2605624.30 | norm 0.4196 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13235 | train 3.135058 | val 3.340246 | tok/s 2626956.19 | norm 0.4456 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13236 | train 3.111560 | val 3.340246 | tok/s 2628408.00 | norm 0.6280 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13237 | train 2.883056 | val 3.340246 | tok/s 2631700.32 | norm 0.3786 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13238 | train 3.010547 | val 3.340246 | tok/s 2630940.92 | norm 0.3526 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13239 | train 3.009841 | val 3.340246 | tok/s 2624442.87 | norm 0.3597 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13240 | train 2.956721 | val 3.340246 | tok/s 2632236.13 | norm 0.3660 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13241 | train 2.997883 | val 3.340246 | tok/s 2628669.63 | norm 0.3902 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13242 | train 2.961015 | val 3.340246 | tok/s 2630417.02 | norm 0.3643 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13243 | train 3.116753 | val 3.340246 | tok/s 2634573.15 | norm 0.3950 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13244 | train 2.804928 | val 3.340246 | tok/s 2632501.07 | norm 0.4174 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13245 | train 2.988253 | val 3.340246 | tok/s 2633380.64 | norm 0.3372 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13246 | train 2.981255 | val 3.340246 | tok/s 2634723.58 | norm 0.3637 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13247 | train 2.895755 | val 3.340246 | tok/s 2628148.78 | norm 0.3950 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13248 | train 3.064186 | val 3.340246 | tok/s 2629018.20 | norm 0.3906 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13249 | train 2.932163 | val 3.340246 | tok/s 2627599.95 | norm 0.3951 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 13250 | val_loss 3.349404 step 13250 | train 3.025184 | val 3.349404 | tok/s 2614807.40 | norm 0.4392 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13251 | train 2.821551 | val 3.349404 | tok/s 2617128.92 | norm 0.3904 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13252 | train 2.985793 | val 3.349404 | tok/s 2627126.35 | norm 0.3727 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13253 | train 3.112473 | val 3.349404 | tok/s 2628748.68 | norm 0.4310 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13254 | train 2.944233 | val 3.349404 | tok/s 2626735.22 | norm 0.3849 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13255 | train 3.087581 | val 3.349404 | tok/s 2627695.73 | norm 0.3739 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13256 | train 3.079396 | val 3.349404 | tok/s 2613439.32 | norm 0.4001 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13257 | train 2.953801 | val 3.349404 | tok/s 2631719.34 | norm 0.3646 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13258 | train 2.973353 | val 3.349404 | tok/s 2627884.01 | norm 0.3551 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13259 | train 3.023637 | val 3.349404 | tok/s 2632137.94 | norm 0.3477 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13260 | train 3.073241 | val 3.349404 | tok/s 2634455.51 | norm 0.3697 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13261 | train 2.919548 | val 3.349404 | tok/s 2631139.88 | norm 0.4010 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13262 | train 2.940455 | val 3.349404 | tok/s 2631306.25 | norm 0.3894 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13263 | train 3.059402 | val 3.349404 | tok/s 2635147.36 | norm 0.3995 | dt 198.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13264 | train 3.127736 | val 3.349404 | tok/s 2632413.18 | norm 0.3933 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13265 | train 3.002726 | val 3.349404 | tok/s 2633458.95 | norm 0.3451 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13266 | train 2.895538 | val 3.349404 | tok/s 2630761.19 | norm 0.3391 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13267 | train 3.017602 | val 3.349404 | tok/s 2630612.92 | norm 0.3568 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13268 | train 2.907382 | val 3.349404 | tok/s 2634206.68 | norm 0.3933 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13269 | train 2.974716 | val 3.349404 | tok/s 2631742.24 | norm 0.4012 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13270 | train 3.179347 | val 3.349404 | tok/s 2632228.73 | norm 0.4669 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13271 | train 2.885388 | val 3.349404 | tok/s 2634935.32 | norm 0.3579 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13272 | train 2.940923 | val 3.349404 | tok/s 2630594.65 | norm 0.3805 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13273 | train 2.968909 | val 3.349404 | tok/s 2634862.32 | norm 0.4509 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13274 | train 2.911944 | val 3.349404 | tok/s 2630303.44 | norm 0.4130 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13275 | train 2.798663 | val 3.349404 | tok/s 2633295.94 | norm 0.4102 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13276 | train 3.032412 | val 3.349404 | tok/s 2628362.83 | norm 0.3953 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13277 | train 3.080060 | val 3.349404 | tok/s 2633036.19 | norm 0.4033 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13278 | train 2.913647 | val 3.349404 | tok/s 2631423.41 | norm 0.4276 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13279 | train 3.026649 | val 3.349404 | tok/s 2632248.70 | norm 0.3658 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13280 | train 3.015460 | val 3.349404 | tok/s 2635137.15 | norm 0.3745 | dt 198.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13281 | train 2.955446 | val 3.349404 | tok/s 2615895.77 | norm 0.3822 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13282 | train 2.974466 | val 3.349404 | tok/s 2630326.40 | norm 0.3900 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13283 | train 3.136696 | val 3.349404 | tok/s 2631539.73 | norm 0.4380 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13284 | train 2.958211 | val 3.349404 | tok/s 2628747.93 | norm 0.3784 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13285 | train 2.899030 | val 3.349404 | tok/s 2628322.78 | norm 0.3613 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13286 | train 2.999373 | val 3.349404 | tok/s 2631005.76 | norm 0.3902 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13287 | train 2.897590 | val 3.349404 | tok/s 2632976.74 | norm 0.3745 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13288 | train 2.967397 | val 3.349404 | tok/s 2628601.93 | norm 0.3613 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13289 | train 3.074284 | val 3.349404 | tok/s 2633159.87 | norm 0.3788 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13290 | train 3.089548 | val 3.349404 | tok/s 2632556.46 | norm 0.3531 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13291 | train 2.943368 | val 3.349404 | tok/s 2629974.29 | norm 0.3833 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13292 | train 2.922319 | val 3.349404 | tok/s 2631027.86 | norm 0.3985 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13293 | train 3.167653 | val 3.349404 | tok/s 2628103.50 | norm 0.5626 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13294 | train 2.918095 | val 3.349404 | tok/s 2632316.88 | norm 0.3792 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13295 | train 2.973544 | val 3.349404 | tok/s 2629557.34 | norm 0.3246 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13296 | train 2.938855 | val 3.349404 | tok/s 2631794.58 | norm 0.4101 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13297 | train 2.863638 | val 3.349404 | tok/s 2586442.04 | norm 0.4021 | dt 202.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13298 | train 2.887981 | val 3.349404 | tok/s 2457571.27 | norm 0.3913 | dt 213.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13299 | train 3.021314 | val 3.349404 | tok/s 2625878.23 | norm 0.4094 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13300 | train 2.951260 | val 3.349404 | tok/s 2630522.83 | norm 0.4008 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13301 | train 3.058211 | val 3.349404 | tok/s 2629562.86 | norm 0.4137 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13302 | train 2.954559 | val 3.349404 | tok/s 2632425.03 | norm 0.4245 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13303 | train 2.977275 | val 3.349404 | tok/s 2631265.57 | norm 0.3622 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13304 | train 2.893281 | val 3.349404 | tok/s 2630983.84 | norm 0.4267 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13305 | train 2.888793 | val 3.349404 | tok/s 2553296.45 | norm 0.4242 | dt 205.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13306 | train 3.022099 | val 3.349404 | tok/s 2633513.20 | norm 0.4081 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13307 | train 2.888337 | val 3.349404 | tok/s 2631318.88 | norm 0.3868 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13308 | train 2.913861 | val 3.349404 | tok/s 2631841.22 | norm 0.3879 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13309 | train 2.923841 | val 3.349404 | tok/s 2630347.76 | norm 0.3779 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13310 | train 2.971330 | val 3.349404 | tok/s 2630065.00 | norm 0.3755 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13311 | train 3.020714 | val 3.349404 | tok/s 2630755.19 | norm 0.3591 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13312 | train 2.919935 | val 3.349404 | tok/s 2602339.57 | norm 0.3752 | dt 201.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13313 | train 3.004345 | val 3.349404 | tok/s 2621071.62 | norm 0.3852 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13314 | train 2.942110 | val 3.349404 | tok/s 2633444.05 | norm 0.3664 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13315 | train 3.030776 | val 3.349404 | tok/s 2629431.91 | norm 0.3922 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13316 | train 3.214521 | val 3.349404 | tok/s 2628573.01 | norm 0.4637 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13317 | train 3.152350 | val 3.349404 | tok/s 2630882.14 | norm 0.3804 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13318 | train 3.057755 | val 3.349404 | tok/s 2631690.55 | norm 0.3446 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13319 | train 3.057611 | val 3.349404 | tok/s 2633607.87 | norm 0.3412 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13320 | train 3.007225 | val 3.349404 | tok/s 2627303.93 | norm 0.3471 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13321 | train 3.037408 | val 3.349404 | tok/s 2631295.58 | norm 0.3812 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13322 | train 3.013942 | val 3.349404 | tok/s 2628417.12 | norm 0.3980 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13323 | train 3.017028 | val 3.349404 | tok/s 2627615.88 | norm 0.3666 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13324 | train 2.805553 | val 3.349404 | tok/s 2628144.07 | norm 0.3860 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13325 | train 3.237495 | val 3.349404 | tok/s 2632477.91 | norm 0.4900 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13326 | train 3.001060 | val 3.349404 | tok/s 2629713.08 | norm 0.4301 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13327 | train 2.861385 | val 3.349404 | tok/s 2631493.57 | norm 0.3491 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13328 | train 3.009621 | val 3.349404 | tok/s 2626918.04 | norm 0.3994 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13329 | train 3.253773 | val 3.349404 | tok/s 2630189.59 | norm 0.4855 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13330 | train 3.062761 | val 3.349404 | tok/s 2633577.74 | norm 0.4273 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13331 | train 2.967479 | val 3.349404 | tok/s 2629852.54 | norm 0.3689 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13332 | train 3.035276 | val 3.349404 | tok/s 2631527.18 | norm 0.4015 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13333 | train 3.033508 | val 3.349404 | tok/s 2630148.58 | norm 0.3790 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13334 | train 2.952167 | val 3.349404 | tok/s 2626038.81 | norm 0.3767 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13335 | train 2.933270 | val 3.349404 | tok/s 2629213.57 | norm 0.3802 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13336 | train 3.000806 | val 3.349404 | tok/s 2630551.01 | norm 0.3491 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13337 | train 3.024076 | val 3.349404 | tok/s 2628485.74 | norm 0.3608 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13338 | train 2.997719 | val 3.349404 | tok/s 2631932.86 | norm 0.3755 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13339 | train 2.874456 | val 3.349404 | tok/s 2633339.05 | norm 0.3668 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13340 | train 2.912147 | val 3.349404 | tok/s 2631443.37 | norm 0.3767 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13341 | train 3.001241 | val 3.349404 | tok/s 2634062.07 | norm 0.4165 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13342 | train 2.920782 | val 3.349404 | tok/s 2627103.37 | norm 0.3827 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13343 | train 3.032413 | val 3.349404 | tok/s 2630315.43 | norm 0.3577 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13344 | train 3.013620 | val 3.349404 | tok/s 2630633.97 | norm 0.3889 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13345 | train 3.110099 | val 3.349404 | tok/s 2628064.00 | norm 0.4079 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13346 | train 3.027039 | val 3.349404 | tok/s 2630967.96 | norm 0.3755 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13347 | train 3.057443 | val 3.349404 | tok/s 2632270.62 | norm 0.3659 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13348 | train 2.923100 | val 3.349404 | tok/s 2626197.11 | norm 0.3606 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13349 | train 2.985241 | val 3.349404 | tok/s 2632492.41 | norm 0.3864 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13350 | train 2.921169 | val 3.349404 | tok/s 2631632.87 | norm 0.3871 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13351 | train 3.064143 | val 3.349404 | tok/s 2631331.81 | norm 0.4044 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13352 | train 2.898965 | val 3.349404 | tok/s 2630906.12 | norm 0.3427 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13353 | train 2.881870 | val 3.349404 | tok/s 2627738.12 | norm 0.3399 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13354 | train 3.016176 | val 3.349404 | tok/s 2631244.35 | norm 0.3844 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13355 | train 2.976704 | val 3.349404 | tok/s 2633909.68 | norm 0.3811 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13356 | train 2.973760 | val 3.349404 | tok/s 2632034.99 | norm 0.3969 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13357 | train 2.954623 | val 3.349404 | tok/s 2630853.22 | norm 0.4325 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13358 | train 2.984452 | val 3.349404 | tok/s 2629451.47 | norm 0.4183 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13359 | train 2.926314 | val 3.349404 | tok/s 2629751.76 | norm 0.3978 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13360 | train 2.937992 | val 3.349404 | tok/s 2610180.54 | norm 0.3875 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13361 | train 3.002104 | val 3.349404 | tok/s 2632490.66 | norm 0.3898 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13362 | train 2.931483 | val 3.349404 | tok/s 2632703.68 | norm 0.4131 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13363 | train 2.980480 | val 3.349404 | tok/s 2633906.82 | norm 0.4058 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13364 | train 2.950167 | val 3.349404 | tok/s 2633276.00 | norm 0.4033 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13365 | train 2.950446 | val 3.349404 | tok/s 2633589.58 | norm 0.3672 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13366 | train 2.953614 | val 3.349404 | tok/s 2634125.29 | norm 0.3774 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13367 | train 2.958319 | val 3.349404 | tok/s 2630761.47 | norm 0.4108 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13368 | train 3.061547 | val 3.349404 | tok/s 2631018.53 | norm 0.3997 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13369 | train 2.848760 | val 3.349404 | tok/s 2629954.29 | norm 0.4692 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13370 | train 3.014854 | val 3.349404 | tok/s 2626506.39 | norm 0.3891 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13371 | train 3.088640 | val 3.349404 | tok/s 2633080.45 | norm 0.3963 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13372 | train 2.905631 | val 3.349404 | tok/s 2630725.22 | norm 0.4659 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13373 | train 2.978902 | val 3.349404 | tok/s 2631759.71 | norm 0.3676 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13374 | train 2.986724 | val 3.349404 | tok/s 2597124.37 | norm 0.3551 | dt 201.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13375 | train 2.910742 | val 3.349404 | tok/s 2626916.28 | norm 0.3450 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13376 | train 2.845509 | val 3.349404 | tok/s 2626506.37 | norm 0.3785 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13377 | train 2.984442 | val 3.349404 | tok/s 2627039.35 | norm 0.4037 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13378 | train 3.011024 | val 3.349404 | tok/s 2628691.26 | norm 0.4123 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13379 | train 2.949277 | val 3.349404 | tok/s 2628720.90 | norm 0.4142 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13380 | train 2.973298 | val 3.349404 | tok/s 2630494.75 | norm 0.3913 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13381 | train 2.914071 | val 3.349404 | tok/s 2631689.60 | norm 0.3648 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13382 | train 3.003168 | val 3.349404 | tok/s 2630114.02 | norm 0.3847 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13383 | train 3.075963 | val 3.349404 | tok/s 2630445.05 | norm 0.3770 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13384 | train 2.893934 | val 3.349404 | tok/s 2629146.01 | norm 0.3613 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13385 | train 3.005970 | val 3.349404 | tok/s 2628874.92 | norm 0.3624 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13386 | train 2.952058 | val 3.349404 | tok/s 2630001.81 | norm 0.3758 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13387 | train 2.945350 | val 3.349404 | tok/s 2629594.28 | norm 0.3878 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13388 | train 2.905107 | val 3.349404 | tok/s 2629902.47 | norm 0.3717 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13389 | train 2.977775 | val 3.349404 | tok/s 2629913.58 | norm 0.3960 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13390 | train 2.944793 | val 3.349404 | tok/s 2630116.79 | norm 0.3960 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13391 | train 2.889639 | val 3.349404 | tok/s 2631525.29 | norm 0.4409 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13392 | train 2.931460 | val 3.349404 | tok/s 2437727.95 | norm 0.4289 | dt 215.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13393 | train 3.011281 | val 3.349404 | tok/s 2620764.90 | norm 0.3788 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13394 | train 2.892501 | val 3.349404 | tok/s 2631533.07 | norm 0.4064 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13395 | train 3.041073 | val 3.349404 | tok/s 2628025.00 | norm 0.4263 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13396 | train 3.112490 | val 3.349404 | tok/s 2629417.71 | norm 0.4116 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13397 | train 2.914630 | val 3.349404 | tok/s 2631681.95 | norm 0.3607 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13398 | train 2.930436 | val 3.349404 | tok/s 2632620.62 | norm 0.3711 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13399 | train 2.979388 | val 3.349404 | tok/s 2627996.68 | norm 0.4013 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13400 | train 2.859863 | val 3.349404 | tok/s 2482366.30 | norm 0.4036 | dt 211.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13401 | train 2.915822 | val 3.349404 | tok/s 2582717.80 | norm 0.3806 | dt 203.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13402 | train 2.906126 | val 3.349404 | tok/s 2629844.67 | norm 0.3839 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13403 | train 3.022035 | val 3.349404 | tok/s 2632630.07 | norm 0.3891 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13404 | train 2.950121 | val 3.349404 | tok/s 2632362.91 | norm 0.3732 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13405 | train 2.925532 | val 3.349404 | tok/s 2632194.00 | norm 0.3683 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13406 | train 3.107472 | val 3.349404 | tok/s 2633880.86 | norm 0.4469 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13407 | train 2.970076 | val 3.349404 | tok/s 2630574.74 | norm 0.4370 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13408 | train 2.862253 | val 3.349404 | tok/s 2625425.02 | norm 0.3834 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13409 | train 2.956902 | val 3.349404 | tok/s 2622381.36 | norm 0.4124 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13410 | train 3.028533 | val 3.349404 | tok/s 2631458.55 | norm 0.3849 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13411 | train 2.925250 | val 3.349404 | tok/s 2626794.10 | norm 0.3705 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13412 | train 3.064632 | val 3.349404 | tok/s 2628303.75 | norm 0.4358 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13413 | train 2.962423 | val 3.349404 | tok/s 2626259.44 | norm 0.3192 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13414 | train 2.884966 | val 3.349404 | tok/s 2629026.37 | norm 0.3351 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13415 | train 2.965003 | val 3.349404 | tok/s 2630489.66 | norm 0.3750 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13416 | train 2.919719 | val 3.349404 | tok/s 2627654.89 | norm 0.4018 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13417 | train 3.071539 | val 3.349404 | tok/s 2631087.37 | norm 0.3949 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13418 | train 3.018564 | val 3.349404 | tok/s 2632310.68 | norm 0.3713 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13419 | train 2.944973 | val 3.349404 | tok/s 2628426.79 | norm 0.3926 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13420 | train 2.922852 | val 3.349404 | tok/s 2626837.65 | norm 0.3721 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13421 | train 2.898946 | val 3.349404 | tok/s 2623191.91 | norm 0.3602 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13422 | train 3.016864 | val 3.349404 | tok/s 2623322.75 | norm 0.3780 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13423 | train 3.190995 | val 3.349404 | tok/s 2628056.92 | norm 0.4735 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13424 | train 3.161876 | val 3.349404 | tok/s 2628970.63 | norm 0.4502 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13425 | train 2.924205 | val 3.349404 | tok/s 2628671.45 | norm 0.3510 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13426 | train 3.142531 | val 3.349404 | tok/s 2624651.72 | norm 0.3726 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13427 | train 3.005740 | val 3.349404 | tok/s 2626296.40 | norm 0.4067 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13428 | train 3.036392 | val 3.349404 | tok/s 2626035.93 | norm 0.4191 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13429 | train 3.008634 | val 3.349404 | tok/s 2624983.02 | norm 0.4122 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13430 | train 2.938011 | val 3.349404 | tok/s 2624217.29 | norm 0.4091 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13431 | train 3.140208 | val 3.349404 | tok/s 2614920.30 | norm 0.4565 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13432 | train 2.991476 | val 3.349404 | tok/s 2625040.07 | norm 0.3787 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13433 | train 2.964396 | val 3.349404 | tok/s 2632451.66 | norm 0.3913 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13434 | train 3.031688 | val 3.349404 | tok/s 2627890.49 | norm 0.4107 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13435 | train 3.004474 | val 3.349404 | tok/s 2632730.57 | norm 0.3965 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13436 | train 2.933026 | val 3.349404 | tok/s 2630505.46 | norm 0.4257 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13437 | train 2.849793 | val 3.349404 | tok/s 2628168.29 | norm 0.4093 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13438 | train 2.949821 | val 3.349404 | tok/s 2627227.36 | norm 0.3983 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13439 | train 3.047625 | val 3.349404 | tok/s 2622575.35 | norm 0.3598 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13440 | train 3.029087 | val 3.349404 | tok/s 2629774.16 | norm 0.3530 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13441 | train 3.061383 | val 3.349404 | tok/s 2630581.86 | norm 0.3458 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13442 | train 3.005483 | val 3.349404 | tok/s 2633747.08 | norm 0.3423 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13443 | train 2.965666 | val 3.349404 | tok/s 2634117.39 | norm 0.4067 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13444 | train 2.927220 | val 3.349404 | tok/s 2630901.14 | norm 0.3890 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13445 | train 3.032884 | val 3.349404 | tok/s 2632210.69 | norm 0.3749 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13446 | train 2.995476 | val 3.349404 | tok/s 2626417.06 | norm 0.3727 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13447 | train 2.998027 | val 3.349404 | tok/s 2629593.33 | norm 0.3702 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13448 | train 3.023397 | val 3.349404 | tok/s 2629341.87 | norm 0.3745 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13449 | train 2.818325 | val 3.349404 | tok/s 2491716.46 | norm 0.4305 | dt 210.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13450 | train 3.069693 | val 3.349404 | tok/s 2624529.50 | norm 0.4202 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13451 | train 2.929726 | val 3.349404 | tok/s 2629761.78 | norm 0.3379 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13452 | train 2.945689 | val 3.349404 | tok/s 2621304.28 | norm 0.3405 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13453 | train 2.941742 | val 3.349404 | tok/s 2622908.20 | norm 0.3951 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13454 | train 2.976583 | val 3.349404 | tok/s 2630008.08 | norm 0.3873 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13455 | train 2.947690 | val 3.349404 | tok/s 2629213.89 | norm 0.3647 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13456 | train 2.905846 | val 3.349404 | tok/s 2522437.75 | norm 0.3483 | dt 207.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13457 | train 3.007357 | val 3.349404 | tok/s 2580518.36 | norm 0.4503 | dt 203.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13458 | train 2.877589 | val 3.349404 | tok/s 2631033.13 | norm 0.3835 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13459 | train 2.995908 | val 3.349404 | tok/s 2630720.93 | norm 0.4016 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13460 | train 2.888290 | val 3.349404 | tok/s 2632402.03 | norm 0.3801 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13461 | train 3.042651 | val 3.349404 | tok/s 2631548.20 | norm 0.3973 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13462 | train 2.906107 | val 3.349404 | tok/s 2634416.05 | norm 0.4077 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13463 | train 2.768570 | val 3.349404 | tok/s 2625961.84 | norm 0.3838 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13464 | train 3.053844 | val 3.349404 | tok/s 2631624.66 | norm 0.3659 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13465 | train 3.016518 | val 3.349404 | tok/s 2629062.37 | norm 0.3494 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13466 | train 2.893794 | val 3.349404 | tok/s 2627876.34 | norm 0.3452 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13467 | train 3.233498 | val 3.349404 | tok/s 2630004.25 | norm 0.4305 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13468 | train 3.035841 | val 3.349404 | tok/s 2630354.82 | norm 0.3706 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13469 | train 2.933647 | val 3.349404 | tok/s 2631914.93 | norm 0.3459 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13470 | train 2.900721 | val 3.349404 | tok/s 2627765.16 | norm 0.3971 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13471 | train 2.918290 | val 3.349404 | tok/s 2629975.89 | norm 0.4079 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13472 | train 2.965841 | val 3.349404 | tok/s 2626245.64 | norm 0.3863 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13473 | train 3.004751 | val 3.349404 | tok/s 2628073.24 | norm 0.3902 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13474 | train 2.985143 | val 3.349404 | tok/s 2627891.55 | norm 0.3721 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13475 | train 3.030509 | val 3.349404 | tok/s 2629802.13 | norm 0.3447 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13476 | train 2.983409 | val 3.349404 | tok/s 2631337.90 | norm 0.3716 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13477 | train 2.902530 | val 3.349404 | tok/s 2621365.36 | norm 0.4197 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13478 | train 3.034903 | val 3.349404 | tok/s 2622125.36 | norm 0.4240 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13479 | train 2.894062 | val 3.349404 | tok/s 2624531.26 | norm 0.4035 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13480 | train 2.979232 | val 3.349404 | tok/s 2631340.26 | norm 0.3895 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13481 | train 3.009133 | val 3.349404 | tok/s 2628726.13 | norm 0.3946 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13482 | train 2.885229 | val 3.349404 | tok/s 2633775.05 | norm 0.3867 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13483 | train 3.029489 | val 3.349404 | tok/s 2628658.98 | norm 0.4366 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13484 | train 2.972640 | val 3.349404 | tok/s 2627461.64 | norm 0.3711 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13485 | train 2.893172 | val 3.349404 | tok/s 2633291.31 | norm 0.3984 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13486 | train 2.972060 | val 3.349404 | tok/s 2628698.30 | norm 0.3860 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13487 | train 2.868256 | val 3.349404 | tok/s 2626264.32 | norm 0.3643 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13488 | train 2.993635 | val 3.349404 | tok/s 2631972.46 | norm 0.3854 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13489 | train 2.994974 | val 3.349404 | tok/s 2630858.37 | norm 0.3971 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13490 | train 2.959929 | val 3.349404 | tok/s 2630019.00 | norm 0.3899 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13491 | train 3.168736 | val 3.349404 | tok/s 2630992.87 | norm 0.4474 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13492 | train 3.040381 | val 3.349404 | tok/s 2628284.41 | norm 0.3973 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13493 | train 3.184536 | val 3.349404 | tok/s 2631223.86 | norm 0.4262 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13494 | train 2.969614 | val 3.349404 | tok/s 2630252.74 | norm 0.3676 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13495 | train 3.009167 | val 3.349404 | tok/s 2631267.24 | norm 0.3486 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13496 | train 2.960294 | val 3.349404 | tok/s 2632256.34 | norm 0.3747 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13497 | train 2.949383 | val 3.349404 | tok/s 2632204.48 | norm 0.3681 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13498 | train 3.081892 | val 3.349404 | tok/s 2632286.24 | norm 0.3841 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13499 | train 2.875892 | val 3.349404 | tok/s 2629517.34 | norm 0.4232 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 13500 | val_loss 3.340889 step 13500 | train 2.949793 | val 3.340889 | tok/s 2610890.96 | norm 0.3748 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13501 | train 2.914613 | val 3.340889 | tok/s 2628328.71 | norm 0.3261 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13502 | train 2.986611 | val 3.340889 | tok/s 2625222.92 | norm 0.3379 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13503 | train 2.993453 | val 3.340889 | tok/s 2630142.33 | norm 0.3643 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13504 | train 2.963581 | val 3.340889 | tok/s 2631082.83 | norm 0.3790 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13505 | train 3.136933 | val 3.340889 | tok/s 2629236.12 | norm 0.3806 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13506 | train 2.933474 | val 3.340889 | tok/s 2631936.03 | norm 0.3634 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13507 | train 2.930656 | val 3.340889 | tok/s 2628761.21 | norm 0.3608 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13508 | train 2.923826 | val 3.340889 | tok/s 2630376.86 | norm 0.3838 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13509 | train 3.034270 | val 3.340889 | tok/s 2490579.55 | norm 0.4138 | dt 210.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13510 | train 2.714546 | val 3.340889 | tok/s 2621176.89 | norm 0.4156 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13511 | train 2.924959 | val 3.340889 | tok/s 2629299.78 | norm 0.4094 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13512 | train 2.835085 | val 3.340889 | tok/s 2623833.07 | norm 0.3660 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13513 | train 2.888606 | val 3.340889 | tok/s 2631429.45 | norm 0.3478 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13514 | train 3.001259 | val 3.340889 | tok/s 2626621.69 | norm 0.3773 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13515 | train 2.873240 | val 3.340889 | tok/s 2628712.82 | norm 0.3736 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13516 | train 3.041747 | val 3.340889 | tok/s 2633373.21 | norm 0.3963 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13517 | train 2.997520 | val 3.340889 | tok/s 2634717.08 | norm 0.3740 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13518 | train 2.865487 | val 3.340889 | tok/s 2628817.63 | norm 0.3560 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13519 | train 2.955914 | val 3.340889 | tok/s 2626301.99 | norm 0.3564 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13520 | train 2.908778 | val 3.340889 | tok/s 2631197.45 | norm 0.4036 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13521 | train 3.015486 | val 3.340889 | tok/s 2631091.04 | norm 0.4004 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13522 | train 2.997285 | val 3.340889 | tok/s 2629809.45 | norm 0.4150 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13523 | train 2.998619 | val 3.340889 | tok/s 2631378.49 | norm 0.4122 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13524 | train 2.979856 | val 3.340889 | tok/s 2628785.25 | norm 0.3981 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13525 | train 3.088125 | val 3.340889 | tok/s 2630008.71 | norm 0.3930 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13526 | train 2.919865 | val 3.340889 | tok/s 2631436.76 | norm 0.3991 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13527 | train 2.949629 | val 3.340889 | tok/s 2625621.63 | norm 0.4225 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13528 | train 3.051011 | val 3.340889 | tok/s 2622955.16 | norm 0.3965 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13529 | train 2.911280 | val 3.340889 | tok/s 2623259.18 | norm 0.3906 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13530 | train 2.919057 | val 3.340889 | tok/s 2628708.27 | norm 0.3826 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13531 | train 3.001359 | val 3.340889 | tok/s 2627658.55 | norm 0.3774 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13532 | train 2.948262 | val 3.340889 | tok/s 2625029.52 | norm 0.3693 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13533 | train 2.994264 | val 3.340889 | tok/s 2629632.51 | norm 0.3601 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13534 | train 3.016044 | val 3.340889 | tok/s 2624419.56 | norm 0.3946 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13535 | train 3.034307 | val 3.340889 | tok/s 2631120.83 | norm 0.4002 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13536 | train 2.977628 | val 3.340889 | tok/s 2628873.59 | norm 0.4277 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13537 | train 2.915527 | val 3.340889 | tok/s 2628724.97 | norm 0.3656 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13538 | train 3.028004 | val 3.340889 | tok/s 2627655.77 | norm 0.3314 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13539 | train 2.953220 | val 3.340889 | tok/s 2628382.36 | norm 0.3565 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13540 | train 3.067630 | val 3.340889 | tok/s 2626702.76 | norm 0.3729 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13541 | train 2.995329 | val 3.340889 | tok/s 2626709.85 | norm 0.3588 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13542 | train 3.010163 | val 3.340889 | tok/s 2628974.48 | norm 0.3555 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13543 | train 3.070334 | val 3.340889 | tok/s 2627638.34 | norm 0.3748 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13544 | train 2.972038 | val 3.340889 | tok/s 2631766.83 | norm 0.3954 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13545 | train 2.937568 | val 3.340889 | tok/s 2631707.13 | norm 0.3823 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13546 | train 2.817833 | val 3.340889 | tok/s 2619743.13 | norm 0.3526 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13547 | train 3.004139 | val 3.340889 | tok/s 2621678.11 | norm 0.3855 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13548 | train 3.081272 | val 3.340889 | tok/s 2635417.66 | norm 0.3640 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13549 | train 2.958575 | val 3.340889 | tok/s 2624414.85 | norm 0.3267 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13550 | train 3.021316 | val 3.340889 | tok/s 2624788.02 | norm 0.3591 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13551 | train 3.127717 | val 3.340889 | tok/s 2619589.73 | norm 0.3737 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13552 | train 2.921391 | val 3.340889 | tok/s 2625025.11 | norm 0.3905 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13553 | train 3.024581 | val 3.340889 | tok/s 2628384.36 | norm 0.3797 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13554 | train 2.965037 | val 3.340889 | tok/s 2628553.25 | norm 0.3925 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13555 | train 3.081091 | val 3.340889 | tok/s 2631081.60 | norm 0.4450 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13556 | train 2.928570 | val 3.340889 | tok/s 2628898.17 | norm 0.4200 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13557 | train 2.885908 | val 3.340889 | tok/s 2629787.48 | norm 0.4308 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13558 | train 3.016182 | val 3.340889 | tok/s 2630151.80 | norm 0.4612 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13559 | train 3.063502 | val 3.340889 | tok/s 2624975.89 | norm 0.3969 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13560 | train 2.937708 | val 3.340889 | tok/s 2630045.46 | norm 0.3762 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13561 | train 3.081623 | val 3.340889 | tok/s 2628043.00 | norm 0.4297 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13562 | train 2.921161 | val 3.340889 | tok/s 2627349.49 | norm 0.4046 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13563 | train 2.990840 | val 3.340889 | tok/s 2579051.08 | norm 0.3896 | dt 203.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13564 | train 2.996104 | val 3.340889 | tok/s 2623909.27 | norm 0.3971 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13565 | train 2.929323 | val 3.340889 | tok/s 2631314.43 | norm 0.3816 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13566 | train 2.972872 | val 3.340889 | tok/s 2625342.15 | norm 0.3668 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13567 | train 2.915134 | val 3.340889 | tok/s 2625767.00 | norm 0.3598 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13568 | train 2.873074 | val 3.340889 | tok/s 2626107.13 | norm 0.3889 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13569 | train 3.119708 | val 3.340889 | tok/s 2627512.15 | norm 0.4409 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13570 | train 3.019011 | val 3.340889 | tok/s 2631565.51 | norm 0.3901 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13571 | train 3.085710 | val 3.340889 | tok/s 2623266.07 | norm 0.3942 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13572 | train 2.797886 | val 3.340889 | tok/s 2626379.16 | norm 0.3920 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13573 | train 3.020954 | val 3.340889 | tok/s 2630052.99 | norm 0.3764 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13574 | train 2.943148 | val 3.340889 | tok/s 2629193.93 | norm 0.4033 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13575 | train 3.017453 | val 3.340889 | tok/s 2631924.52 | norm 0.4281 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13576 | train 2.997120 | val 3.340889 | tok/s 2624276.52 | norm 0.4081 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13577 | train 3.059966 | val 3.340889 | tok/s 2627726.03 | norm 0.3882 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13578 | train 2.915945 | val 3.340889 | tok/s 2626477.43 | norm 0.3830 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13579 | train 2.964867 | val 3.340889 | tok/s 2627421.26 | norm 0.4783 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13580 | train 2.823892 | val 3.340889 | tok/s 2631205.48 | norm 0.4205 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13581 | train 2.989427 | val 3.340889 | tok/s 2624915.91 | norm 0.3648 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13582 | train 2.911051 | val 3.340889 | tok/s 2624659.95 | norm 0.3948 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13583 | train 2.931387 | val 3.340889 | tok/s 2630608.43 | norm 0.4013 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13584 | train 2.896886 | val 3.340889 | tok/s 2621422.91 | norm 0.4013 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13585 | train 2.915173 | val 3.340889 | tok/s 2630456.78 | norm 0.3817 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13586 | train 3.035689 | val 3.340889 | tok/s 2628257.35 | norm 0.3745 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13587 | train 2.965348 | val 3.340889 | tok/s 2627218.24 | norm 0.4068 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13588 | train 3.026893 | val 3.340889 | tok/s 2623894.81 | norm 0.3642 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13589 | train 3.013144 | val 3.340889 | tok/s 2628005.85 | norm 0.3874 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13590 | train 2.891007 | val 3.340889 | tok/s 2627039.28 | norm 0.3897 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13591 | train 3.040619 | val 3.340889 | tok/s 2629905.90 | norm 0.3861 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13592 | train 2.983697 | val 3.340889 | tok/s 2632093.37 | norm 0.4119 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13593 | train 2.930725 | val 3.340889 | tok/s 2630590.01 | norm 0.3834 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13594 | train 2.954355 | val 3.340889 | tok/s 2629309.36 | norm 0.4201 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13595 | train 2.853857 | val 3.340889 | tok/s 2628758.36 | norm 0.3989 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13596 | train 2.906779 | val 3.340889 | tok/s 2627461.34 | norm 0.3708 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13597 | train 2.993210 | val 3.340889 | tok/s 2634518.24 | norm 0.4089 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13598 | train 3.056238 | val 3.340889 | tok/s 2624847.27 | norm 0.3739 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13599 | train 2.880053 | val 3.340889 | tok/s 2626307.60 | norm 0.3579 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13600 | train 2.905764 | val 3.340889 | tok/s 2618298.26 | norm 0.3888 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13601 | train 3.120435 | val 3.340889 | tok/s 2623385.11 | norm 0.3729 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13602 | train 3.118980 | val 3.340889 | tok/s 2630288.13 | norm 0.3730 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13603 | train 2.975348 | val 3.340889 | tok/s 2627312.90 | norm 0.4561 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13604 | train 2.986274 | val 3.340889 | tok/s 2630136.39 | norm 0.3913 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13605 | train 2.987529 | val 3.340889 | tok/s 2628850.40 | norm 0.3540 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13606 | train 3.119074 | val 3.340889 | tok/s 2626423.97 | norm 0.4168 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13607 | train 2.982602 | val 3.340889 | tok/s 2615878.07 | norm 0.3732 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13608 | train 2.997301 | val 3.340889 | tok/s 2621548.18 | norm 0.3708 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13609 | train 2.960891 | val 3.340889 | tok/s 2626865.17 | norm 0.3864 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13610 | train 2.901795 | val 3.340889 | tok/s 2630946.50 | norm 0.3331 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13611 | train 2.890230 | val 3.340889 | tok/s 2626458.79 | norm 0.3629 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13612 | train 2.979362 | val 3.340889 | tok/s 2625998.17 | norm 0.3667 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13613 | train 3.151455 | val 3.340889 | tok/s 2619172.11 | norm 0.4440 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13614 | train 3.055710 | val 3.340889 | tok/s 2598557.18 | norm 0.3972 | dt 201.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13615 | train 2.945245 | val 3.340889 | tok/s 2631449.81 | norm 0.3528 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13616 | train 2.968769 | val 3.340889 | tok/s 2628706.97 | norm 0.3654 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13617 | train 2.975811 | val 3.340889 | tok/s 2621790.76 | norm 0.3730 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13618 | train 3.038686 | val 3.340889 | tok/s 2603993.62 | norm 0.3692 | dt 201.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13619 | train 2.974110 | val 3.340889 | tok/s 2627807.74 | norm 0.3592 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13620 | train 2.967410 | val 3.340889 | tok/s 2628327.02 | norm 0.3739 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13621 | train 2.945550 | val 3.340889 | tok/s 2629522.02 | norm 0.3887 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13622 | train 3.083716 | val 3.340889 | tok/s 2630150.80 | norm 0.4138 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13623 | train 2.889348 | val 3.340889 | tok/s 2628194.05 | norm 0.4104 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13624 | train 3.033758 | val 3.340889 | tok/s 2628957.21 | norm 0.3997 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13625 | train 2.921244 | val 3.340889 | tok/s 2612174.76 | norm 0.3981 | dt 200.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13626 | train 3.016976 | val 3.340889 | tok/s 2630653.31 | norm 0.3805 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13627 | train 2.888411 | val 3.340889 | tok/s 2630453.60 | norm 0.4857 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13628 | train 2.938194 | val 3.340889 | tok/s 2624076.63 | norm 0.3901 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13629 | train 2.873447 | val 3.340889 | tok/s 2628167.89 | norm 0.3680 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13630 | train 3.222038 | val 3.340889 | tok/s 2498500.34 | norm 0.3929 | dt 209.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13631 | train 3.014751 | val 3.340889 | tok/s 2474898.73 | norm 0.3921 | dt 211.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13632 | train 2.915378 | val 3.340889 | tok/s 2631779.01 | norm 0.3592 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13633 | train 2.978219 | val 3.340889 | tok/s 2624814.63 | norm 0.3661 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13634 | train 3.019616 | val 3.340889 | tok/s 2626568.05 | norm 0.3795 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13635 | train 3.066365 | val 3.340889 | tok/s 2630902.34 | norm 0.3794 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13636 | train 3.034635 | val 3.340889 | tok/s 2628687.17 | norm 0.3779 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13637 | train 2.967927 | val 3.340889 | tok/s 2631495.28 | norm 0.4061 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13638 | train 2.950490 | val 3.340889 | tok/s 2630158.45 | norm 0.4452 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13639 | train 2.977837 | val 3.340889 | tok/s 2630257.61 | norm 0.4124 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13640 | train 3.010433 | val 3.340889 | tok/s 2625745.91 | norm 0.3838 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13641 | train 2.970708 | val 3.340889 | tok/s 2623294.10 | norm 0.4143 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13642 | train 3.012246 | val 3.340889 | tok/s 2632292.15 | norm 0.4091 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13643 | train 2.944459 | val 3.340889 | tok/s 2629569.76 | norm 0.3585 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13644 | train 2.981936 | val 3.340889 | tok/s 2628879.98 | norm 0.2995 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13645 | train 2.886581 | val 3.340889 | tok/s 2631193.58 | norm 0.3159 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13646 | train 3.075387 | val 3.340889 | tok/s 2632819.69 | norm 0.3773 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13647 | train 2.947403 | val 3.340889 | tok/s 2633077.78 | norm 0.3928 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13648 | train 2.973793 | val 3.340889 | tok/s 2630393.62 | norm 0.3921 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13649 | train 2.874179 | val 3.340889 | tok/s 2630288.94 | norm 0.4053 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13650 | train 2.978291 | val 3.340889 | tok/s 2620071.59 | norm 0.4276 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13651 | train 2.963809 | val 3.340889 | tok/s 2622219.14 | norm 0.3747 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13652 | train 2.965794 | val 3.340889 | tok/s 2598321.90 | norm 0.4123 | dt 201.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13653 | train 3.020238 | val 3.340889 | tok/s 2628259.04 | norm 0.4249 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13654 | train 2.989562 | val 3.340889 | tok/s 2631012.03 | norm 0.3680 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13655 | train 2.958513 | val 3.340889 | tok/s 2625448.13 | norm 0.3650 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13656 | train 2.898520 | val 3.340889 | tok/s 2628587.83 | norm 0.3894 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13657 | train 2.921684 | val 3.340889 | tok/s 2633318.03 | norm 0.4118 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13658 | train 3.099762 | val 3.340889 | tok/s 2631730.59 | norm 0.3899 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13659 | train 3.003004 | val 3.340889 | tok/s 2628179.57 | norm 0.4234 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13660 | train 3.050264 | val 3.340889 | tok/s 2553872.58 | norm 0.4035 | dt 205.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13661 | train 2.939188 | val 3.340889 | tok/s 2626470.49 | norm 0.4259 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13662 | train 2.954057 | val 3.340889 | tok/s 2630166.18 | norm 0.4060 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13663 | train 2.925142 | val 3.340889 | tok/s 2625591.94 | norm 0.3769 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13664 | train 2.939573 | val 3.340889 | tok/s 2626532.02 | norm 0.3987 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13665 | train 2.964042 | val 3.340889 | tok/s 2619086.65 | norm 0.4170 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13666 | train 3.061564 | val 3.340889 | tok/s 2625663.58 | norm 0.3745 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13667 | train 2.989457 | val 3.340889 | tok/s 2627029.98 | norm 0.3939 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13668 | train 3.071547 | val 3.340889 | tok/s 2630165.38 | norm 0.3996 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13669 | train 2.986071 | val 3.340889 | tok/s 2632117.32 | norm 0.4327 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13670 | train 3.081085 | val 3.340889 | tok/s 2634800.19 | norm 0.3918 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13671 | train 2.928027 | val 3.340889 | tok/s 2457370.02 | norm 0.3781 | dt 213.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13672 | train 3.135789 | val 3.340889 | tok/s 2605488.15 | norm 0.3970 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13673 | train 2.913882 | val 3.340889 | tok/s 2625132.53 | norm 0.3650 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13674 | train 2.938756 | val 3.340889 | tok/s 2631710.30 | norm 0.3713 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13675 | train 3.044084 | val 3.340889 | tok/s 2626857.33 | norm 0.4126 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13676 | train 3.026896 | val 3.340889 | tok/s 2627589.41 | norm 0.3864 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13677 | train 2.905078 | val 3.340889 | tok/s 2494487.14 | norm 0.3558 | dt 210.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13678 | train 2.839313 | val 3.340889 | tok/s 2573893.93 | norm 0.3812 | dt 203.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13679 | train 2.910265 | val 3.340889 | tok/s 2630787.94 | norm 0.3630 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13680 | train 2.936669 | val 3.340889 | tok/s 2629215.13 | norm 0.3339 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13681 | train 2.931134 | val 3.340889 | tok/s 2626753.90 | norm 0.3884 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13682 | train 3.082515 | val 3.340889 | tok/s 2628826.83 | norm 0.4085 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13683 | train 2.920418 | val 3.340889 | tok/s 2628663.12 | norm 0.3780 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13684 | train 2.878247 | val 3.340889 | tok/s 2629027.41 | norm 0.4103 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13685 | train 2.943702 | val 3.340889 | tok/s 2626882.85 | norm 0.4299 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13686 | train 3.021639 | val 3.340889 | tok/s 2630378.45 | norm 0.4135 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13687 | train 2.907621 | val 3.340889 | tok/s 2631107.53 | norm 0.4163 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13688 | train 2.997432 | val 3.340889 | tok/s 2628949.48 | norm 0.3376 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13689 | train 3.026463 | val 3.340889 | tok/s 2628172.14 | norm 0.3499 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13690 | train 3.128226 | val 3.340889 | tok/s 2629577.93 | norm 0.3941 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13691 | train 3.016769 | val 3.340889 | tok/s 2628677.37 | norm 0.4244 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13692 | train 2.965259 | val 3.340889 | tok/s 2629903.99 | norm 0.3783 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13693 | train 2.898794 | val 3.340889 | tok/s 2628426.30 | norm 0.4073 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13694 | train 2.902272 | val 3.340889 | tok/s 2630416.14 | norm 0.4212 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13695 | train 2.910645 | val 3.340889 | tok/s 2630965.45 | norm 0.3561 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13696 | train 2.914320 | val 3.340889 | tok/s 2631305.03 | norm 0.3515 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13697 | train 2.972495 | val 3.340889 | tok/s 2628758.94 | norm 0.4066 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13698 | train 3.244566 | val 3.340889 | tok/s 2627659.79 | norm 0.4988 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13699 | train 2.983475 | val 3.340889 | tok/s 2626678.54 | norm 0.4133 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13700 | train 2.831397 | val 3.340889 | tok/s 2628800.47 | norm 0.3268 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13701 | train 3.042444 | val 3.340889 | tok/s 2632344.38 | norm 0.4093 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13702 | train 2.846427 | val 3.340889 | tok/s 2630417.13 | norm 0.3944 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13703 | train 3.052469 | val 3.340889 | tok/s 2600391.70 | norm 0.3909 | dt 201.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13704 | train 2.986306 | val 3.340889 | tok/s 2629651.30 | norm 0.4414 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13705 | train 3.021979 | val 3.340889 | tok/s 2627861.06 | norm 0.4333 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13706 | train 2.847467 | val 3.340889 | tok/s 2625729.72 | norm 0.3983 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13707 | train 3.057367 | val 3.340889 | tok/s 2623511.54 | norm 0.4162 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13708 | train 3.034980 | val 3.340889 | tok/s 2624360.62 | norm 0.4386 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13709 | train 2.916861 | val 3.340889 | tok/s 2630108.89 | norm 0.4108 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13710 | train 3.042892 | val 3.340889 | tok/s 2629091.95 | norm 0.4106 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13711 | train 3.032271 | val 3.340889 | tok/s 2632268.35 | norm 0.4021 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13712 | train 2.960942 | val 3.340889 | tok/s 2626072.09 | norm 0.3944 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13713 | train 2.940747 | val 3.340889 | tok/s 2630711.13 | norm 0.3871 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13714 | train 2.984974 | val 3.340889 | tok/s 2624009.98 | norm 0.3858 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13715 | train 3.019977 | val 3.340889 | tok/s 2603369.70 | norm 0.3650 | dt 201.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13716 | train 2.873649 | val 3.340889 | tok/s 2622207.48 | norm 0.3402 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13717 | train 3.076021 | val 3.340889 | tok/s 2530238.69 | norm 0.3690 | dt 207.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13718 | train 3.042290 | val 3.340889 | tok/s 2504739.45 | norm 0.4210 | dt 209.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13719 | train 2.899421 | val 3.340889 | tok/s 2619934.65 | norm 0.4337 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13720 | train 3.038384 | val 3.340889 | tok/s 2623103.74 | norm 0.3938 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13721 | train 2.995990 | val 3.340889 | tok/s 2625571.43 | norm 0.3801 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13722 | train 2.983214 | val 3.340889 | tok/s 2625690.26 | norm 0.3856 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13723 | train 2.790752 | val 3.340889 | tok/s 2628489.89 | norm 0.3835 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13724 | train 3.102636 | val 3.340889 | tok/s 2620232.41 | norm 0.3832 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13725 | train 2.969627 | val 3.340889 | tok/s 2537139.19 | norm 0.3746 | dt 206.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13726 | train 3.106125 | val 3.340889 | tok/s 2628252.36 | norm 0.3878 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13727 | train 3.015607 | val 3.340889 | tok/s 2622848.18 | norm 0.4287 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13728 | train 2.988258 | val 3.340889 | tok/s 2628168.39 | norm 0.4006 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13729 | train 3.104045 | val 3.340889 | tok/s 2627363.14 | norm 0.3966 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13730 | train 2.934579 | val 3.340889 | tok/s 2627848.85 | norm 0.3878 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13731 | train 3.037025 | val 3.340889 | tok/s 2627846.27 | norm 0.3582 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13732 | train 3.080636 | val 3.340889 | tok/s 2626651.04 | norm 0.3930 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13733 | train 2.924728 | val 3.340889 | tok/s 2626639.02 | norm 0.3768 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13734 | train 3.031769 | val 3.340889 | tok/s 2623398.46 | norm 0.3592 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13735 | train 2.994976 | val 3.340889 | tok/s 2628324.28 | norm 0.3778 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13736 | train 3.051446 | val 3.340889 | tok/s 2630388.90 | norm 0.3918 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13737 | train 3.023090 | val 3.340889 | tok/s 2623574.72 | norm 0.3684 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13738 | train 2.955056 | val 3.340889 | tok/s 2625072.69 | norm 0.3898 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13739 | train 2.981390 | val 3.340889 | tok/s 2627586.13 | norm 0.3810 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13740 | train 2.943922 | val 3.340889 | tok/s 2627237.62 | norm 0.3716 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13741 | train 3.070404 | val 3.340889 | tok/s 2628269.11 | norm 0.3585 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13742 | train 2.930258 | val 3.340889 | tok/s 2628883.63 | norm 0.3679 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13743 | train 2.991842 | val 3.340889 | tok/s 2626150.87 | norm 0.4044 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13744 | train 2.945534 | val 3.340889 | tok/s 2626440.85 | norm 0.3914 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13745 | train 2.979280 | val 3.340889 | tok/s 2627408.88 | norm 0.3674 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13746 | train 2.971801 | val 3.340889 | tok/s 2625236.20 | norm 0.3586 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13747 | train 2.924532 | val 3.340889 | tok/s 2626943.50 | norm 0.3779 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13748 | train 2.949546 | val 3.340889 | tok/s 2627622.95 | norm 0.3641 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13749 | train 2.929290 | val 3.340889 | tok/s 2619339.63 | norm 0.3428 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 13750 | val_loss 3.339385 step 13750 | train 2.920941 | val 3.339385 | tok/s 2615539.46 | norm 0.3668 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13751 | train 3.011622 | val 3.339385 | tok/s 2623109.36 | norm 0.3875 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13752 | train 2.876667 | val 3.339385 | tok/s 2627258.59 | norm 0.4156 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13753 | train 3.055497 | val 3.339385 | tok/s 2624298.62 | norm 0.3852 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13754 | train 2.980566 | val 3.339385 | tok/s 2624630.08 | norm 0.3485 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13755 | train 2.879025 | val 3.339385 | tok/s 2617649.72 | norm 0.3633 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13756 | train 2.917126 | val 3.339385 | tok/s 2624606.18 | norm 0.4173 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13757 | train 2.972324 | val 3.339385 | tok/s 2627094.64 | norm 0.4114 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13758 | train 3.030468 | val 3.339385 | tok/s 2630609.44 | norm 0.3821 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13759 | train 2.974776 | val 3.339385 | tok/s 2627724.47 | norm 0.3970 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13760 | train 2.977169 | val 3.339385 | tok/s 2622332.54 | norm 0.4301 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13761 | train 2.980510 | val 3.339385 | tok/s 2621585.70 | norm 0.3933 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13762 | train 2.934222 | val 3.339385 | tok/s 2627352.96 | norm 0.3937 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13763 | train 2.973649 | val 3.339385 | tok/s 2625974.30 | norm 0.4221 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13764 | train 2.903065 | val 3.339385 | tok/s 2627704.37 | norm 0.3891 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13765 | train 2.921877 | val 3.339385 | tok/s 2625665.23 | norm 0.4227 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13766 | train 3.025932 | val 3.339385 | tok/s 2625352.09 | norm 0.4063 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13767 | train 2.978158 | val 3.339385 | tok/s 2624566.28 | norm 0.3570 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13768 | train 2.957730 | val 3.339385 | tok/s 2624607.49 | norm 0.3950 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13769 | train 2.875405 | val 3.339385 | tok/s 2629526.20 | norm 0.3841 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13770 | train 2.942051 | val 3.339385 | tok/s 2628841.90 | norm 0.3715 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13771 | train 3.057503 | val 3.339385 | tok/s 2625421.50 | norm 0.4105 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13772 | train 2.865714 | val 3.339385 | tok/s 2627863.08 | norm 0.3932 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13773 | train 3.109608 | val 3.339385 | tok/s 2625747.97 | norm 0.4569 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13774 | train 2.839245 | val 3.339385 | tok/s 2626154.99 | norm 0.3979 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13775 | train 2.968671 | val 3.339385 | tok/s 2633602.07 | norm 0.3985 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13776 | train 2.873333 | val 3.339385 | tok/s 2631123.01 | norm 0.3634 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13777 | train 2.973231 | val 3.339385 | tok/s 2576211.88 | norm 0.3694 | dt 203.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13778 | train 2.872597 | val 3.339385 | tok/s 2631939.70 | norm 0.4168 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13779 | train 2.985768 | val 3.339385 | tok/s 2630801.22 | norm 0.3770 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13780 | train 2.991582 | val 3.339385 | tok/s 2626348.17 | norm 0.3824 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13781 | train 2.937260 | val 3.339385 | tok/s 2629162.15 | norm 0.3870 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13782 | train 3.066285 | val 3.339385 | tok/s 2628690.97 | norm 0.4051 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13783 | train 3.015630 | val 3.339385 | tok/s 2631288.70 | norm 0.3698 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13784 | train 3.032804 | val 3.339385 | tok/s 2461466.13 | norm 0.4498 | dt 213.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13785 | train 2.978311 | val 3.339385 | tok/s 2615261.78 | norm 0.3861 | dt 200.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13786 | train 2.990466 | val 3.339385 | tok/s 2618492.80 | norm 0.3393 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13787 | train 3.049597 | val 3.339385 | tok/s 2632258.61 | norm 0.3617 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13788 | train 2.976987 | val 3.339385 | tok/s 2628878.48 | norm 0.3764 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13789 | train 2.940277 | val 3.339385 | tok/s 2633064.49 | norm 0.3897 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13790 | train 3.023839 | val 3.339385 | tok/s 2631540.02 | norm 0.3703 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13791 | train 3.096293 | val 3.339385 | tok/s 2526777.03 | norm 0.4225 | dt 207.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13792 | train 2.908314 | val 3.339385 | tok/s 2561972.44 | norm 0.4116 | dt 204.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13793 | train 3.111681 | val 3.339385 | tok/s 2629461.01 | norm 0.3705 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13794 | train 3.008455 | val 3.339385 | tok/s 2628039.09 | norm 0.3675 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13795 | train 2.959301 | val 3.339385 | tok/s 2626442.06 | norm 0.4026 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13796 | train 2.812035 | val 3.339385 | tok/s 2627447.12 | norm 0.4547 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13797 | train 2.961863 | val 3.339385 | tok/s 2624651.10 | norm 0.3626 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13798 | train 3.075800 | val 3.339385 | tok/s 2626949.61 | norm 0.3758 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13799 | train 3.222243 | val 3.339385 | tok/s 2627497.04 | norm 0.3861 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13800 | train 3.000953 | val 3.339385 | tok/s 2628145.77 | norm 0.3721 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13801 | train 3.011958 | val 3.339385 | tok/s 2631171.81 | norm 0.3677 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13802 | train 2.944965 | val 3.339385 | tok/s 2616055.16 | norm 0.3552 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13803 | train 2.899724 | val 3.339385 | tok/s 2625154.27 | norm 0.3661 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13804 | train 2.950789 | val 3.339385 | tok/s 2631794.56 | norm 0.3928 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13805 | train 2.966171 | val 3.339385 | tok/s 2627616.96 | norm 0.3892 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13806 | train 2.775263 | val 3.339385 | tok/s 2634006.13 | norm 0.3916 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13807 | train 3.073690 | val 3.339385 | tok/s 2626476.39 | norm 0.3915 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13808 | train 3.031027 | val 3.339385 | tok/s 2628568.28 | norm 0.3749 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13809 | train 3.051198 | val 3.339385 | tok/s 2629812.20 | norm 0.3947 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13810 | train 2.969074 | val 3.339385 | tok/s 2623939.71 | norm 0.4074 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13811 | train 3.111418 | val 3.339385 | tok/s 2631847.02 | norm 0.3989 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13812 | train 3.104227 | val 3.339385 | tok/s 2620597.86 | norm 0.4014 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13813 | train 2.924614 | val 3.339385 | tok/s 2626246.19 | norm 0.3525 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13814 | train 2.954057 | val 3.339385 | tok/s 2631577.42 | norm 0.3780 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13815 | train 3.019435 | val 3.339385 | tok/s 2629848.80 | norm 0.4109 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13816 | train 2.964376 | val 3.339385 | tok/s 2630778.58 | norm 0.3838 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13817 | train 2.890909 | val 3.339385 | tok/s 2620703.78 | norm 0.3613 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13818 | train 2.874689 | val 3.339385 | tok/s 2626184.02 | norm 0.3911 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13819 | train 3.046973 | val 3.339385 | tok/s 2628667.03 | norm 0.3628 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13820 | train 2.938295 | val 3.339385 | tok/s 2629830.66 | norm 0.3576 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13821 | train 2.904841 | val 3.339385 | tok/s 2627425.58 | norm 0.3949 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13822 | train 2.977038 | val 3.339385 | tok/s 2630526.79 | norm 0.3626 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13823 | train 2.912243 | val 3.339385 | tok/s 2627009.32 | norm 0.3398 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13824 | train 2.973610 | val 3.339385 | tok/s 2627545.48 | norm 0.3855 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13825 | train 3.111562 | val 3.339385 | tok/s 2628037.29 | norm 0.4345 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13826 | train 2.947812 | val 3.339385 | tok/s 2631325.97 | norm 0.4321 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13827 | train 2.977969 | val 3.339385 | tok/s 2630185.72 | norm 0.3832 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13828 | train 2.874037 | val 3.339385 | tok/s 2632942.91 | norm 0.3704 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13829 | train 2.896601 | val 3.339385 | tok/s 2634252.28 | norm 0.3971 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13830 | train 2.985804 | val 3.339385 | tok/s 2630469.59 | norm 0.3997 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13831 | train 2.920098 | val 3.339385 | tok/s 2629277.59 | norm 0.4102 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13832 | train 3.006407 | val 3.339385 | tok/s 2536869.14 | norm 0.4173 | dt 206.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13833 | train 2.976582 | val 3.339385 | tok/s 2630552.32 | norm 0.3681 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13834 | train 2.977618 | val 3.339385 | tok/s 2634409.41 | norm 0.3846 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13835 | train 3.082547 | val 3.339385 | tok/s 2629779.97 | norm 0.4788 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13836 | train 3.089306 | val 3.339385 | tok/s 2631178.51 | norm 0.4435 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13837 | train 3.080880 | val 3.339385 | tok/s 2625958.57 | norm 0.4185 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13838 | train 2.910080 | val 3.339385 | tok/s 2631828.38 | norm 0.3515 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13839 | train 2.937891 | val 3.339385 | tok/s 2633254.13 | norm 0.3512 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13840 | train 2.994661 | val 3.339385 | tok/s 2627625.07 | norm 0.3533 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13841 | train 3.152252 | val 3.339385 | tok/s 2634103.68 | norm 0.3731 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13842 | train 2.899324 | val 3.339385 | tok/s 2628982.64 | norm 0.3583 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13843 | train 3.011906 | val 3.339385 | tok/s 2631738.33 | norm 0.3634 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13844 | train 3.050251 | val 3.339385 | tok/s 2630933.20 | norm 0.3821 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13845 | train 3.058418 | val 3.339385 | tok/s 2627443.70 | norm 0.3907 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13846 | train 2.986231 | val 3.339385 | tok/s 2628773.36 | norm 0.3744 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13847 | train 3.064052 | val 3.339385 | tok/s 2627141.98 | norm 0.3869 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13848 | train 2.898944 | val 3.339385 | tok/s 2629028.45 | norm 0.3903 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13849 | train 2.930989 | val 3.339385 | tok/s 2628133.42 | norm 0.3988 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13850 | train 2.972577 | val 3.339385 | tok/s 2626130.98 | norm 0.3916 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13851 | train 2.964630 | val 3.339385 | tok/s 2629088.73 | norm 0.4035 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13852 | train 2.949811 | val 3.339385 | tok/s 2624201.41 | norm 0.3798 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13853 | train 2.964887 | val 3.339385 | tok/s 2630872.65 | norm 0.3596 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13854 | train 2.935097 | val 3.339385 | tok/s 2629152.04 | norm 0.3758 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13855 | train 2.957621 | val 3.339385 | tok/s 2626004.64 | norm 0.3671 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13856 | train 3.080373 | val 3.339385 | tok/s 2629609.62 | norm 0.4244 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13857 | train 2.958610 | val 3.339385 | tok/s 2629503.77 | norm 0.3644 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13858 | train 3.111252 | val 3.339385 | tok/s 2631182.89 | norm 0.4029 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13859 | train 2.929134 | val 3.339385 | tok/s 2629986.72 | norm 0.3626 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13860 | train 2.971425 | val 3.339385 | tok/s 2626605.57 | norm 0.3268 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13861 | train 2.908976 | val 3.339385 | tok/s 2629263.08 | norm 0.3844 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13862 | train 3.013167 | val 3.339385 | tok/s 2628478.47 | norm 0.3866 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13863 | train 2.978084 | val 3.339385 | tok/s 2630059.90 | norm 0.3367 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13864 | train 3.086023 | val 3.339385 | tok/s 2625780.05 | norm 0.4095 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13865 | train 2.915751 | val 3.339385 | tok/s 2613863.66 | norm 0.3540 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13866 | train 2.827657 | val 3.339385 | tok/s 2631698.68 | norm 0.3715 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13867 | train 2.968069 | val 3.339385 | tok/s 2627428.66 | norm 0.3816 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13868 | train 3.101135 | val 3.339385 | tok/s 2624271.08 | norm 0.3861 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13869 | train 2.883511 | val 3.339385 | tok/s 2621008.23 | norm 0.3950 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13870 | train 2.893444 | val 3.339385 | tok/s 2628451.05 | norm 0.4306 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13871 | train 2.963648 | val 3.339385 | tok/s 2615386.53 | norm 0.4088 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13872 | train 2.996286 | val 3.339385 | tok/s 2628191.78 | norm 0.3654 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13873 | train 2.977759 | val 3.339385 | tok/s 2632427.95 | norm 0.3459 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13874 | train 2.964833 | val 3.339385 | tok/s 2631440.83 | norm 0.3528 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13875 | train 2.912098 | val 3.339385 | tok/s 2630295.63 | norm 0.3770 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13876 | train 2.845172 | val 3.339385 | tok/s 2631310.02 | norm 0.3532 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13877 | train 3.004421 | val 3.339385 | tok/s 2632746.53 | norm 0.3166 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13878 | train 2.930121 | val 3.339385 | tok/s 2631393.32 | norm 0.3715 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13879 | train 2.998190 | val 3.339385 | tok/s 2625056.22 | norm 0.3697 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13880 | train 2.908113 | val 3.339385 | tok/s 2631989.00 | norm 0.3658 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13881 | train 2.929627 | val 3.339385 | tok/s 2630684.28 | norm 0.4004 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13882 | train 2.901253 | val 3.339385 | tok/s 2631892.77 | norm 0.4082 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13883 | train 3.121250 | val 3.339385 | tok/s 2629300.07 | norm 0.4164 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13884 | train 2.928553 | val 3.339385 | tok/s 2627570.70 | norm 0.3896 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13885 | train 3.055085 | val 3.339385 | tok/s 2628552.13 | norm 0.3723 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13886 | train 3.052428 | val 3.339385 | tok/s 2626774.43 | norm 0.4037 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13887 | train 2.982545 | val 3.339385 | tok/s 2628986.86 | norm 0.4094 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13888 | train 2.896280 | val 3.339385 | tok/s 2626689.12 | norm 0.3981 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13889 | train 2.924991 | val 3.339385 | tok/s 2628136.21 | norm 0.4064 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13890 | train 2.920025 | val 3.339385 | tok/s 2577671.83 | norm 0.4240 | dt 203.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13891 | train 3.019671 | val 3.339385 | tok/s 2634867.71 | norm 0.3792 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13892 | train 2.958785 | val 3.339385 | tok/s 2624419.19 | norm 0.3726 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13893 | train 2.917245 | val 3.339385 | tok/s 2632088.72 | norm 0.3769 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13894 | train 3.024194 | val 3.339385 | tok/s 2632638.85 | norm 0.3959 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13895 | train 3.003633 | val 3.339385 | tok/s 2628874.43 | norm 0.4064 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13896 | train 2.857133 | val 3.339385 | tok/s 2538897.32 | norm 0.3776 | dt 206.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13897 | train 2.955770 | val 3.339385 | tok/s 2455414.39 | norm 0.3592 | dt 213.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13898 | train 3.010261 | val 3.339385 | tok/s 2629456.16 | norm 0.3902 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13899 | train 3.058169 | val 3.339385 | tok/s 2619103.71 | norm 0.3903 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13900 | train 3.001954 | val 3.339385 | tok/s 2630625.72 | norm 0.3804 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13901 | train 2.995756 | val 3.339385 | tok/s 2621224.88 | norm 0.4112 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13902 | train 2.836556 | val 3.339385 | tok/s 2621861.83 | norm 0.4331 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13903 | train 2.906616 | val 3.339385 | tok/s 2631668.96 | norm 0.4404 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13904 | train 3.032508 | val 3.339385 | tok/s 2589229.93 | norm 0.4100 | dt 202.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13905 | train 2.983473 | val 3.339385 | tok/s 2627753.81 | norm 0.4005 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13906 | train 2.983888 | val 3.339385 | tok/s 2629094.73 | norm 0.4107 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13907 | train 2.989723 | val 3.339385 | tok/s 2627567.47 | norm 0.4086 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13908 | train 2.917645 | val 3.339385 | tok/s 2629482.55 | norm 0.4096 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13909 | train 2.988595 | val 3.339385 | tok/s 2624386.63 | norm 0.3938 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13910 | train 2.958736 | val 3.339385 | tok/s 2627527.44 | norm 0.4040 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13911 | train 3.202229 | val 3.339385 | tok/s 2619944.21 | norm 0.4043 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13912 | train 2.967571 | val 3.339385 | tok/s 2629592.55 | norm 0.3650 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13913 | train 2.982432 | val 3.339385 | tok/s 2626765.37 | norm 0.3532 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13914 | train 2.956934 | val 3.339385 | tok/s 2624360.62 | norm 0.4147 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13915 | train 3.049257 | val 3.339385 | tok/s 2628668.50 | norm 0.4015 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13916 | train 2.931810 | val 3.339385 | tok/s 2627391.32 | norm 0.3752 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13917 | train 2.903780 | val 3.339385 | tok/s 2627918.09 | norm 0.3940 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13918 | train 2.900452 | val 3.339385 | tok/s 2629004.82 | norm 0.4088 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13919 | train 2.914941 | val 3.339385 | tok/s 2628051.43 | norm 0.3978 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13920 | train 2.962224 | val 3.339385 | tok/s 2629588.95 | norm 0.3607 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13921 | train 2.935152 | val 3.339385 | tok/s 2628820.78 | norm 0.3647 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13922 | train 2.990302 | val 3.339385 | tok/s 2628641.97 | norm 0.3929 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13923 | train 2.954395 | val 3.339385 | tok/s 2628213.93 | norm 0.3876 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13924 | train 3.010656 | val 3.339385 | tok/s 2628261.98 | norm 0.3551 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13925 | train 2.817804 | val 3.339385 | tok/s 2624124.99 | norm 0.3985 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13926 | train 2.996942 | val 3.339385 | tok/s 2631936.95 | norm 0.3996 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13927 | train 2.977960 | val 3.339385 | tok/s 2631943.72 | norm 0.3970 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13928 | train 3.014898 | val 3.339385 | tok/s 2631215.25 | norm 0.4449 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13929 | train 3.065006 | val 3.339385 | tok/s 2631295.56 | norm 0.4089 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13930 | train 2.950879 | val 3.339385 | tok/s 2631895.86 | norm 0.3461 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13931 | train 2.977441 | val 3.339385 | tok/s 2630657.79 | norm 0.3750 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13932 | train 2.912553 | val 3.339385 | tok/s 2632076.48 | norm 0.3812 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13933 | train 2.880077 | val 3.339385 | tok/s 2630151.48 | norm 0.3610 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13934 | train 2.898612 | val 3.339385 | tok/s 2626624.96 | norm 0.3928 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13935 | train 2.967659 | val 3.339385 | tok/s 2628601.10 | norm 0.3731 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13936 | train 2.884622 | val 3.339385 | tok/s 2622912.57 | norm 0.3875 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13937 | train 2.971113 | val 3.339385 | tok/s 2628568.93 | norm 0.3645 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13938 | train 3.009240 | val 3.339385 | tok/s 2627573.45 | norm 0.3562 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13939 | train 2.899599 | val 3.339385 | tok/s 2623468.05 | norm 0.3675 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13940 | train 2.952346 | val 3.339385 | tok/s 2618959.45 | norm 0.3824 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13941 | train 3.093827 | val 3.339385 | tok/s 2631944.41 | norm 0.3855 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13942 | train 3.094787 | val 3.339385 | tok/s 2630784.72 | norm 0.4587 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13943 | train 3.085893 | val 3.339385 | tok/s 2630863.19 | norm 0.3858 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13944 | train 2.931178 | val 3.339385 | tok/s 2627440.81 | norm 0.3405 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13945 | train 3.108331 | val 3.339385 | tok/s 2630767.93 | norm 0.3231 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13946 | train 2.940367 | val 3.339385 | tok/s 2634367.54 | norm 0.3448 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13947 | train 2.966058 | val 3.339385 | tok/s 2625182.66 | norm 0.3935 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13948 | train 2.916133 | val 3.339385 | tok/s 2629325.29 | norm 0.4159 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13949 | train 3.054326 | val 3.339385 | tok/s 2623210.52 | norm 0.3800 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13950 | train 2.928800 | val 3.339385 | tok/s 2632019.15 | norm 0.3463 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13951 | train 2.908073 | val 3.339385 | tok/s 2628277.56 | norm 0.3658 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13952 | train 2.993187 | val 3.339385 | tok/s 2629851.79 | norm 0.3524 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13953 | train 2.958145 | val 3.339385 | tok/s 2633651.06 | norm 0.3714 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13954 | train 2.865295 | val 3.339385 | tok/s 2626513.78 | norm 0.3830 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13955 | train 2.989820 | val 3.339385 | tok/s 2628868.38 | norm 0.3593 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13956 | train 3.054566 | val 3.339385 | tok/s 2627173.64 | norm 0.3754 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13957 | train 2.890976 | val 3.339385 | tok/s 2631984.08 | norm 0.3947 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13958 | train 2.933201 | val 3.339385 | tok/s 2626273.33 | norm 0.3821 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13959 | train 2.916732 | val 3.339385 | tok/s 2624955.59 | norm 0.3838 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13960 | train 2.984477 | val 3.339385 | tok/s 2630966.77 | norm 0.4124 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13961 | train 2.998313 | val 3.339385 | tok/s 2623650.10 | norm 0.4012 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13962 | train 2.950561 | val 3.339385 | tok/s 2626336.23 | norm 0.3783 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13963 | train 2.953772 | val 3.339385 | tok/s 2632166.34 | norm 0.3815 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13964 | train 2.964069 | val 3.339385 | tok/s 2629007.20 | norm 0.3511 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13965 | train 3.013544 | val 3.339385 | tok/s 2628038.11 | norm 0.3525 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13966 | train 3.076545 | val 3.339385 | tok/s 2627359.41 | norm 0.3735 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13967 | train 3.076471 | val 3.339385 | tok/s 2629156.32 | norm 0.3907 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13968 | train 2.956063 | val 3.339385 | tok/s 2631277.54 | norm 0.3613 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13969 | train 2.977580 | val 3.339385 | tok/s 2612149.56 | norm 0.3839 | dt 200.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13970 | train 2.837776 | val 3.339385 | tok/s 2618706.57 | norm 0.3890 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13971 | train 2.904261 | val 3.339385 | tok/s 2632803.72 | norm 0.4042 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13972 | train 2.856118 | val 3.339385 | tok/s 2627904.86 | norm 0.4309 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13973 | train 2.914112 | val 3.339385 | tok/s 2619982.36 | norm 0.4393 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13974 | train 2.977112 | val 3.339385 | tok/s 2627376.86 | norm 0.4350 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13975 | train 3.038296 | val 3.339385 | tok/s 2629957.33 | norm 0.3604 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13976 | train 3.077728 | val 3.339385 | tok/s 2632211.84 | norm 0.4138 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13977 | train 3.065745 | val 3.339385 | tok/s 2632620.87 | norm 0.4198 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13978 | train 2.927091 | val 3.339385 | tok/s 2625166.78 | norm 0.3746 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13979 | train 3.076944 | val 3.339385 | tok/s 2624855.28 | norm 0.4208 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13980 | train 3.293646 | val 3.339385 | tok/s 2634542.83 | norm 0.5245 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13981 | train 3.012873 | val 3.339385 | tok/s 2630983.71 | norm 0.4034 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13982 | train 2.981110 | val 3.339385 | tok/s 2628798.31 | norm 0.3409 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13983 | train 2.819696 | val 3.339385 | tok/s 2632694.49 | norm 0.3467 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13984 | train 3.023698 | val 3.339385 | tok/s 2617055.14 | norm 0.3380 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13985 | train 3.016252 | val 3.339385 | tok/s 2633325.87 | norm 0.3753 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13986 | train 2.880605 | val 3.339385 | tok/s 2627753.52 | norm 0.3511 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13987 | train 3.007513 | val 3.339385 | tok/s 2631531.30 | norm 0.3426 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13988 | train 2.939913 | val 3.339385 | tok/s 2629164.97 | norm 0.3926 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13989 | train 3.097793 | val 3.339385 | tok/s 2626533.87 | norm 0.4986 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13990 | train 2.960082 | val 3.339385 | tok/s 2631032.73 | norm 0.4407 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13991 | train 3.052862 | val 3.339385 | tok/s 2629166.44 | norm 0.3826 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13992 | train 3.307514 | val 3.339385 | tok/s 2632115.45 | norm 0.6012 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13993 | train 3.085882 | val 3.339385 | tok/s 2629968.95 | norm 0.4377 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13994 | train 2.994735 | val 3.339385 | tok/s 2624607.98 | norm 0.3550 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13995 | train 2.958866 | val 3.339385 | tok/s 2628111.83 | norm 0.3615 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13996 | train 2.815589 | val 3.339385 | tok/s 2626253.81 | norm 0.3669 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13997 | train 2.927679 | val 3.339385 | tok/s 2616394.73 | norm 0.4179 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13998 | train 3.015790 | val 3.339385 | tok/s 2628513.92 | norm 0.4029 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 13999 | train 3.166766 | val 3.339385 | tok/s 2623276.44 | norm 0.3748 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 14000 | val_loss 3.346537 step 14000 | train 2.987198 | val 3.346537 | tok/s 2616478.06 | norm 0.3556 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14001 | train 3.049662 | val 3.346537 | tok/s 2630687.05 | norm 0.3512 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14002 | train 3.025491 | val 3.346537 | tok/s 2625102.66 | norm 0.3615 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14003 | train 2.956513 | val 3.346537 | tok/s 2628536.35 | norm 0.3526 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14004 | train 2.923852 | val 3.346537 | tok/s 2623794.91 | norm 0.3727 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14005 | train 2.903471 | val 3.346537 | tok/s 2627459.02 | norm 0.3676 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14006 | train 2.952931 | val 3.346537 | tok/s 2630039.02 | norm 0.3456 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14007 | train 2.991111 | val 3.346537 | tok/s 2625152.14 | norm 0.3483 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14008 | train 2.901430 | val 3.346537 | tok/s 2628875.67 | norm 0.4003 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14009 | train 2.949748 | val 3.346537 | tok/s 2625694.89 | norm 0.3966 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14010 | train 3.003984 | val 3.346537 | tok/s 2628506.87 | norm 0.3791 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14011 | train 2.936898 | val 3.346537 | tok/s 2625137.44 | norm 0.3737 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14012 | train 2.929566 | val 3.346537 | tok/s 2630852.12 | norm 0.3979 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14013 | train 3.110245 | val 3.346537 | tok/s 2633536.33 | norm 0.4322 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14014 | train 2.920238 | val 3.346537 | tok/s 2630480.53 | norm 0.3552 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14015 | train 3.057924 | val 3.346537 | tok/s 2632774.33 | norm 0.3261 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14016 | train 3.064109 | val 3.346537 | tok/s 2632082.46 | norm 0.3560 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14017 | train 2.941600 | val 3.346537 | tok/s 2629949.27 | norm 0.3844 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14018 | train 3.004799 | val 3.346537 | tok/s 2621974.17 | norm 0.3874 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14019 | train 3.098707 | val 3.346537 | tok/s 2629357.75 | norm 0.4208 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14020 | train 3.075349 | val 3.346537 | tok/s 2630084.17 | norm 0.4402 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14021 | train 2.951737 | val 3.346537 | tok/s 2623928.73 | norm 0.3730 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14022 | train 3.017123 | val 3.346537 | tok/s 2624352.19 | norm 0.3597 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14023 | train 3.091926 | val 3.346537 | tok/s 2627621.76 | norm 0.4164 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14024 | train 2.939951 | val 3.346537 | tok/s 2624805.89 | norm 0.3778 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14025 | train 2.966198 | val 3.346537 | tok/s 2624487.80 | norm 0.3884 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14026 | train 2.870813 | val 3.346537 | tok/s 2632395.89 | norm 0.3946 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14027 | train 2.838683 | val 3.346537 | tok/s 2622359.19 | norm 0.4262 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14028 | train 2.913051 | val 3.346537 | tok/s 2629786.98 | norm 0.4019 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14029 | train 2.986180 | val 3.346537 | tok/s 2629278.83 | norm 0.3798 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14030 | train 2.984752 | val 3.346537 | tok/s 2630118.27 | norm 0.3750 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14031 | train 3.044787 | val 3.346537 | tok/s 2630980.54 | norm 0.3961 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14032 | train 2.932805 | val 3.346537 | tok/s 2629428.11 | norm 0.4020 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14033 | train 2.935575 | val 3.346537 | tok/s 2626984.72 | norm 0.3800 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14034 | train 2.924329 | val 3.346537 | tok/s 2627387.93 | norm 0.3837 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14035 | train 2.998200 | val 3.346537 | tok/s 2628785.93 | norm 0.3786 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14036 | train 3.030265 | val 3.346537 | tok/s 2626043.77 | norm 0.3957 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14037 | train 2.958644 | val 3.346537 | tok/s 2624434.02 | norm 0.3616 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14038 | train 2.931444 | val 3.346537 | tok/s 2629093.76 | norm 0.3660 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14039 | train 3.002707 | val 3.346537 | tok/s 2626839.67 | norm 0.3671 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14040 | train 2.828639 | val 3.346537 | tok/s 2629142.63 | norm 0.3935 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14041 | train 3.010406 | val 3.346537 | tok/s 2622991.22 | norm 0.3698 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14042 | train 2.978431 | val 3.346537 | tok/s 2624244.41 | norm 0.3814 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14043 | train 2.993197 | val 3.346537 | tok/s 2626975.94 | norm 0.3821 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14044 | train 2.960228 | val 3.346537 | tok/s 2625503.81 | norm 0.3901 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14045 | train 2.923231 | val 3.346537 | tok/s 2627675.12 | norm 0.4140 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14046 | train 3.047331 | val 3.346537 | tok/s 2625980.73 | norm 0.4110 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14047 | train 3.014473 | val 3.346537 | tok/s 2629020.35 | norm 0.3925 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14048 | train 2.967759 | val 3.346537 | tok/s 2630584.85 | norm 0.3864 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14049 | train 3.004450 | val 3.346537 | tok/s 2626173.86 | norm 0.3819 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14050 | train 3.001057 | val 3.346537 | tok/s 2631586.52 | norm 0.3867 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14051 | train 2.946773 | val 3.346537 | tok/s 2628521.54 | norm 0.3797 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14052 | train 2.908844 | val 3.346537 | tok/s 2625714.78 | norm 0.3887 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14053 | train 3.013220 | val 3.346537 | tok/s 2631967.22 | norm 0.3973 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14054 | train 3.014314 | val 3.346537 | tok/s 2624943.42 | norm 0.4143 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14055 | train 2.884652 | val 3.346537 | tok/s 2628391.83 | norm 0.3980 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14056 | train 3.009798 | val 3.346537 | tok/s 2623938.53 | norm 0.3879 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14057 | train 3.025817 | val 3.346537 | tok/s 2615613.94 | norm 0.3886 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14058 | train 2.954275 | val 3.346537 | tok/s 2625494.01 | norm 0.4123 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14059 | train 2.898380 | val 3.346537 | tok/s 2629931.88 | norm 0.4139 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14060 | train 3.022872 | val 3.346537 | tok/s 2634035.07 | norm 0.4015 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14061 | train 3.031219 | val 3.346537 | tok/s 2622262.22 | norm 0.3814 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14062 | train 2.864874 | val 3.346537 | tok/s 2627833.24 | norm 0.3607 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14063 | train 3.099197 | val 3.346537 | tok/s 2632109.63 | norm 0.3994 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14064 | train 3.165565 | val 3.346537 | tok/s 2633916.19 | norm 0.4015 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14065 | train 2.961404 | val 3.346537 | tok/s 2634279.90 | norm 0.3794 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14066 | train 2.953517 | val 3.346537 | tok/s 2635226.04 | norm 0.3748 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14067 | train 2.967862 | val 3.346537 | tok/s 2633566.52 | norm 0.3727 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14068 | train 3.164279 | val 3.346537 | tok/s 2630634.78 | norm 0.3948 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14069 | train 3.031955 | val 3.346537 | tok/s 2625946.32 | norm 0.3837 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14070 | train 2.867950 | val 3.346537 | tok/s 2634533.90 | norm 0.3781 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14071 | train 3.037492 | val 3.346537 | tok/s 2625934.94 | norm 0.3930 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14072 | train 2.994242 | val 3.346537 | tok/s 2625318.26 | norm 0.3676 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14073 | train 2.971482 | val 3.346537 | tok/s 2627397.95 | norm 0.3825 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14074 | train 2.952579 | val 3.346537 | tok/s 2624502.48 | norm 0.3552 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14075 | train 3.029852 | val 3.346537 | tok/s 2627151.48 | norm 0.3663 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14076 | train 2.989078 | val 3.346537 | tok/s 2625912.89 | norm 0.3710 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14077 | train 2.832720 | val 3.346537 | tok/s 2631242.40 | norm 0.3579 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14078 | train 3.057185 | val 3.346537 | tok/s 2629261.46 | norm 0.3762 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14079 | train 2.949039 | val 3.346537 | tok/s 2627662.29 | norm 0.3658 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14080 | train 3.000479 | val 3.346537 | tok/s 2627219.58 | norm 0.3910 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14081 | train 3.030134 | val 3.346537 | tok/s 2625672.26 | norm 0.4184 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14082 | train 3.020192 | val 3.346537 | tok/s 2629125.37 | norm 0.3779 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14083 | train 3.009560 | val 3.346537 | tok/s 2630276.15 | norm 0.3559 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14084 | train 2.968775 | val 3.346537 | tok/s 2538900.89 | norm 0.3436 | dt 206.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14085 | train 2.994299 | val 3.346537 | tok/s 2627560.78 | norm 0.3575 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14086 | train 2.931362 | val 3.346537 | tok/s 2626670.58 | norm 0.3583 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14087 | train 2.943609 | val 3.346537 | tok/s 2626619.85 | norm 0.3828 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14088 | train 2.987537 | val 3.346537 | tok/s 2628474.35 | norm 0.3921 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14089 | train 2.897682 | val 3.346537 | tok/s 2625843.91 | norm 0.3730 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14090 | train 3.075782 | val 3.346537 | tok/s 2628851.07 | norm 0.3581 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14091 | train 3.011823 | val 3.346537 | tok/s 2630571.17 | norm 0.3688 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14092 | train 2.940078 | val 3.346537 | tok/s 2630230.03 | norm 0.3831 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14093 | train 2.965725 | val 3.346537 | tok/s 2627751.87 | norm 0.3713 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14094 | train 3.007542 | val 3.346537 | tok/s 2626464.04 | norm 0.4151 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14095 | train 2.975096 | val 3.346537 | tok/s 2627433.16 | norm 0.3770 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14096 | train 2.987743 | val 3.346537 | tok/s 2624710.23 | norm 0.3783 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14097 | train 2.917765 | val 3.346537 | tok/s 2629264.44 | norm 0.3729 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14098 | train 3.090152 | val 3.346537 | tok/s 2628441.97 | norm 0.3803 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14099 | train 2.917179 | val 3.346537 | tok/s 2628498.52 | norm 0.3775 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14100 | train 3.017592 | val 3.346537 | tok/s 2592315.50 | norm 0.3377 | dt 202.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14101 | train 3.288138 | val 3.346537 | tok/s 2627843.44 | norm 0.5442 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14102 | train 3.127608 | val 3.346537 | tok/s 2631814.90 | norm 0.4534 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14103 | train 2.982639 | val 3.346537 | tok/s 2623877.61 | norm 0.3809 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14104 | train 2.940582 | val 3.346537 | tok/s 2628390.09 | norm 0.4027 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14105 | train 2.847465 | val 3.346537 | tok/s 2628152.68 | norm 0.4242 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14106 | train 2.975000 | val 3.346537 | tok/s 2613506.61 | norm 0.4145 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14107 | train 3.096470 | val 3.346537 | tok/s 2631659.99 | norm 0.4061 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14108 | train 2.974883 | val 3.346537 | tok/s 2622519.03 | norm 0.4049 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14109 | train 2.987306 | val 3.346537 | tok/s 2624003.11 | norm 0.3796 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14110 | train 2.980180 | val 3.346537 | tok/s 2631313.16 | norm 0.3761 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14111 | train 2.964481 | val 3.346537 | tok/s 2630028.82 | norm 0.3876 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14112 | train 2.960737 | val 3.346537 | tok/s 2629655.10 | norm 0.3855 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14113 | train 3.024468 | val 3.346537 | tok/s 2625702.24 | norm 0.3752 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14114 | train 3.292241 | val 3.346537 | tok/s 2624950.54 | norm 0.4650 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14115 | train 2.985760 | val 3.346537 | tok/s 2628568.40 | norm 0.3426 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14116 | train 2.890617 | val 3.346537 | tok/s 2630212.39 | norm 0.3376 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14117 | train 2.998205 | val 3.346537 | tok/s 2630096.94 | norm 0.3719 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14118 | train 3.106928 | val 3.346537 | tok/s 2627020.47 | norm 0.3758 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14119 | train 2.985849 | val 3.346537 | tok/s 2625703.22 | norm 0.3476 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14120 | train 2.925480 | val 3.346537 | tok/s 2628760.70 | norm 0.3643 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14121 | train 2.893145 | val 3.346537 | tok/s 2628544.88 | norm 0.4159 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14122 | train 2.995665 | val 3.346537 | tok/s 2630269.82 | norm 0.3895 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14123 | train 3.022995 | val 3.346537 | tok/s 2624452.54 | norm 0.3784 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14124 | train 2.863615 | val 3.346537 | tok/s 2628356.16 | norm 0.3595 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14125 | train 2.987695 | val 3.346537 | tok/s 2628997.82 | norm 0.3775 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14126 | train 2.920275 | val 3.346537 | tok/s 2634195.23 | norm 0.3952 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14127 | train 2.947310 | val 3.346537 | tok/s 2629778.73 | norm 0.3852 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14128 | train 2.947852 | val 3.346537 | tok/s 2627227.44 | norm 0.4111 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14129 | train 2.930987 | val 3.346537 | tok/s 2631143.24 | norm 0.3566 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14130 | train 2.908570 | val 3.346537 | tok/s 2631498.76 | norm 0.3389 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14131 | train 2.993900 | val 3.346537 | tok/s 2625925.16 | norm 0.3748 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14132 | train 2.933204 | val 3.346537 | tok/s 2626668.86 | norm 0.3608 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14133 | train 2.950400 | val 3.346537 | tok/s 2622529.03 | norm 0.3456 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14134 | train 2.903818 | val 3.346537 | tok/s 2627196.79 | norm 0.3619 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14135 | train 3.047550 | val 3.346537 | tok/s 2631394.47 | norm 0.3587 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14136 | train 2.881336 | val 3.346537 | tok/s 2626131.48 | norm 0.4068 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14137 | train 2.976278 | val 3.346537 | tok/s 2627212.39 | norm 0.4168 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14138 | train 3.047480 | val 3.346537 | tok/s 2628682.66 | norm 0.3816 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14139 | train 2.959131 | val 3.346537 | tok/s 2627387.74 | norm 0.3768 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14140 | train 2.929413 | val 3.346537 | tok/s 2633607.70 | norm 0.3768 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14141 | train 2.971410 | val 3.346537 | tok/s 2629767.34 | norm 0.4002 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14142 | train 2.895137 | val 3.346537 | tok/s 2627341.57 | norm 0.4368 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14143 | train 3.019287 | val 3.346537 | tok/s 2626203.01 | norm 0.4472 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14144 | train 3.076683 | val 3.346537 | tok/s 2627446.04 | norm 0.4319 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14145 | train 2.955331 | val 3.346537 | tok/s 2624541.20 | norm 0.4038 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14146 | train 2.908519 | val 3.346537 | tok/s 2622008.88 | norm 0.3950 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14147 | train 2.970454 | val 3.346537 | tok/s 2632093.75 | norm 0.3707 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14148 | train 3.038006 | val 3.346537 | tok/s 2633437.44 | norm 0.3725 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14149 | train 2.968662 | val 3.346537 | tok/s 2634163.30 | norm 0.3705 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14150 | train 2.925613 | val 3.346537 | tok/s 2633775.05 | norm 0.3409 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14151 | train 2.966926 | val 3.346537 | tok/s 2627334.54 | norm 0.3700 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14152 | train 3.034644 | val 3.346537 | tok/s 2628657.70 | norm 0.3631 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14153 | train 2.863434 | val 3.346537 | tok/s 2629345.64 | norm 0.3612 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14154 | train 2.990197 | val 3.346537 | tok/s 2632007.30 | norm 0.4133 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14155 | train 3.141510 | val 3.346537 | tok/s 2630139.11 | norm 0.4417 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14156 | train 2.906775 | val 3.346537 | tok/s 2624990.23 | norm 0.3587 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14157 | train 3.168452 | val 3.346537 | tok/s 2630009.93 | norm 0.4803 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14158 | train 2.826655 | val 3.346537 | tok/s 2633807.83 | norm 0.4548 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14159 | train 2.874901 | val 3.346537 | tok/s 2630454.13 | norm 0.3639 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14160 | train 2.978045 | val 3.346537 | tok/s 2626702.52 | norm 0.3730 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14161 | train 2.991269 | val 3.346537 | tok/s 2631638.56 | norm 0.3958 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14162 | train 2.854830 | val 3.346537 | tok/s 2631073.56 | norm 0.3992 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14163 | train 3.048761 | val 3.346537 | tok/s 2553079.99 | norm 0.4042 | dt 205.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14164 | train 2.917178 | val 3.346537 | tok/s 2586597.60 | norm 0.3655 | dt 202.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14165 | train 2.882032 | val 3.346537 | tok/s 2633428.51 | norm 0.3808 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14166 | train 3.111202 | val 3.346537 | tok/s 2633597.54 | norm 0.4568 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14167 | train 2.991473 | val 3.346537 | tok/s 2635418.64 | norm 0.3828 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14168 | train 3.003822 | val 3.346537 | tok/s 2631826.07 | norm 0.3857 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14169 | train 2.972352 | val 3.346537 | tok/s 2628034.70 | norm 0.3916 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14170 | train 3.016412 | val 3.346537 | tok/s 2622363.12 | norm 0.3564 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14171 | train 2.880437 | val 3.346537 | tok/s 2626956.99 | norm 0.3575 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14172 | train 3.045291 | val 3.346537 | tok/s 2624467.01 | norm 0.4121 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14173 | train 2.940522 | val 3.346537 | tok/s 2621715.94 | norm 0.3890 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14174 | train 2.967556 | val 3.346537 | tok/s 2622412.43 | norm 0.3576 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14175 | train 2.978591 | val 3.346537 | tok/s 2626679.87 | norm 0.4012 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14176 | train 2.961936 | val 3.346537 | tok/s 2627482.62 | norm 0.4249 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14177 | train 3.054902 | val 3.346537 | tok/s 2631114.41 | norm 0.3897 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14178 | train 3.091709 | val 3.346537 | tok/s 2630035.09 | norm 0.4010 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14179 | train 2.901134 | val 3.346537 | tok/s 2633778.12 | norm 0.4003 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14180 | train 2.878572 | val 3.346537 | tok/s 2628628.72 | norm 0.3710 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14181 | train 2.930871 | val 3.346537 | tok/s 2627564.92 | norm 0.3636 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14182 | train 3.058867 | val 3.346537 | tok/s 2630163.00 | norm 0.3603 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14183 | train 2.936069 | val 3.346537 | tok/s 2628361.91 | norm 0.3374 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14184 | train 3.031114 | val 3.346537 | tok/s 2631503.55 | norm 0.3441 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14185 | train 3.005889 | val 3.346537 | tok/s 2625564.72 | norm 0.4093 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14186 | train 2.976460 | val 3.346537 | tok/s 2628085.80 | norm 0.4296 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14187 | train 2.965210 | val 3.346537 | tok/s 2627632.19 | norm 0.4417 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14188 | train 3.054287 | val 3.346537 | tok/s 2629152.37 | norm 0.4234 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14189 | train 2.907565 | val 3.346537 | tok/s 2627723.07 | norm 0.4120 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14190 | train 3.019917 | val 3.346537 | tok/s 2627814.37 | norm 0.4127 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14191 | train 3.093524 | val 3.346537 | tok/s 2628240.93 | norm 0.3796 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14192 | train 2.890050 | val 3.346537 | tok/s 2633934.29 | norm 0.3968 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14193 | train 2.831074 | val 3.346537 | tok/s 2625438.77 | norm 0.4052 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14194 | train 3.149618 | val 3.346537 | tok/s 2630825.46 | norm 0.4458 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14195 | train 2.966354 | val 3.346537 | tok/s 2628369.22 | norm 0.3730 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14196 | train 2.920173 | val 3.346537 | tok/s 2630365.87 | norm 0.3592 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14197 | train 2.874040 | val 3.346537 | tok/s 2627883.95 | norm 0.3796 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14198 | train 2.972745 | val 3.346537 | tok/s 2623702.41 | norm 0.3631 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14199 | train 2.978939 | val 3.346537 | tok/s 2623372.63 | norm 0.3462 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14200 | train 3.012427 | val 3.346537 | tok/s 2628184.87 | norm 0.3702 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14201 | train 3.048280 | val 3.346537 | tok/s 2628828.44 | norm 0.3950 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14202 | train 2.898411 | val 3.346537 | tok/s 2627678.16 | norm 0.3471 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14203 | train 2.953318 | val 3.346537 | tok/s 2626963.74 | norm 0.3073 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14204 | train 2.975688 | val 3.346537 | tok/s 2626977.18 | norm 0.3374 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14205 | train 2.975028 | val 3.346537 | tok/s 2628031.18 | norm 0.4033 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14206 | train 3.043526 | val 3.346537 | tok/s 2627476.52 | norm 0.4048 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14207 | train 2.987986 | val 3.346537 | tok/s 2626387.84 | norm 0.6269 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14208 | train 2.959521 | val 3.346537 | tok/s 2624532.13 | norm 0.3898 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14209 | train 3.078433 | val 3.346537 | tok/s 2626525.99 | norm 0.3815 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14210 | train 3.040096 | val 3.346537 | tok/s 2624388.88 | norm 0.3554 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14211 | train 2.989093 | val 3.346537 | tok/s 2626896.39 | norm 0.3680 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14212 | train 2.939188 | val 3.346537 | tok/s 2627307.34 | norm 0.3700 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14213 | train 2.886526 | val 3.346537 | tok/s 2623939.85 | norm 0.3530 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14214 | train 2.972686 | val 3.346537 | tok/s 2628672.52 | norm 0.3879 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14215 | train 3.003892 | val 3.346537 | tok/s 2624993.69 | norm 0.3846 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14216 | train 2.918294 | val 3.346537 | tok/s 2629017.21 | norm 0.3770 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14217 | train 2.797859 | val 3.346537 | tok/s 2628672.00 | norm 0.3958 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14218 | train 3.004769 | val 3.346537 | tok/s 2627757.81 | norm 0.4212 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14219 | train 3.020453 | val 3.346537 | tok/s 2629314.75 | norm 0.3916 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14220 | train 3.059219 | val 3.346537 | tok/s 2628901.39 | norm 0.3839 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14221 | train 3.013119 | val 3.346537 | tok/s 2628355.18 | norm 0.3768 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14222 | train 2.942753 | val 3.346537 | tok/s 2626614.26 | norm 0.3963 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14223 | train 2.925576 | val 3.346537 | tok/s 2626953.09 | norm 0.3941 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14224 | train 2.774160 | val 3.346537 | tok/s 2630093.81 | norm 0.3634 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14225 | train 2.880619 | val 3.346537 | tok/s 2628438.68 | norm 0.3985 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14226 | train 2.966037 | val 3.346537 | tok/s 2630576.18 | norm 0.3966 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14227 | train 3.016407 | val 3.346537 | tok/s 2627866.88 | norm 0.3692 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14228 | train 2.992890 | val 3.346537 | tok/s 2629274.22 | norm 0.3422 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14229 | train 3.009213 | val 3.346537 | tok/s 2631169.46 | norm 0.3531 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14230 | train 3.001318 | val 3.346537 | tok/s 2622519.30 | norm 0.3887 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14231 | train 3.034622 | val 3.346537 | tok/s 2625098.90 | norm 0.4355 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14232 | train 2.961438 | val 3.346537 | tok/s 2625018.00 | norm 0.4341 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14233 | train 2.940346 | val 3.346537 | tok/s 2627609.36 | norm 0.4105 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14234 | train 2.983780 | val 3.346537 | tok/s 2630348.36 | norm 0.3841 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14235 | train 2.946649 | val 3.346537 | tok/s 2628586.94 | norm 0.4264 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14236 | train 2.924765 | val 3.346537 | tok/s 2629765.68 | norm 0.3824 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14237 | train 2.956922 | val 3.346537 | tok/s 2629075.18 | norm 0.3616 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14238 | train 3.004411 | val 3.346537 | tok/s 2629281.04 | norm 0.3501 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14239 | train 3.029010 | val 3.346537 | tok/s 2628610.96 | norm 0.3940 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14240 | train 2.938396 | val 3.346537 | tok/s 2625719.74 | norm 0.4190 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14241 | train 3.052120 | val 3.346537 | tok/s 2627855.84 | norm 0.4097 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14242 | train 2.940752 | val 3.346537 | tok/s 2631121.38 | norm 0.4124 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14243 | train 2.867884 | val 3.346537 | tok/s 2631029.35 | norm 0.4042 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14244 | train 3.060809 | val 3.346537 | tok/s 2631587.61 | norm 0.3865 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14245 | train 3.053376 | val 3.346537 | tok/s 2628683.42 | norm 0.3499 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14246 | train 2.886321 | val 3.346537 | tok/s 2628217.42 | norm 0.3741 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14247 | train 2.959921 | val 3.346537 | tok/s 2629450.45 | norm 0.3934 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14248 | train 2.994653 | val 3.346537 | tok/s 2631148.24 | norm 0.3737 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14249 | train 3.008388 | val 3.346537 | tok/s 2633254.28 | norm 0.4024 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 14250 | val_loss 3.334813 step 14250 | train 2.939793 | val 3.334813 | tok/s 2610868.90 | norm 0.4061 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14251 | train 2.969066 | val 3.334813 | tok/s 2631519.30 | norm 0.3983 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14252 | train 2.976538 | val 3.334813 | tok/s 2630398.72 | norm 0.4067 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14253 | train 2.987876 | val 3.334813 | tok/s 2623353.25 | norm 0.3765 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14254 | train 2.856673 | val 3.334813 | tok/s 2624565.27 | norm 0.3836 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14255 | train 2.983688 | val 3.334813 | tok/s 2630535.62 | norm 0.3458 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14256 | train 2.908242 | val 3.334813 | tok/s 2625655.49 | norm 0.3748 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14257 | train 2.798513 | val 3.334813 | tok/s 2628447.36 | norm 0.4072 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14258 | train 2.928883 | val 3.334813 | tok/s 2629039.87 | norm 0.3684 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14259 | train 2.934690 | val 3.334813 | tok/s 2631360.39 | norm 0.3490 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14260 | train 2.897831 | val 3.334813 | tok/s 2632354.43 | norm 0.3600 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14261 | train 2.988022 | val 3.334813 | tok/s 2624788.22 | norm 0.3835 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14262 | train 3.043592 | val 3.334813 | tok/s 2629665.75 | norm 0.4010 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14263 | train 2.999429 | val 3.334813 | tok/s 2623115.17 | norm 0.3871 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14264 | train 3.071276 | val 3.334813 | tok/s 2625659.47 | norm 0.3587 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14265 | train 3.050653 | val 3.334813 | tok/s 2630632.27 | norm 0.3751 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14266 | train 3.081250 | val 3.334813 | tok/s 2626181.74 | norm 0.4194 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14267 | train 2.952032 | val 3.334813 | tok/s 2629038.88 | norm 0.4261 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14268 | train 2.989661 | val 3.334813 | tok/s 2624774.59 | norm 0.4075 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14269 | train 2.979952 | val 3.334813 | tok/s 2625261.24 | norm 0.4399 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14270 | train 2.803817 | val 3.334813 | tok/s 2631973.29 | norm 0.4453 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14271 | train 2.996258 | val 3.334813 | tok/s 2567517.99 | norm 0.4140 | dt 204.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14272 | train 2.935803 | val 3.334813 | tok/s 2626707.98 | norm 0.4062 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14273 | train 2.953139 | val 3.334813 | tok/s 2565675.34 | norm 0.3653 | dt 204.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14274 | train 2.829533 | val 3.334813 | tok/s 2589789.22 | norm 0.3728 | dt 202.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14275 | train 2.853366 | val 3.334813 | tok/s 2631067.85 | norm 0.3664 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14276 | train 2.921448 | val 3.334813 | tok/s 2626087.40 | norm 0.3910 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14277 | train 3.003722 | val 3.334813 | tok/s 2621518.61 | norm 0.4111 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14278 | train 2.944058 | val 3.334813 | tok/s 2627382.32 | norm 0.4317 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14279 | train 3.024583 | val 3.334813 | tok/s 2627387.10 | norm 0.4074 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14280 | train 2.963225 | val 3.334813 | tok/s 2481478.86 | norm 0.3980 | dt 211.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14281 | train 2.955339 | val 3.334813 | tok/s 2628745.94 | norm 0.3877 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14282 | train 2.911530 | val 3.334813 | tok/s 2629842.03 | norm 0.3603 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14283 | train 2.953504 | val 3.334813 | tok/s 2631880.82 | norm 0.3468 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14284 | train 2.920530 | val 3.334813 | tok/s 2628684.34 | norm 0.3615 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14285 | train 2.929302 | val 3.334813 | tok/s 2625499.02 | norm 0.3957 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14286 | train 2.962738 | val 3.334813 | tok/s 2627598.00 | norm 0.4291 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14287 | train 2.971917 | val 3.334813 | tok/s 2626447.68 | norm 0.3948 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14288 | train 3.015259 | val 3.334813 | tok/s 2632392.00 | norm 0.3602 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14289 | train 3.027657 | val 3.334813 | tok/s 2628963.13 | norm 0.3610 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14290 | train 3.000684 | val 3.334813 | tok/s 2630126.06 | norm 0.4177 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14291 | train 2.985713 | val 3.334813 | tok/s 2628813.03 | norm 0.4476 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14292 | train 2.975390 | val 3.334813 | tok/s 2633859.21 | norm 0.3826 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14293 | train 3.006968 | val 3.334813 | tok/s 2628361.28 | norm 0.3707 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14294 | train 3.022790 | val 3.334813 | tok/s 2632517.57 | norm 0.3972 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14295 | train 2.990285 | val 3.334813 | tok/s 2632394.64 | norm 0.4767 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14296 | train 2.887526 | val 3.334813 | tok/s 2632629.40 | norm 0.3940 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14297 | train 3.127360 | val 3.334813 | tok/s 2632226.31 | norm 0.4635 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14298 | train 3.123761 | val 3.334813 | tok/s 2625049.10 | norm 0.4607 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14299 | train 3.087867 | val 3.334813 | tok/s 2627323.40 | norm 0.4064 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14300 | train 3.038125 | val 3.334813 | tok/s 2629889.02 | norm 0.3530 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14301 | train 2.869975 | val 3.334813 | tok/s 2631995.78 | norm 0.3764 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14302 | train 2.969690 | val 3.334813 | tok/s 2630066.47 | norm 0.3902 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14303 | train 2.903252 | val 3.334813 | tok/s 2631288.06 | norm 0.3842 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14304 | train 3.035551 | val 3.334813 | tok/s 2628734.05 | norm 0.3673 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14305 | train 2.916470 | val 3.334813 | tok/s 2626537.79 | norm 0.3524 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14306 | train 2.948593 | val 3.334813 | tok/s 2625944.01 | norm 0.3735 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14307 | train 2.952673 | val 3.334813 | tok/s 2630096.83 | norm 0.3421 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14308 | train 2.893936 | val 3.334813 | tok/s 2629232.55 | norm 0.3574 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14309 | train 3.050291 | val 3.334813 | tok/s 2627693.37 | norm 0.3764 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14310 | train 2.870103 | val 3.334813 | tok/s 2627903.38 | norm 0.3478 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14311 | train 3.003895 | val 3.334813 | tok/s 2627270.87 | norm 0.3848 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14312 | train 3.026179 | val 3.334813 | tok/s 2634176.39 | norm 0.4376 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14313 | train 3.015070 | val 3.334813 | tok/s 2627554.96 | norm 0.3846 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14314 | train 2.858957 | val 3.334813 | tok/s 2629808.39 | norm 0.3726 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14315 | train 2.811524 | val 3.334813 | tok/s 2621121.04 | norm 0.3707 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14316 | train 3.011705 | val 3.334813 | tok/s 2629598.54 | norm 0.3745 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14317 | train 2.999631 | val 3.334813 | tok/s 2630071.70 | norm 0.4297 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14318 | train 3.052200 | val 3.334813 | tok/s 2627168.20 | norm 0.3914 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14319 | train 2.921370 | val 3.334813 | tok/s 2629854.30 | norm 0.3871 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14320 | train 3.062711 | val 3.334813 | tok/s 2628869.12 | norm 0.3754 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14321 | train 2.827220 | val 3.334813 | tok/s 2627342.53 | norm 0.3549 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14322 | train 3.149447 | val 3.334813 | tok/s 2630183.39 | norm 0.3967 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14323 | train 3.048761 | val 3.334813 | tok/s 2627210.73 | norm 0.4398 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14324 | train 2.897472 | val 3.334813 | tok/s 2628325.90 | norm 0.3596 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14325 | train 3.143440 | val 3.334813 | tok/s 2629546.66 | norm 0.3542 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14326 | train 2.969574 | val 3.334813 | tok/s 2627477.84 | norm 0.3544 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14327 | train 3.023049 | val 3.334813 | tok/s 2629949.42 | norm 0.3448 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14328 | train 2.969368 | val 3.334813 | tok/s 2625374.83 | norm 0.4100 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14329 | train 2.956782 | val 3.334813 | tok/s 2628816.68 | norm 0.3550 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14330 | train 3.011002 | val 3.334813 | tok/s 2623537.19 | norm 0.3511 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14331 | train 2.903507 | val 3.334813 | tok/s 2628357.96 | norm 0.4011 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14332 | train 3.057859 | val 3.334813 | tok/s 2626034.76 | norm 0.3844 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14333 | train 3.040736 | val 3.334813 | tok/s 2626434.19 | norm 0.3954 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14334 | train 3.020844 | val 3.334813 | tok/s 2519180.34 | norm 0.4387 | dt 208.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14335 | train 2.852242 | val 3.334813 | tok/s 2612916.44 | norm 0.4049 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14336 | train 3.023570 | val 3.334813 | tok/s 2628615.60 | norm 0.4045 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14337 | train 2.962774 | val 3.334813 | tok/s 2620459.20 | norm 0.4048 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14338 | train 3.098344 | val 3.334813 | tok/s 2622628.29 | norm 0.3749 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14339 | train 2.955906 | val 3.334813 | tok/s 2629473.52 | norm 0.3489 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14340 | train 2.900275 | val 3.334813 | tok/s 2624686.46 | norm 0.3742 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14341 | train 2.999910 | val 3.334813 | tok/s 2630977.94 | norm 0.3565 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14342 | train 2.840401 | val 3.334813 | tok/s 2626942.13 | norm 0.3499 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14343 | train 3.034300 | val 3.334813 | tok/s 2606307.57 | norm 0.3694 | dt 201.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14344 | train 3.042361 | val 3.334813 | tok/s 2625101.95 | norm 0.3934 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14345 | train 2.978040 | val 3.334813 | tok/s 2626589.24 | norm 0.3953 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14346 | train 2.952799 | val 3.334813 | tok/s 2589645.78 | norm 0.4111 | dt 202.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14347 | train 2.903034 | val 3.334813 | tok/s 2626369.68 | norm 0.3866 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14348 | train 2.987527 | val 3.334813 | tok/s 2614639.65 | norm 0.3893 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14349 | train 3.017991 | val 3.334813 | tok/s 2631671.79 | norm 0.3777 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14350 | train 2.860485 | val 3.334813 | tok/s 2628443.09 | norm 0.3747 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14351 | train 2.920896 | val 3.334813 | tok/s 2627965.40 | norm 0.4007 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14352 | train 3.121864 | val 3.334813 | tok/s 2612490.65 | norm 0.3717 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14353 | train 2.968900 | val 3.334813 | tok/s 2580073.31 | norm 0.3517 | dt 203.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14354 | train 2.936833 | val 3.334813 | tok/s 2627508.39 | norm 0.3813 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14355 | train 2.982757 | val 3.334813 | tok/s 2627236.06 | norm 0.3945 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14356 | train 2.877687 | val 3.334813 | tok/s 2627826.18 | norm 0.3945 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14357 | train 2.806541 | val 3.334813 | tok/s 2617583.85 | norm 0.4178 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14358 | train 2.936404 | val 3.334813 | tok/s 2610773.09 | norm 0.4314 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14359 | train 3.073584 | val 3.334813 | tok/s 2631498.38 | norm 0.4016 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14360 | train 2.949356 | val 3.334813 | tok/s 2628564.03 | norm 0.3905 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14361 | train 2.916603 | val 3.334813 | tok/s 2628560.23 | norm 0.3877 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14362 | train 2.895732 | val 3.334813 | tok/s 2610173.27 | norm 0.3862 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14363 | train 2.995127 | val 3.334813 | tok/s 2618521.54 | norm 0.3816 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14364 | train 3.105765 | val 3.334813 | tok/s 2627307.87 | norm 0.3760 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14365 | train 3.025584 | val 3.334813 | tok/s 2621157.53 | norm 0.3702 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14366 | train 3.010689 | val 3.334813 | tok/s 2625218.42 | norm 0.3740 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14367 | train 2.903487 | val 3.334813 | tok/s 2624111.63 | norm 0.4049 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14368 | train 2.992405 | val 3.334813 | tok/s 2615145.64 | norm 0.3556 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14369 | train 3.031745 | val 3.334813 | tok/s 2630527.43 | norm 0.3530 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14370 | train 3.446010 | val 3.334813 | tok/s 2628725.71 | norm 0.4773 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14371 | train 2.931261 | val 3.334813 | tok/s 2628174.26 | norm 0.3871 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14372 | train 3.115390 | val 3.334813 | tok/s 2626244.61 | norm 0.3971 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14373 | train 3.012642 | val 3.334813 | tok/s 2630125.07 | norm 0.3748 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14374 | train 2.894725 | val 3.334813 | tok/s 2632051.18 | norm 0.3655 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14375 | train 2.929220 | val 3.334813 | tok/s 2630378.67 | norm 0.3739 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14376 | train 2.943962 | val 3.334813 | tok/s 2630341.91 | norm 0.3882 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14377 | train 3.010738 | val 3.334813 | tok/s 2630347.16 | norm 0.3586 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14378 | train 2.920500 | val 3.334813 | tok/s 2627687.81 | norm 0.3753 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14379 | train 2.940092 | val 3.334813 | tok/s 2631324.07 | norm 0.3879 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14380 | train 2.973352 | val 3.334813 | tok/s 2626873.32 | norm 0.3781 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14381 | train 3.056043 | val 3.334813 | tok/s 2631872.89 | norm 0.3941 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14382 | train 3.003685 | val 3.334813 | tok/s 2632639.42 | norm 0.4310 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14383 | train 2.977152 | val 3.334813 | tok/s 2632202.11 | norm 0.4195 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14384 | train 3.027278 | val 3.334813 | tok/s 2628623.96 | norm 0.3533 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14385 | train 3.011788 | val 3.334813 | tok/s 2586336.20 | norm 0.3711 | dt 202.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14386 | train 2.918013 | val 3.334813 | tok/s 2629072.21 | norm 0.4043 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14387 | train 3.073298 | val 3.334813 | tok/s 2627051.86 | norm 0.3959 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14388 | train 2.868539 | val 3.334813 | tok/s 2628532.35 | norm 0.3833 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14389 | train 3.097759 | val 3.334813 | tok/s 2633670.93 | norm 0.3819 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14390 | train 2.911308 | val 3.334813 | tok/s 2624056.87 | norm 0.3875 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14391 | train 3.090608 | val 3.334813 | tok/s 2630595.04 | norm 0.4285 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14392 | train 2.943045 | val 3.334813 | tok/s 2625031.94 | norm 0.4065 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14393 | train 2.900198 | val 3.334813 | tok/s 2627309.38 | norm 0.3631 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14394 | train 2.991439 | val 3.334813 | tok/s 2630764.35 | norm 0.4774 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14395 | train 2.966795 | val 3.334813 | tok/s 2627547.08 | norm 0.3655 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14396 | train 2.920386 | val 3.334813 | tok/s 2628169.06 | norm 0.3444 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14397 | train 3.079712 | val 3.334813 | tok/s 2627941.01 | norm 0.4033 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14398 | train 2.969269 | val 3.334813 | tok/s 2630219.67 | norm 0.3898 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14399 | train 2.956444 | val 3.334813 | tok/s 2626911.98 | norm 0.3969 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14400 | train 3.042213 | val 3.334813 | tok/s 2630184.67 | norm 0.4700 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14401 | train 2.963845 | val 3.334813 | tok/s 2629653.84 | norm 0.3992 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14402 | train 3.088943 | val 3.334813 | tok/s 2630291.98 | norm 0.3698 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14403 | train 2.968162 | val 3.334813 | tok/s 2630978.03 | norm 0.3892 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14404 | train 2.983174 | val 3.334813 | tok/s 2619762.98 | norm 0.4177 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14405 | train 2.892479 | val 3.334813 | tok/s 2625986.40 | norm 0.4266 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14406 | train 2.899118 | val 3.334813 | tok/s 2629339.83 | norm 0.4115 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14407 | train 2.873060 | val 3.334813 | tok/s 2622325.44 | norm 0.3990 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14408 | train 3.014232 | val 3.334813 | tok/s 2628106.64 | norm 0.3835 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14409 | train 2.916120 | val 3.334813 | tok/s 2629483.88 | norm 0.3877 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14410 | train 3.021231 | val 3.334813 | tok/s 2629822.99 | norm 0.3805 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14411 | train 2.868784 | val 3.334813 | tok/s 2625828.06 | norm 0.3742 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14412 | train 3.001280 | val 3.334813 | tok/s 2627462.47 | norm 0.3878 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14413 | train 3.001361 | val 3.334813 | tok/s 2633593.20 | norm 0.3999 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14414 | train 2.930205 | val 3.334813 | tok/s 2628392.02 | norm 0.4051 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14415 | train 3.012587 | val 3.334813 | tok/s 2630167.77 | norm 0.3993 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14416 | train 3.062394 | val 3.334813 | tok/s 2630452.63 | norm 0.3598 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14417 | train 3.033653 | val 3.334813 | tok/s 2628866.98 | norm 0.3951 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14418 | train 3.097960 | val 3.334813 | tok/s 2627467.62 | norm 0.4250 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14419 | train 3.022824 | val 3.334813 | tok/s 2618261.94 | norm 0.3949 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14420 | train 2.923240 | val 3.334813 | tok/s 2626023.34 | norm 0.4043 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14421 | train 3.016981 | val 3.334813 | tok/s 2629191.20 | norm 0.3819 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14422 | train 3.009725 | val 3.334813 | tok/s 2630195.61 | norm 0.4009 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14423 | train 2.973386 | val 3.334813 | tok/s 2630304.34 | norm 0.3631 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14424 | train 3.034641 | val 3.334813 | tok/s 2622701.06 | norm 0.3793 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14425 | train 2.908101 | val 3.334813 | tok/s 2626831.69 | norm 0.3700 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14426 | train 3.005757 | val 3.334813 | tok/s 2626894.80 | norm 0.3464 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14427 | train 2.944215 | val 3.334813 | tok/s 2629677.17 | norm 0.3669 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14428 | train 3.109162 | val 3.334813 | tok/s 2633948.26 | norm 0.4097 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14429 | train 2.933147 | val 3.334813 | tok/s 2623614.26 | norm 0.3832 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14430 | train 3.009288 | val 3.334813 | tok/s 2625781.68 | norm 0.3497 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14431 | train 2.775083 | val 3.334813 | tok/s 2632344.62 | norm 0.3536 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14432 | train 2.958800 | val 3.334813 | tok/s 2623738.28 | norm 0.3791 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14433 | train 3.027132 | val 3.334813 | tok/s 2629850.58 | norm 0.3688 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14434 | train 2.901613 | val 3.334813 | tok/s 2628551.06 | norm 0.3880 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14435 | train 3.064904 | val 3.334813 | tok/s 2630476.30 | norm 0.3924 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14436 | train 2.951173 | val 3.334813 | tok/s 2627478.73 | norm 0.3611 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14437 | train 3.045793 | val 3.334813 | tok/s 2627977.53 | norm 0.3628 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14438 | train 2.969672 | val 3.334813 | tok/s 2626638.00 | norm 0.3683 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14439 | train 3.104788 | val 3.334813 | tok/s 2628084.01 | norm 0.3631 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14440 | train 2.980579 | val 3.334813 | tok/s 2626765.53 | norm 0.3726 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14441 | train 3.031020 | val 3.334813 | tok/s 2629753.45 | norm 0.3952 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14442 | train 2.932025 | val 3.334813 | tok/s 2630329.42 | norm 0.3855 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14443 | train 2.935479 | val 3.334813 | tok/s 2626091.48 | norm 0.3986 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14444 | train 2.991954 | val 3.334813 | tok/s 2624876.48 | norm 0.3715 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14445 | train 2.964165 | val 3.334813 | tok/s 2631187.50 | norm 0.3956 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14446 | train 2.862321 | val 3.334813 | tok/s 2632870.38 | norm 0.4273 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14447 | train 3.006087 | val 3.334813 | tok/s 2630994.14 | norm 0.3969 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14448 | train 2.951699 | val 3.334813 | tok/s 2630778.50 | norm 0.3836 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14449 | train 2.952837 | val 3.334813 | tok/s 2623563.58 | norm 0.3769 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14450 | train 2.938208 | val 3.334813 | tok/s 2628648.91 | norm 0.4005 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14451 | train 3.032727 | val 3.334813 | tok/s 2631855.33 | norm 0.4039 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14452 | train 2.920097 | val 3.334813 | tok/s 2621805.27 | norm 0.3950 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14453 | train 3.024265 | val 3.334813 | tok/s 2528683.57 | norm 0.3798 | dt 207.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14454 | train 2.871684 | val 3.334813 | tok/s 2626997.54 | norm 0.3939 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14455 | train 3.066725 | val 3.334813 | tok/s 2626779.80 | norm 0.3707 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14456 | train 2.972003 | val 3.334813 | tok/s 2628298.43 | norm 0.3729 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14457 | train 3.047689 | val 3.334813 | tok/s 2630028.81 | norm 0.4131 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14458 | train 2.929462 | val 3.334813 | tok/s 2629659.83 | norm 0.3597 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14459 | train 3.036135 | val 3.334813 | tok/s 2465463.75 | norm 0.3677 | dt 212.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14460 | train 3.011039 | val 3.334813 | tok/s 2553502.43 | norm 0.4145 | dt 205.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14461 | train 2.968716 | val 3.334813 | tok/s 2628323.66 | norm 0.4033 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14462 | train 2.823734 | val 3.334813 | tok/s 2629924.57 | norm 0.3891 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14463 | train 2.932087 | val 3.334813 | tok/s 2627916.20 | norm 0.3775 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14464 | train 2.937626 | val 3.334813 | tok/s 2629059.40 | norm 0.4360 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14465 | train 2.963735 | val 3.334813 | tok/s 2630704.54 | norm 0.4303 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14466 | train 2.995419 | val 3.334813 | tok/s 2624238.15 | norm 0.3739 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14467 | train 2.996656 | val 3.334813 | tok/s 2621217.92 | norm 0.3943 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14468 | train 3.003814 | val 3.334813 | tok/s 2632325.84 | norm 0.3970 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14469 | train 2.996423 | val 3.334813 | tok/s 2628734.61 | norm 0.3890 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14470 | train 2.944803 | val 3.334813 | tok/s 2630235.89 | norm 0.3815 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14471 | train 3.062842 | val 3.334813 | tok/s 2627596.30 | norm 0.4209 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14472 | train 3.065207 | val 3.334813 | tok/s 2627734.71 | norm 0.3898 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14473 | train 2.985078 | val 3.334813 | tok/s 2631295.93 | norm 0.3716 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14474 | train 2.977069 | val 3.334813 | tok/s 2626930.31 | norm 0.3974 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14475 | train 2.945405 | val 3.334813 | tok/s 2631112.31 | norm 0.3845 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14476 | train 2.940298 | val 3.334813 | tok/s 2629854.38 | norm 0.3862 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14477 | train 2.866022 | val 3.334813 | tok/s 2628323.45 | norm 0.3875 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14478 | train 3.005106 | val 3.334813 | tok/s 2627754.23 | norm 0.3812 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14479 | train 3.086954 | val 3.334813 | tok/s 2625275.12 | norm 0.4321 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14480 | train 3.030997 | val 3.334813 | tok/s 2623962.81 | norm 0.4280 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14481 | train 2.979701 | val 3.334813 | tok/s 2627489.80 | norm 0.3672 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14482 | train 3.030814 | val 3.334813 | tok/s 2628448.90 | norm 0.3976 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14483 | train 3.090952 | val 3.334813 | tok/s 2630194.97 | norm 0.3736 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14484 | train 3.212818 | val 3.334813 | tok/s 2627923.31 | norm 0.5062 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14485 | train 2.946301 | val 3.334813 | tok/s 2631587.19 | norm 0.4244 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14486 | train 3.056031 | val 3.334813 | tok/s 2626868.71 | norm 0.3679 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14487 | train 3.008721 | val 3.334813 | tok/s 2626360.51 | norm 0.3996 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14488 | train 2.851875 | val 3.334813 | tok/s 2630994.77 | norm 0.4065 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14489 | train 2.912804 | val 3.334813 | tok/s 2621515.36 | norm 0.3731 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14490 | train 2.895989 | val 3.334813 | tok/s 2629628.60 | norm 0.3796 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14491 | train 2.928634 | val 3.334813 | tok/s 2632854.69 | norm 0.4253 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14492 | train 2.871887 | val 3.334813 | tok/s 2629278.46 | norm 0.3492 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14493 | train 2.862926 | val 3.334813 | tok/s 2629560.66 | norm 0.3518 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14494 | train 3.031811 | val 3.334813 | tok/s 2631149.27 | norm 0.3583 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14495 | train 2.943817 | val 3.334813 | tok/s 2630443.14 | norm 0.3464 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14496 | train 3.023103 | val 3.334813 | tok/s 2627117.27 | norm 0.3469 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14497 | train 2.917449 | val 3.334813 | tok/s 2612761.36 | norm 0.3592 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14498 | train 3.023763 | val 3.334813 | tok/s 2632582.87 | norm 0.3440 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14499 | train 2.942168 | val 3.334813 | tok/s 2622435.08 | norm 0.3528 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 14500 | val_loss 3.336481 step 14500 | train 2.992173 | val 3.336481 | tok/s 2614884.24 | norm 0.3555 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14501 | train 2.986856 | val 3.336481 | tok/s 2626417.85 | norm 0.3600 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14502 | train 2.881016 | val 3.336481 | tok/s 2632892.86 | norm 0.3618 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14503 | train 2.958604 | val 3.336481 | tok/s 2628766.79 | norm 0.3889 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14504 | train 2.943515 | val 3.336481 | tok/s 2624739.56 | norm 0.4043 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14505 | train 2.938146 | val 3.336481 | tok/s 2622006.07 | norm 0.3625 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14506 | train 2.906919 | val 3.336481 | tok/s 2631178.08 | norm 0.3758 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14507 | train 3.027054 | val 3.336481 | tok/s 2630572.45 | norm 0.4179 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14508 | train 2.926822 | val 3.336481 | tok/s 2626063.63 | norm 0.3627 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14509 | train 3.012362 | val 3.336481 | tok/s 2626267.20 | norm 0.3928 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14510 | train 2.935507 | val 3.336481 | tok/s 2620276.41 | norm 0.4071 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14511 | train 3.141635 | val 3.336481 | tok/s 2633183.76 | norm 0.4198 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14512 | train 3.008778 | val 3.336481 | tok/s 2631807.91 | norm 0.3991 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14513 | train 2.882513 | val 3.336481 | tok/s 2633576.55 | norm 0.3964 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14514 | train 2.937831 | val 3.336481 | tok/s 2628229.70 | norm 0.4221 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14515 | train 2.968496 | val 3.336481 | tok/s 2628920.45 | norm 0.3824 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14516 | train 2.931623 | val 3.336481 | tok/s 2628402.70 | norm 0.3534 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14517 | train 2.860918 | val 3.336481 | tok/s 2625206.33 | norm 0.3702 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14518 | train 2.864687 | val 3.336481 | tok/s 2484422.66 | norm 0.4049 | dt 211.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14519 | train 2.932205 | val 3.336481 | tok/s 2627730.95 | norm 0.4517 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14520 | train 2.971886 | val 3.336481 | tok/s 2556245.76 | norm 0.3858 | dt 205.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14521 | train 2.944894 | val 3.336481 | tok/s 2631879.39 | norm 0.3970 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14522 | train 2.947572 | val 3.336481 | tok/s 2625363.21 | norm 0.3753 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14523 | train 2.930417 | val 3.336481 | tok/s 2630603.85 | norm 0.3699 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14524 | train 2.968859 | val 3.336481 | tok/s 2628074.34 | norm 0.4216 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14525 | train 2.958301 | val 3.336481 | tok/s 2628134.19 | norm 0.4099 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14526 | train 3.078658 | val 3.336481 | tok/s 2627419.99 | norm 0.3565 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14527 | train 2.996491 | val 3.336481 | tok/s 2626930.15 | norm 0.3973 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14528 | train 2.945228 | val 3.336481 | tok/s 2624507.13 | norm 0.3698 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14529 | train 3.016742 | val 3.336481 | tok/s 2628965.27 | norm 0.3592 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14530 | train 2.847019 | val 3.336481 | tok/s 2624863.04 | norm 0.4087 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14531 | train 3.088988 | val 3.336481 | tok/s 2630563.02 | norm 0.4406 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14532 | train 2.909145 | val 3.336481 | tok/s 2629370.17 | norm 0.4262 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14533 | train 2.906043 | val 3.336481 | tok/s 2628750.20 | norm 0.3896 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14534 | train 2.949569 | val 3.336481 | tok/s 2630435.02 | norm 0.3704 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14535 | train 2.929708 | val 3.336481 | tok/s 2615371.62 | norm 0.4132 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14536 | train 2.945927 | val 3.336481 | tok/s 2627921.06 | norm 0.3975 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14537 | train 2.932119 | val 3.336481 | tok/s 2628851.09 | norm 0.3900 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14538 | train 3.009945 | val 3.336481 | tok/s 2630295.23 | norm 0.3814 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14539 | train 2.997869 | val 3.336481 | tok/s 2627256.97 | norm 0.3874 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14540 | train 2.992414 | val 3.336481 | tok/s 2621109.27 | norm 0.3486 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14541 | train 2.992345 | val 3.336481 | tok/s 2627819.32 | norm 0.3613 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14542 | train 3.012384 | val 3.336481 | tok/s 2629441.02 | norm 0.3915 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14543 | train 3.018334 | val 3.336481 | tok/s 2627430.67 | norm 0.3749 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14544 | train 3.041471 | val 3.336481 | tok/s 2629793.08 | norm 0.5596 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14545 | train 2.984792 | val 3.336481 | tok/s 2630101.60 | norm 0.3956 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14546 | train 2.967488 | val 3.336481 | tok/s 2630760.53 | norm 0.3360 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14547 | train 2.887152 | val 3.336481 | tok/s 2622305.11 | norm 0.3747 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14548 | train 2.969517 | val 3.336481 | tok/s 2630113.42 | norm 0.3663 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14549 | train 2.867519 | val 3.336481 | tok/s 2630369.22 | norm 0.3427 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14550 | train 2.953966 | val 3.336481 | tok/s 2631250.13 | norm 0.3498 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14551 | train 3.031339 | val 3.336481 | tok/s 2634326.47 | norm 0.3763 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14552 | train 3.046259 | val 3.336481 | tok/s 2622509.04 | norm 0.3536 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14553 | train 2.976525 | val 3.336481 | tok/s 2630907.79 | norm 0.3545 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14554 | train 2.895710 | val 3.336481 | tok/s 2626061.58 | norm 0.3751 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14555 | train 2.906523 | val 3.336481 | tok/s 2625274.96 | norm 0.3918 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14556 | train 2.903502 | val 3.336481 | tok/s 2629363.63 | norm 0.3689 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14557 | train 2.983208 | val 3.336481 | tok/s 2626638.29 | norm 0.3762 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14558 | train 2.866279 | val 3.336481 | tok/s 2630184.16 | norm 0.3976 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14559 | train 2.961489 | val 3.336481 | tok/s 2625163.01 | norm 0.4136 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14560 | train 3.072055 | val 3.336481 | tok/s 2626813.29 | norm 0.4043 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14561 | train 3.012686 | val 3.336481 | tok/s 2631280.37 | norm 0.3740 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14562 | train 3.001883 | val 3.336481 | tok/s 2625538.40 | norm 0.4122 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14563 | train 3.128206 | val 3.336481 | tok/s 2543507.76 | norm 0.4106 | dt 206.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14564 | train 2.978353 | val 3.336481 | tok/s 2630570.51 | norm 0.3757 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14565 | train 2.908066 | val 3.336481 | tok/s 2630063.77 | norm 0.3619 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14566 | train 2.945968 | val 3.336481 | tok/s 2624603.57 | norm 0.3423 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14567 | train 2.844964 | val 3.336481 | tok/s 2626199.29 | norm 0.3822 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14568 | train 3.139403 | val 3.336481 | tok/s 2630358.47 | norm 0.3842 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14569 | train 2.903189 | val 3.336481 | tok/s 2632046.98 | norm 0.3589 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14570 | train 2.877188 | val 3.336481 | tok/s 2628336.06 | norm 0.3912 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14571 | train 2.931278 | val 3.336481 | tok/s 2629131.66 | norm 0.4872 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14572 | train 2.980221 | val 3.336481 | tok/s 2629393.93 | norm 0.3921 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14573 | train 2.912951 | val 3.336481 | tok/s 2632791.39 | norm 0.3553 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14574 | train 2.994264 | val 3.336481 | tok/s 2626529.99 | norm 0.3892 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14575 | train 2.973998 | val 3.336481 | tok/s 2627266.09 | norm 0.3913 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14576 | train 2.822687 | val 3.336481 | tok/s 2622975.50 | norm 0.3569 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14577 | train 3.008057 | val 3.336481 | tok/s 2624220.39 | norm 0.3672 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14578 | train 2.913134 | val 3.336481 | tok/s 2631400.27 | norm 0.4427 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14579 | train 2.898661 | val 3.336481 | tok/s 2503925.58 | norm 0.3847 | dt 209.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14580 | train 2.948737 | val 3.336481 | tok/s 2630520.29 | norm 0.3491 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14581 | train 2.959631 | val 3.336481 | tok/s 2626932.17 | norm 0.3590 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14582 | train 3.020740 | val 3.336481 | tok/s 2627011.02 | norm 0.4392 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14583 | train 2.854761 | val 3.336481 | tok/s 2627574.39 | norm 0.3571 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14584 | train 2.911091 | val 3.336481 | tok/s 2630962.91 | norm 0.3332 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14585 | train 3.031534 | val 3.336481 | tok/s 2629789.48 | norm 0.3889 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14586 | train 3.085907 | val 3.336481 | tok/s 2625842.05 | norm 0.4090 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14587 | train 2.941917 | val 3.336481 | tok/s 2630780.55 | norm 0.3644 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14588 | train 3.046970 | val 3.336481 | tok/s 2628925.50 | norm 0.3809 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14589 | train 2.890275 | val 3.336481 | tok/s 2630965.11 | norm 0.3725 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14590 | train 3.015415 | val 3.336481 | tok/s 2631266.50 | norm 0.3638 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14591 | train 2.988523 | val 3.336481 | tok/s 2627875.05 | norm 0.3887 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14592 | train 2.879661 | val 3.336481 | tok/s 2627054.20 | norm 0.4054 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14593 | train 3.059783 | val 3.336481 | tok/s 2629714.40 | norm 0.4118 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14594 | train 2.885812 | val 3.336481 | tok/s 2494738.58 | norm 0.3819 | dt 210.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14595 | train 2.994272 | val 3.336481 | tok/s 2626628.91 | norm 0.3751 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14596 | train 2.993457 | val 3.336481 | tok/s 2628989.38 | norm 0.3764 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14597 | train 2.933422 | val 3.336481 | tok/s 2629972.89 | norm 0.3760 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14598 | train 2.935101 | val 3.336481 | tok/s 2633858.14 | norm 0.3819 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14599 | train 2.951851 | val 3.336481 | tok/s 2623593.66 | norm 0.3856 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14600 | train 2.966098 | val 3.336481 | tok/s 2631428.77 | norm 0.3590 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14601 | train 2.912623 | val 3.336481 | tok/s 2625416.38 | norm 0.3606 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14602 | train 3.042648 | val 3.336481 | tok/s 2630773.81 | norm 0.3744 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14603 | train 2.810050 | val 3.336481 | tok/s 2626453.20 | norm 0.4036 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14604 | train 2.839699 | val 3.336481 | tok/s 2619732.73 | norm 0.3655 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14605 | train 3.114098 | val 3.336481 | tok/s 2629708.38 | norm 0.4305 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14606 | train 3.028196 | val 3.336481 | tok/s 2630700.34 | norm 0.4570 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14607 | train 3.054840 | val 3.336481 | tok/s 2630829.38 | norm 0.5070 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14608 | train 3.011290 | val 3.336481 | tok/s 2625050.80 | norm 0.4589 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14609 | train 3.007953 | val 3.336481 | tok/s 2624522.50 | norm 0.4016 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14610 | train 2.929288 | val 3.336481 | tok/s 2629452.32 | norm 0.4060 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14611 | train 3.065528 | val 3.336481 | tok/s 2626629.38 | norm 0.4302 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14612 | train 2.943008 | val 3.336481 | tok/s 2630775.85 | norm 0.3796 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14613 | train 2.958169 | val 3.336481 | tok/s 2621685.28 | norm 0.3666 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14614 | train 2.956993 | val 3.336481 | tok/s 2626161.55 | norm 0.3594 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14615 | train 3.005693 | val 3.336481 | tok/s 2627710.83 | norm 0.3689 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14616 | train 2.986877 | val 3.336481 | tok/s 2626349.86 | norm 0.3811 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14617 | train 2.890794 | val 3.336481 | tok/s 2630349.86 | norm 0.3922 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14618 | train 2.977488 | val 3.336481 | tok/s 2483330.45 | norm 0.3907 | dt 211.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14619 | train 2.959827 | val 3.336481 | tok/s 2621939.03 | norm 0.3749 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14620 | train 2.921680 | val 3.336481 | tok/s 2633347.64 | norm 0.3820 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14621 | train 2.898284 | val 3.336481 | tok/s 2627029.53 | norm 0.3911 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14622 | train 3.078888 | val 3.336481 | tok/s 2628522.25 | norm 0.3980 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14623 | train 3.043359 | val 3.336481 | tok/s 2630867.79 | norm 0.5495 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14624 | train 2.808294 | val 3.336481 | tok/s 2628475.51 | norm 0.3352 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14625 | train 2.985924 | val 3.336481 | tok/s 2627524.66 | norm 0.3832 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14626 | train 2.876114 | val 3.336481 | tok/s 2629869.69 | norm 0.4128 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14627 | train 2.959920 | val 3.336481 | tok/s 2563809.96 | norm 0.3932 | dt 204.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14628 | train 2.995649 | val 3.336481 | tok/s 2619449.72 | norm 0.3790 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14629 | train 2.957127 | val 3.336481 | tok/s 2623527.75 | norm 0.3971 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14630 | train 2.985244 | val 3.336481 | tok/s 2625174.38 | norm 0.3623 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14631 | train 3.099304 | val 3.336481 | tok/s 2626459.14 | norm 0.3697 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14632 | train 2.988357 | val 3.336481 | tok/s 2621974.38 | norm 0.3610 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14633 | train 2.875089 | val 3.336481 | tok/s 2622425.28 | norm 0.3827 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14634 | train 2.903242 | val 3.336481 | tok/s 2623955.01 | norm 0.3581 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14635 | train 2.961129 | val 3.336481 | tok/s 2627648.40 | norm 0.3501 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14636 | train 3.031760 | val 3.336481 | tok/s 2623027.30 | norm 0.4142 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14637 | train 3.010549 | val 3.336481 | tok/s 2627253.11 | norm 0.3814 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14638 | train 2.920944 | val 3.336481 | tok/s 2624145.66 | norm 0.3811 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14639 | train 3.108742 | val 3.336481 | tok/s 2625853.46 | norm 0.4349 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14640 | train 3.048648 | val 3.336481 | tok/s 2627566.71 | norm 0.3748 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14641 | train 3.142748 | val 3.336481 | tok/s 2630696.04 | norm 0.4104 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14642 | train 2.987769 | val 3.336481 | tok/s 2629531.63 | norm 0.4202 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14643 | train 2.901454 | val 3.336481 | tok/s 2622280.83 | norm 0.3778 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14644 | train 2.966072 | val 3.336481 | tok/s 2626724.46 | norm 0.3681 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14645 | train 2.940499 | val 3.336481 | tok/s 2624090.11 | norm 0.3380 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14646 | train 2.942419 | val 3.336481 | tok/s 2626052.02 | norm 0.3680 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14647 | train 2.899163 | val 3.336481 | tok/s 2624180.04 | norm 0.3533 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14648 | train 2.985831 | val 3.336481 | tok/s 2618867.84 | norm 0.3837 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14649 | train 2.970118 | val 3.336481 | tok/s 2623884.21 | norm 0.3643 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14650 | train 2.999076 | val 3.336481 | tok/s 2625243.35 | norm 0.3752 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14651 | train 2.959995 | val 3.336481 | tok/s 2621294.57 | norm 0.3616 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14652 | train 3.050639 | val 3.336481 | tok/s 2627583.07 | norm 0.4068 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14653 | train 2.865646 | val 3.336481 | tok/s 2626253.24 | norm 0.3931 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14654 | train 2.947298 | val 3.336481 | tok/s 2625503.06 | norm 0.3758 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14655 | train 2.922501 | val 3.336481 | tok/s 2623025.36 | norm 0.3649 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14656 | train 3.010658 | val 3.336481 | tok/s 2619926.69 | norm 0.3594 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14657 | train 3.002268 | val 3.336481 | tok/s 2622702.50 | norm 0.3678 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14658 | train 3.047092 | val 3.336481 | tok/s 2623045.78 | norm 0.3757 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14659 | train 2.867986 | val 3.336481 | tok/s 2625432.40 | norm 0.3795 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14660 | train 3.111476 | val 3.336481 | tok/s 2624141.92 | norm 0.4538 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14661 | train 2.952840 | val 3.336481 | tok/s 2621371.03 | norm 0.3436 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14662 | train 2.926177 | val 3.336481 | tok/s 2623977.12 | norm 0.3963 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14663 | train 3.013137 | val 3.336481 | tok/s 2623397.51 | norm 0.3915 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14664 | train 2.929682 | val 3.336481 | tok/s 2626758.94 | norm 0.4117 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14665 | train 2.818336 | val 3.336481 | tok/s 2620593.30 | norm 0.4144 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14666 | train 2.946788 | val 3.336481 | tok/s 2622930.98 | norm 0.3739 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14667 | train 3.082661 | val 3.336481 | tok/s 2623007.26 | norm 0.3901 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14668 | train 2.928761 | val 3.336481 | tok/s 2623158.39 | norm 0.3725 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14669 | train 2.890195 | val 3.336481 | tok/s 2621435.50 | norm 0.3656 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14670 | train 3.016246 | val 3.336481 | tok/s 2622947.50 | norm 0.3640 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14671 | train 2.911667 | val 3.336481 | tok/s 2623970.48 | norm 0.3378 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14672 | train 2.905357 | val 3.336481 | tok/s 2626505.86 | norm 0.3724 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14673 | train 2.876724 | val 3.336481 | tok/s 2623297.06 | norm 0.3974 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14674 | train 2.923946 | val 3.336481 | tok/s 2482188.68 | norm 0.3978 | dt 211.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14675 | train 2.878698 | val 3.336481 | tok/s 2623956.54 | norm 0.3971 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14676 | train 3.026767 | val 3.336481 | tok/s 2623112.47 | norm 0.3795 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14677 | train 3.002285 | val 3.336481 | tok/s 2628529.25 | norm 0.3853 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14678 | train 2.870199 | val 3.336481 | tok/s 2622989.87 | norm 0.3931 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14679 | train 2.997680 | val 3.336481 | tok/s 2628494.26 | norm 0.4095 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14680 | train 3.042982 | val 3.336481 | tok/s 2626396.56 | norm 0.3802 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14681 | train 2.793211 | val 3.336481 | tok/s 2627105.46 | norm 0.3697 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14682 | train 2.993468 | val 3.336481 | tok/s 2624892.71 | norm 0.3575 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14683 | train 2.947653 | val 3.336481 | tok/s 2629089.40 | norm 0.3578 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14684 | train 2.854066 | val 3.336481 | tok/s 2625016.61 | norm 0.3459 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14685 | train 2.973432 | val 3.336481 | tok/s 2624752.95 | norm 0.3919 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14686 | train 3.042431 | val 3.336481 | tok/s 2622443.77 | norm 0.5104 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14687 | train 2.921479 | val 3.336481 | tok/s 2625941.69 | norm 0.3686 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14688 | train 2.919138 | val 3.336481 | tok/s 2623733.21 | norm 0.3787 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14689 | train 2.953814 | val 3.336481 | tok/s 2625190.72 | norm 0.3885 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14690 | train 2.863590 | val 3.336481 | tok/s 2617747.69 | norm 0.3951 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14691 | train 3.210476 | val 3.336481 | tok/s 2622155.52 | norm 0.3800 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14692 | train 3.126199 | val 3.336481 | tok/s 2627028.62 | norm 0.3658 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14693 | train 2.932809 | val 3.336481 | tok/s 2624540.80 | norm 0.3371 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14694 | train 2.935942 | val 3.336481 | tok/s 2626017.33 | norm 0.3818 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14695 | train 2.986958 | val 3.336481 | tok/s 2618487.81 | norm 0.4070 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14696 | train 3.124877 | val 3.336481 | tok/s 2606658.62 | norm 0.3671 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14697 | train 2.984452 | val 3.336481 | tok/s 2624157.31 | norm 0.3762 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14698 | train 2.957781 | val 3.336481 | tok/s 2625112.36 | norm 0.3600 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14699 | train 2.991040 | val 3.336481 | tok/s 2621011.28 | norm 0.3632 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14700 | train 2.869194 | val 3.336481 | tok/s 2620177.73 | norm 0.3573 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14701 | train 3.152488 | val 3.336481 | tok/s 2627036.20 | norm 0.3666 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14702 | train 2.863251 | val 3.336481 | tok/s 2624362.45 | norm 0.3461 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14703 | train 2.943890 | val 3.336481 | tok/s 2624876.34 | norm 0.3868 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14704 | train 2.972511 | val 3.336481 | tok/s 2617540.47 | norm 0.4285 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14705 | train 2.944920 | val 3.336481 | tok/s 2556306.18 | norm 0.4206 | dt 205.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14706 | train 2.892125 | val 3.336481 | tok/s 2629283.02 | norm 0.4137 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14707 | train 2.839023 | val 3.336481 | tok/s 2623724.84 | norm 0.4557 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14708 | train 2.915437 | val 3.336481 | tok/s 2625255.35 | norm 0.3803 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14709 | train 2.928398 | val 3.336481 | tok/s 2625633.61 | norm 0.3706 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14710 | train 2.928739 | val 3.336481 | tok/s 2623300.39 | norm 0.3925 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14711 | train 2.981081 | val 3.336481 | tok/s 2627498.13 | norm 0.3697 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14712 | train 2.971142 | val 3.336481 | tok/s 2627186.67 | norm 0.3820 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14713 | train 3.043665 | val 3.336481 | tok/s 2624692.97 | norm 0.4156 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14714 | train 2.987355 | val 3.336481 | tok/s 2625390.68 | norm 0.3947 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14715 | train 3.006485 | val 3.336481 | tok/s 2629457.50 | norm 0.3946 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14716 | train 2.885048 | val 3.336481 | tok/s 2627196.57 | norm 0.3785 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14717 | train 2.873246 | val 3.336481 | tok/s 2629235.83 | norm 0.3429 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14718 | train 2.973232 | val 3.336481 | tok/s 2625074.06 | norm 0.3542 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14719 | train 2.958267 | val 3.336481 | tok/s 2627589.65 | norm 0.3581 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14720 | train 3.011140 | val 3.336481 | tok/s 2631573.25 | norm 0.3580 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14721 | train 2.942934 | val 3.336481 | tok/s 2630353.48 | norm 0.3713 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14722 | train 2.986053 | val 3.336481 | tok/s 2623852.20 | norm 0.3589 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14723 | train 3.136071 | val 3.336481 | tok/s 2620318.37 | norm 0.4151 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14724 | train 3.012446 | val 3.336481 | tok/s 2624009.48 | norm 0.4156 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14725 | train 3.034818 | val 3.336481 | tok/s 2622290.93 | norm 0.3887 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14726 | train 3.003247 | val 3.336481 | tok/s 2624335.13 | norm 0.4086 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14727 | train 2.946595 | val 3.336481 | tok/s 2622000.35 | norm 0.4085 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14728 | train 2.914524 | val 3.336481 | tok/s 2630939.81 | norm 0.3683 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14729 | train 2.892684 | val 3.336481 | tok/s 2621438.35 | norm 0.3889 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14730 | train 2.994479 | val 3.336481 | tok/s 2628323.70 | norm 0.3899 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14731 | train 2.873241 | val 3.336481 | tok/s 2626093.42 | norm 0.4020 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14732 | train 2.823749 | val 3.336481 | tok/s 2623578.68 | norm 0.3979 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14733 | train 3.181418 | val 3.336481 | tok/s 2628756.92 | norm 0.3972 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14734 | train 2.990844 | val 3.336481 | tok/s 2623116.00 | norm 0.3753 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14735 | train 2.998014 | val 3.336481 | tok/s 2625064.08 | norm 0.3518 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14736 | train 2.999647 | val 3.336481 | tok/s 2625909.76 | norm 0.3636 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14737 | train 2.983548 | val 3.336481 | tok/s 2622803.71 | norm 0.3791 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14738 | train 2.981153 | val 3.336481 | tok/s 2631537.47 | norm 0.3636 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14739 | train 2.914062 | val 3.336481 | tok/s 2628753.57 | norm 0.4287 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14740 | train 2.918909 | val 3.336481 | tok/s 2627090.56 | norm 0.4121 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14741 | train 2.981702 | val 3.336481 | tok/s 2625527.63 | norm 0.4228 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14742 | train 3.060289 | val 3.336481 | tok/s 2621409.36 | norm 0.3918 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14743 | train 2.897353 | val 3.336481 | tok/s 2620103.69 | norm 0.3787 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14744 | train 2.892707 | val 3.336481 | tok/s 2611801.56 | norm 0.4278 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14745 | train 3.079415 | val 3.336481 | tok/s 2618802.36 | norm 0.3926 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14746 | train 3.065096 | val 3.336481 | tok/s 2615645.34 | norm 0.4090 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14747 | train 2.977722 | val 3.336481 | tok/s 2621031.16 | norm 0.4245 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14748 | train 2.850596 | val 3.336481 | tok/s 2623319.56 | norm 0.3488 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14749 | train 2.976789 | val 3.336481 | tok/s 2623272.81 | norm 0.3730 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 14750 | val_loss 3.332533 step 14750 | train 3.029193 | val 3.332533 | tok/s 2621993.40 | norm 0.4255 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14751 | train 2.994149 | val 3.332533 | tok/s 2620776.85 | norm 0.4140 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14752 | train 2.921418 | val 3.332533 | tok/s 2623781.42 | norm 0.3713 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14753 | train 3.086494 | val 3.332533 | tok/s 2624401.86 | norm 0.3786 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14754 | train 2.940904 | val 3.332533 | tok/s 2580784.76 | norm 0.3959 | dt 203.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14755 | train 2.842564 | val 3.332533 | tok/s 2624108.57 | norm 0.3824 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14756 | train 3.096354 | val 3.332533 | tok/s 2626953.41 | norm 0.3386 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14757 | train 2.882638 | val 3.332533 | tok/s 2625965.54 | norm 0.3416 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14758 | train 3.074221 | val 3.332533 | tok/s 2623104.98 | norm 0.4101 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14759 | train 2.900247 | val 3.332533 | tok/s 2621652.59 | norm 0.3798 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14760 | train 2.904052 | val 3.332533 | tok/s 2621734.67 | norm 0.3919 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14761 | train 2.894066 | val 3.332533 | tok/s 2626201.95 | norm 0.4024 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14762 | train 3.002821 | val 3.332533 | tok/s 2623826.16 | norm 0.3893 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14763 | train 2.877210 | val 3.332533 | tok/s 2598836.27 | norm 0.3695 | dt 201.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14764 | train 2.983721 | val 3.332533 | tok/s 2625380.67 | norm 0.3602 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14765 | train 2.946113 | val 3.332533 | tok/s 2624253.08 | norm 0.3791 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14766 | train 2.901193 | val 3.332533 | tok/s 2625627.00 | norm 0.3568 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14767 | train 3.055463 | val 3.332533 | tok/s 2622524.74 | norm 0.3760 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14768 | train 2.886306 | val 3.332533 | tok/s 2627552.20 | norm 0.3893 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14769 | train 2.972669 | val 3.332533 | tok/s 2630092.77 | norm 0.3868 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14770 | train 2.966423 | val 3.332533 | tok/s 2507329.84 | norm 0.3801 | dt 209.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14771 | train 2.991243 | val 3.332533 | tok/s 2552904.91 | norm 0.3846 | dt 205.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14772 | train 2.877617 | val 3.332533 | tok/s 2626654.20 | norm 0.3871 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14773 | train 2.939357 | val 3.332533 | tok/s 2629011.78 | norm 0.3800 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14774 | train 3.174331 | val 3.332533 | tok/s 2629029.81 | norm 0.3796 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14775 | train 2.938565 | val 3.332533 | tok/s 2628028.45 | norm 0.4015 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14776 | train 3.034362 | val 3.332533 | tok/s 2628190.97 | norm 0.4019 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14777 | train 3.068552 | val 3.332533 | tok/s 2627011.33 | norm 0.3940 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14778 | train 2.847788 | val 3.332533 | tok/s 2625497.93 | norm 0.3922 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14779 | train 2.739974 | val 3.332533 | tok/s 2626667.24 | norm 0.4002 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14780 | train 2.903686 | val 3.332533 | tok/s 2622841.62 | norm 0.3960 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14781 | train 2.967514 | val 3.332533 | tok/s 2620549.37 | norm 0.3635 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14782 | train 2.941911 | val 3.332533 | tok/s 2628319.85 | norm 0.3602 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14783 | train 2.955354 | val 3.332533 | tok/s 2626456.57 | norm 0.3840 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14784 | train 3.080173 | val 3.332533 | tok/s 2625269.70 | norm 0.4107 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14785 | train 3.052970 | val 3.332533 | tok/s 2612460.92 | norm 0.4163 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14786 | train 2.963656 | val 3.332533 | tok/s 2627600.75 | norm 0.3448 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14787 | train 3.007640 | val 3.332533 | tok/s 2621428.68 | norm 0.3790 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14788 | train 2.795519 | val 3.332533 | tok/s 2624241.13 | norm 0.3980 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14789 | train 2.940437 | val 3.332533 | tok/s 2628983.61 | norm 0.3979 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14790 | train 2.942427 | val 3.332533 | tok/s 2624504.27 | norm 0.4334 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14791 | train 3.150219 | val 3.332533 | tok/s 2624980.76 | norm 0.3981 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14792 | train 2.962855 | val 3.332533 | tok/s 2622766.99 | norm 0.3892 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14793 | train 3.058302 | val 3.332533 | tok/s 2628056.07 | norm 0.3747 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14794 | train 2.913388 | val 3.332533 | tok/s 2624067.43 | norm 0.3413 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14795 | train 2.981097 | val 3.332533 | tok/s 2624409.94 | norm 0.3314 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14796 | train 3.172517 | val 3.332533 | tok/s 2626933.05 | norm 0.4230 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14797 | train 3.043897 | val 3.332533 | tok/s 2624418.91 | norm 0.4091 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14798 | train 2.963427 | val 3.332533 | tok/s 2626582.22 | norm 0.3600 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14799 | train 3.090880 | val 3.332533 | tok/s 2623886.42 | norm 0.4799 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14800 | train 2.935488 | val 3.332533 | tok/s 2625567.69 | norm 0.4446 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14801 | train 2.958143 | val 3.332533 | tok/s 2626306.55 | norm 0.3326 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14802 | train 2.887557 | val 3.332533 | tok/s 2626561.06 | norm 0.3572 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14803 | train 2.982098 | val 3.332533 | tok/s 2627625.71 | norm 0.3468 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14804 | train 3.055671 | val 3.332533 | tok/s 2621882.40 | norm 0.3840 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14805 | train 2.962067 | val 3.332533 | tok/s 2553521.47 | norm 0.3836 | dt 205.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14806 | train 2.933196 | val 3.332533 | tok/s 2627865.01 | norm 0.3460 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14807 | train 2.840832 | val 3.332533 | tok/s 2626277.74 | norm 0.3801 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14808 | train 2.993593 | val 3.332533 | tok/s 2627030.14 | norm 0.3942 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14809 | train 3.016440 | val 3.332533 | tok/s 2620256.36 | norm 0.3672 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14810 | train 3.325864 | val 3.332533 | tok/s 2624828.29 | norm 0.5368 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14811 | train 2.985128 | val 3.332533 | tok/s 2626935.39 | norm 0.4235 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14812 | train 3.109989 | val 3.332533 | tok/s 2629957.47 | norm 0.3802 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14813 | train 2.904307 | val 3.332533 | tok/s 2626417.56 | norm 0.4152 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14814 | train 2.991647 | val 3.332533 | tok/s 2629996.87 | norm 0.4946 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14815 | train 2.862573 | val 3.332533 | tok/s 2627921.73 | norm 0.3722 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14816 | train 2.985655 | val 3.332533 | tok/s 2627814.89 | norm 0.3904 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14817 | train 3.002020 | val 3.332533 | tok/s 2623996.15 | norm 0.3895 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14818 | train 2.939194 | val 3.332533 | tok/s 2628808.67 | norm 0.3560 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14819 | train 2.935900 | val 3.332533 | tok/s 2473516.44 | norm 0.3877 | dt 211.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14820 | train 2.859992 | val 3.332533 | tok/s 2622200.21 | norm 0.3803 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14821 | train 3.030281 | val 3.332533 | tok/s 2629700.23 | norm 0.3792 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14822 | train 2.945072 | val 3.332533 | tok/s 2625938.73 | norm 0.4292 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14823 | train 2.985860 | val 3.332533 | tok/s 2626032.98 | norm 0.3992 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14824 | train 2.901559 | val 3.332533 | tok/s 2623760.64 | norm 0.3487 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14825 | train 3.235801 | val 3.332533 | tok/s 2617845.45 | norm 0.4049 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14826 | train 3.050827 | val 3.332533 | tok/s 2629643.04 | norm 0.3540 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14827 | train 2.972380 | val 3.332533 | tok/s 2629381.14 | norm 0.3489 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14828 | train 3.053099 | val 3.332533 | tok/s 2624882.09 | norm 0.4140 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14829 | train 2.938840 | val 3.332533 | tok/s 2619145.69 | norm 0.3852 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14830 | train 3.026979 | val 3.332533 | tok/s 2625260.62 | norm 0.3612 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14831 | train 2.820992 | val 3.332533 | tok/s 2625953.65 | norm 0.3828 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14832 | train 2.988436 | val 3.332533 | tok/s 2625085.53 | norm 0.4036 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14833 | train 2.777146 | val 3.332533 | tok/s 2629053.41 | norm 0.4047 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14834 | train 2.954850 | val 3.332533 | tok/s 2625473.98 | norm 0.4247 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14835 | train 2.983047 | val 3.332533 | tok/s 2627276.01 | norm 0.4015 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14836 | train 2.982249 | val 3.332533 | tok/s 2630541.71 | norm 0.3951 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14837 | train 3.145115 | val 3.332533 | tok/s 2623742.65 | norm 0.3839 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14838 | train 3.019812 | val 3.332533 | tok/s 2622985.91 | norm 0.3809 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14839 | train 2.899303 | val 3.332533 | tok/s 2624223.54 | norm 0.3809 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14840 | train 2.959986 | val 3.332533 | tok/s 2627206.23 | norm 0.4127 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14841 | train 2.995664 | val 3.332533 | tok/s 2626833.45 | norm 0.4261 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14842 | train 3.016678 | val 3.332533 | tok/s 2623976.47 | norm 0.4528 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14843 | train 2.908439 | val 3.332533 | tok/s 2629751.05 | norm 0.3701 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14844 | train 2.977785 | val 3.332533 | tok/s 2624622.64 | norm 0.3613 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14845 | train 3.034916 | val 3.332533 | tok/s 2627106.30 | norm 0.3741 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14846 | train 2.938009 | val 3.332533 | tok/s 2621020.40 | norm 0.3759 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14847 | train 3.033976 | val 3.332533 | tok/s 2602574.10 | norm 0.3648 | dt 201.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14848 | train 2.896078 | val 3.332533 | tok/s 2619345.47 | norm 0.3722 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14849 | train 3.151496 | val 3.332533 | tok/s 2627354.86 | norm 0.4210 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14850 | train 2.929974 | val 3.332533 | tok/s 2627504.31 | norm 0.3292 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14851 | train 3.097720 | val 3.332533 | tok/s 2622601.80 | norm 0.3109 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14852 | train 3.015223 | val 3.332533 | tok/s 2622578.28 | norm 0.3605 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14853 | train 2.948377 | val 3.332533 | tok/s 2625402.70 | norm 0.3700 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14854 | train 2.821449 | val 3.332533 | tok/s 2626821.02 | norm 0.4058 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14855 | train 2.878476 | val 3.332533 | tok/s 2597838.12 | norm 0.4182 | dt 201.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14856 | train 2.950371 | val 3.332533 | tok/s 2544630.11 | norm 0.3967 | dt 206.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14857 | train 3.012873 | val 3.332533 | tok/s 2625540.23 | norm 0.3886 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14858 | train 2.981414 | val 3.332533 | tok/s 2625829.13 | norm 0.4016 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14859 | train 2.986715 | val 3.332533 | tok/s 2630042.45 | norm 0.3867 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14860 | train 2.896663 | val 3.332533 | tok/s 2626233.82 | norm 0.4364 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14861 | train 2.979511 | val 3.332533 | tok/s 2622542.62 | norm 0.3520 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14862 | train 2.942719 | val 3.332533 | tok/s 2621625.20 | norm 0.3622 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14863 | train 2.853852 | val 3.332533 | tok/s 2624564.20 | norm 0.4060 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14864 | train 3.075158 | val 3.332533 | tok/s 2574538.36 | norm 0.3505 | dt 203.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14865 | train 3.034154 | val 3.332533 | tok/s 2627056.19 | norm 0.3670 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14866 | train 2.935935 | val 3.332533 | tok/s 2616773.98 | norm 0.3949 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14867 | train 3.006050 | val 3.332533 | tok/s 2609294.97 | norm 0.3934 | dt 200.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14868 | train 3.042436 | val 3.332533 | tok/s 2623692.94 | norm 0.4216 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14869 | train 2.930034 | val 3.332533 | tok/s 2518464.68 | norm 0.4137 | dt 208.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14870 | train 2.956067 | val 3.332533 | tok/s 2579344.26 | norm 0.4040 | dt 203.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14871 | train 2.993454 | val 3.332533 | tok/s 2627801.78 | norm 0.4070 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14872 | train 3.080019 | val 3.332533 | tok/s 2625194.28 | norm 0.3630 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14873 | train 3.007404 | val 3.332533 | tok/s 2626555.15 | norm 0.3471 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14874 | train 2.998540 | val 3.332533 | tok/s 2623273.45 | norm 0.3843 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14875 | train 2.960473 | val 3.332533 | tok/s 2630339.98 | norm 0.4075 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14876 | train 2.987253 | val 3.332533 | tok/s 2619649.89 | norm 0.3540 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14877 | train 3.030833 | val 3.332533 | tok/s 2613039.94 | norm 0.3693 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14878 | train 2.966380 | val 3.332533 | tok/s 2343594.54 | norm 0.3718 | dt 223.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14879 | train 2.877751 | val 3.332533 | tok/s 2618991.43 | norm 0.3465 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14880 | train 2.908142 | val 3.332533 | tok/s 2617147.66 | norm 0.3912 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14881 | train 3.297339 | val 3.332533 | tok/s 2621472.94 | norm 0.4449 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14882 | train 2.936801 | val 3.332533 | tok/s 2614340.73 | norm 0.4379 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14883 | train 2.927238 | val 3.332533 | tok/s 2616199.93 | norm 0.3923 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14884 | train 3.024913 | val 3.332533 | tok/s 2621329.04 | norm 0.3836 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14885 | train 2.956504 | val 3.332533 | tok/s 2623161.88 | norm 0.4044 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14886 | train 3.005426 | val 3.332533 | tok/s 2620712.46 | norm 0.4229 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14887 | train 3.020046 | val 3.332533 | tok/s 2621069.34 | norm 0.4067 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14888 | train 3.020164 | val 3.332533 | tok/s 2611438.97 | norm 0.3968 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14889 | train 3.247969 | val 3.332533 | tok/s 2622778.02 | norm 0.4012 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14890 | train 2.920332 | val 3.332533 | tok/s 2624152.51 | norm 0.3707 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14891 | train 2.994791 | val 3.332533 | tok/s 2619901.92 | norm 0.3901 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14892 | train 2.927636 | val 3.332533 | tok/s 2622242.38 | norm 0.4140 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14893 | train 3.036333 | val 3.332533 | tok/s 2619801.00 | norm 0.3856 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14894 | train 3.088996 | val 3.332533 | tok/s 2614456.55 | norm 0.4313 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14895 | train 3.015099 | val 3.332533 | tok/s 2626572.22 | norm 0.3674 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14896 | train 2.965615 | val 3.332533 | tok/s 2620490.64 | norm 0.3656 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14897 | train 2.960477 | val 3.332533 | tok/s 2625619.80 | norm 0.4162 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14898 | train 2.929038 | val 3.332533 | tok/s 2616982.51 | norm 0.4334 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14899 | train 2.985648 | val 3.332533 | tok/s 2607800.33 | norm 0.4122 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14900 | train 3.403707 | val 3.332533 | tok/s 2623503.39 | norm 0.6250 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14901 | train 2.853164 | val 3.332533 | tok/s 2625125.60 | norm 0.4022 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14902 | train 3.016942 | val 3.332533 | tok/s 2626713.85 | norm 0.3886 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14903 | train 3.041529 | val 3.332533 | tok/s 2474567.88 | norm 0.4245 | dt 211.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14904 | train 2.934103 | val 3.332533 | tok/s 2623315.84 | norm 0.3907 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14905 | train 2.931507 | val 3.332533 | tok/s 2626691.12 | norm 0.3900 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14906 | train 2.938376 | val 3.332533 | tok/s 2625099.51 | norm 0.4013 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14907 | train 3.037942 | val 3.332533 | tok/s 2627536.48 | norm 0.3373 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14908 | train 2.956944 | val 3.332533 | tok/s 2625485.22 | norm 0.3557 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14909 | train 2.967398 | val 3.332533 | tok/s 2627202.95 | norm 0.3949 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14910 | train 2.973782 | val 3.332533 | tok/s 2624365.34 | norm 0.3804 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14911 | train 3.036426 | val 3.332533 | tok/s 2625673.12 | norm 0.3597 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14912 | train 2.947212 | val 3.332533 | tok/s 2555890.28 | norm 0.4073 | dt 205.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14913 | train 2.893558 | val 3.332533 | tok/s 2624366.90 | norm 0.4007 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14914 | train 2.884616 | val 3.332533 | tok/s 2623035.76 | norm 0.3432 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14915 | train 2.922230 | val 3.332533 | tok/s 2624064.05 | norm 0.3743 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14916 | train 2.963069 | val 3.332533 | tok/s 2617411.01 | norm 0.3897 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14917 | train 3.071363 | val 3.332533 | tok/s 2631500.20 | norm 0.3725 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14918 | train 2.929666 | val 3.332533 | tok/s 2628490.14 | norm 0.3756 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14919 | train 3.046326 | val 3.332533 | tok/s 2626808.92 | norm 0.4166 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14920 | train 3.074894 | val 3.332533 | tok/s 2628048.08 | norm 0.4319 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14921 | train 3.126700 | val 3.332533 | tok/s 2621912.38 | norm 0.4040 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14922 | train 2.990127 | val 3.332533 | tok/s 2627434.24 | norm 0.3935 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14923 | train 2.948154 | val 3.332533 | tok/s 2626578.51 | norm 0.3795 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14924 | train 2.978476 | val 3.332533 | tok/s 2626082.47 | norm 0.3856 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14925 | train 2.978430 | val 3.332533 | tok/s 2626830.95 | norm 0.3726 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14926 | train 2.893127 | val 3.332533 | tok/s 2622445.22 | norm 0.4017 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14927 | train 3.008174 | val 3.332533 | tok/s 2626941.64 | norm 0.3743 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14928 | train 2.920029 | val 3.332533 | tok/s 2619913.04 | norm 0.3338 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14929 | train 2.844254 | val 3.332533 | tok/s 2627015.22 | norm 0.3437 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14930 | train 2.952668 | val 3.332533 | tok/s 2621816.10 | norm 0.4067 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14931 | train 2.914748 | val 3.332533 | tok/s 2625897.47 | norm 0.3968 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14932 | train 3.012353 | val 3.332533 | tok/s 2624914.36 | norm 0.3706 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14933 | train 2.994140 | val 3.332533 | tok/s 2622001.49 | norm 0.3457 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14934 | train 2.874958 | val 3.332533 | tok/s 2575288.78 | norm 0.3577 | dt 203.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14935 | train 2.893803 | val 3.332533 | tok/s 2616830.85 | norm 0.3987 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14936 | train 2.946650 | val 3.332533 | tok/s 2577868.53 | norm 0.3735 | dt 203.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14937 | train 2.935171 | val 3.332533 | tok/s 2622482.08 | norm 0.3601 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14938 | train 2.944474 | val 3.332533 | tok/s 2625250.87 | norm 0.4379 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14939 | train 3.118902 | val 3.332533 | tok/s 2622454.39 | norm 0.4259 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14940 | train 2.988368 | val 3.332533 | tok/s 2595052.98 | norm 0.3632 | dt 202.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14941 | train 2.942986 | val 3.332533 | tok/s 2619636.11 | norm 0.3942 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14942 | train 3.057549 | val 3.332533 | tok/s 2625613.10 | norm 0.4617 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14943 | train 2.914135 | val 3.332533 | tok/s 2625500.47 | norm 0.3875 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14944 | train 3.029140 | val 3.332533 | tok/s 2488304.29 | norm 0.3737 | dt 210.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14945 | train 3.088676 | val 3.332533 | tok/s 2611207.30 | norm 0.3857 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14946 | train 2.985255 | val 3.332533 | tok/s 2612224.85 | norm 0.3521 | dt 200.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14947 | train 2.908931 | val 3.332533 | tok/s 2624441.38 | norm 0.3775 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14948 | train 2.973250 | val 3.332533 | tok/s 2628364.20 | norm 0.3674 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14949 | train 2.989601 | val 3.332533 | tok/s 2626972.13 | norm 0.3434 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14950 | train 2.988775 | val 3.332533 | tok/s 2624974.15 | norm 0.3720 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14951 | train 2.993948 | val 3.332533 | tok/s 2605983.66 | norm 0.3923 | dt 201.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14952 | train 3.031372 | val 3.332533 | tok/s 2628369.53 | norm 0.3710 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14953 | train 2.964669 | val 3.332533 | tok/s 2626793.41 | norm 0.3715 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14954 | train 3.030995 | val 3.332533 | tok/s 2629689.76 | norm 0.3692 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14955 | train 2.943501 | val 3.332533 | tok/s 2625013.38 | norm 0.3874 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14956 | train 2.980073 | val 3.332533 | tok/s 2625401.95 | norm 0.4210 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14957 | train 2.950354 | val 3.332533 | tok/s 2626692.82 | norm 0.3717 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14958 | train 2.955285 | val 3.332533 | tok/s 2627733.28 | norm 0.3636 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14959 | train 3.024599 | val 3.332533 | tok/s 2611993.10 | norm 0.3916 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14960 | train 2.979455 | val 3.332533 | tok/s 2622792.99 | norm 0.3932 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14961 | train 2.918256 | val 3.332533 | tok/s 2625625.55 | norm 0.3678 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14962 | train 3.017310 | val 3.332533 | tok/s 2628412.92 | norm 0.3822 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14963 | train 3.048233 | val 3.332533 | tok/s 2626398.38 | norm 0.5519 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14964 | train 2.977685 | val 3.332533 | tok/s 2628524.85 | norm 0.3897 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14965 | train 3.182775 | val 3.332533 | tok/s 2628793.96 | norm 0.3441 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14966 | train 3.012224 | val 3.332533 | tok/s 2628799.31 | norm 0.3414 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14967 | train 2.893339 | val 3.332533 | tok/s 2626655.57 | norm 0.3457 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14968 | train 2.891869 | val 3.332533 | tok/s 2622369.38 | norm 0.3696 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14969 | train 2.904435 | val 3.332533 | tok/s 2627900.92 | norm 0.4236 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14970 | train 3.115156 | val 3.332533 | tok/s 2629791.22 | norm 0.4022 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14971 | train 2.992902 | val 3.332533 | tok/s 2626199.87 | norm 0.3731 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14972 | train 2.913925 | val 3.332533 | tok/s 2629289.10 | norm 0.3664 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14973 | train 2.911878 | val 3.332533 | tok/s 2622262.52 | norm 0.3460 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14974 | train 3.189762 | val 3.332533 | tok/s 2626829.93 | norm 0.4167 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14975 | train 3.018408 | val 3.332533 | tok/s 2625710.56 | norm 0.3894 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14976 | train 2.862838 | val 3.332533 | tok/s 2629577.17 | norm 0.3502 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14977 | train 3.027108 | val 3.332533 | tok/s 2627189.20 | norm 0.4057 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14978 | train 3.004205 | val 3.332533 | tok/s 2626137.23 | norm 0.4229 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14979 | train 3.116047 | val 3.332533 | tok/s 2626165.23 | norm 0.3889 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14980 | train 2.978605 | val 3.332533 | tok/s 2626428.53 | norm 0.3762 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14981 | train 2.965304 | val 3.332533 | tok/s 2620716.91 | norm 0.3849 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14982 | train 3.064359 | val 3.332533 | tok/s 2624426.08 | norm 0.4006 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14983 | train 2.990880 | val 3.332533 | tok/s 2627930.79 | norm 0.3978 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14984 | train 2.910049 | val 3.332533 | tok/s 2628332.47 | norm 0.3784 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14985 | train 2.948339 | val 3.332533 | tok/s 2622067.05 | norm 0.4132 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14986 | train 2.980227 | val 3.332533 | tok/s 2624083.44 | norm 0.4092 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14987 | train 2.863916 | val 3.332533 | tok/s 2625225.29 | norm 0.3741 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14988 | train 2.995068 | val 3.332533 | tok/s 2624767.59 | norm 0.3762 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14989 | train 2.933996 | val 3.332533 | tok/s 2630421.59 | norm 0.3762 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14990 | train 2.997417 | val 3.332533 | tok/s 2625869.71 | norm 0.3524 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14991 | train 2.782221 | val 3.332533 | tok/s 2626164.13 | norm 0.3727 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14992 | train 2.956376 | val 3.332533 | tok/s 2623816.77 | norm 0.3900 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14993 | train 2.898962 | val 3.332533 | tok/s 2625377.65 | norm 0.3738 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14994 | train 2.897821 | val 3.332533 | tok/s 2628613.47 | norm 0.3833 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14995 | train 3.054179 | val 3.332533 | tok/s 2624095.16 | norm 0.3907 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14996 | train 2.937264 | val 3.332533 | tok/s 2620795.44 | norm 0.3952 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14997 | train 2.920156 | val 3.332533 | tok/s 2621994.97 | norm 0.4013 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14998 | train 2.861425 | val 3.332533 | tok/s 2622889.61 | norm 0.4251 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 14999 | train 3.025464 | val 3.332533 | tok/s 2626973.22 | norm 0.4925 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 15000 | val_loss 3.330439 step 15000 | train 2.866756 | val 3.330439 | tok/s 2616876.70 | norm 0.4396 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15001 | train 3.044382 | val 3.330439 | tok/s 2625903.58 | norm 0.3619 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15002 | train 2.897851 | val 3.330439 | tok/s 2627889.68 | norm 0.3786 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15003 | train 2.980457 | val 3.330439 | tok/s 2622862.18 | norm 0.3929 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15004 | train 2.800397 | val 3.330439 | tok/s 2624617.36 | norm 0.3741 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15005 | train 2.926715 | val 3.330439 | tok/s 2631172.96 | norm 0.4083 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15006 | train 2.967343 | val 3.330439 | tok/s 2561811.54 | norm 0.3888 | dt 204.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15007 | train 2.913492 | val 3.330439 | tok/s 2585159.64 | norm 0.3775 | dt 202.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15008 | train 2.995934 | val 3.330439 | tok/s 2619949.79 | norm 0.3887 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15009 | train 2.891896 | val 3.330439 | tok/s 2624784.91 | norm 0.3847 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15010 | train 2.998480 | val 3.330439 | tok/s 2628033.92 | norm 0.4058 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15011 | train 2.811525 | val 3.330439 | tok/s 2623982.84 | norm 0.3850 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15012 | train 2.931824 | val 3.330439 | tok/s 2624164.34 | norm 0.3691 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15013 | train 2.886479 | val 3.330439 | tok/s 2623113.06 | norm 0.3849 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15014 | train 3.008430 | val 3.330439 | tok/s 2625528.46 | norm 0.4608 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15015 | train 2.899260 | val 3.330439 | tok/s 2631727.73 | norm 0.4557 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15016 | train 3.011031 | val 3.330439 | tok/s 2627581.74 | norm 0.4110 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15017 | train 2.981176 | val 3.330439 | tok/s 2626093.94 | norm 0.3864 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15018 | train 2.888043 | val 3.330439 | tok/s 2631513.35 | norm 0.4126 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15019 | train 2.942060 | val 3.330439 | tok/s 2627995.90 | norm 0.4004 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15020 | train 2.939116 | val 3.330439 | tok/s 2626495.69 | norm 0.3673 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15021 | train 2.993929 | val 3.330439 | tok/s 2629668.89 | norm 0.3740 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15022 | train 2.880434 | val 3.330439 | tok/s 2560198.55 | norm 0.3760 | dt 204.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15023 | train 2.966245 | val 3.330439 | tok/s 2624813.48 | norm 0.3735 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15024 | train 3.014182 | val 3.330439 | tok/s 2624651.42 | norm 0.3867 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15025 | train 2.955601 | val 3.330439 | tok/s 2629920.98 | norm 0.3785 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15026 | train 3.069101 | val 3.330439 | tok/s 2627222.10 | norm 0.3869 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15027 | train 2.916154 | val 3.330439 | tok/s 2623019.54 | norm 0.4136 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15028 | train 3.053794 | val 3.330439 | tok/s 2620803.39 | norm 0.4195 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15029 | train 2.978531 | val 3.330439 | tok/s 2626851.37 | norm 0.3993 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15030 | train 2.912847 | val 3.330439 | tok/s 2626556.99 | norm 0.3993 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15031 | train 3.045380 | val 3.330439 | tok/s 2625502.47 | norm 0.3991 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15032 | train 2.972074 | val 3.330439 | tok/s 2628981.73 | norm 0.3557 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15033 | train 2.998919 | val 3.330439 | tok/s 2623579.21 | norm 0.3360 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15034 | train 2.907017 | val 3.330439 | tok/s 2622250.94 | norm 0.3777 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15035 | train 2.963707 | val 3.330439 | tok/s 2625833.16 | norm 0.4083 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15036 | train 2.956208 | val 3.330439 | tok/s 2627035.19 | norm 0.4265 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15037 | train 2.985269 | val 3.330439 | tok/s 2628319.47 | norm 0.4397 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15038 | train 2.966581 | val 3.330439 | tok/s 2629096.02 | norm 0.3983 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15039 | train 3.025909 | val 3.330439 | tok/s 2625166.94 | norm 0.3776 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15040 | train 3.032433 | val 3.330439 | tok/s 2625552.49 | norm 0.3602 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15041 | train 3.108257 | val 3.330439 | tok/s 2623495.52 | norm 0.3843 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15042 | train 3.014804 | val 3.330439 | tok/s 2630998.18 | norm 0.3521 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15043 | train 2.974027 | val 3.330439 | tok/s 2625393.76 | norm 0.3493 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15044 | train 3.056838 | val 3.330439 | tok/s 2627455.20 | norm 0.3571 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15045 | train 2.976675 | val 3.330439 | tok/s 2623605.56 | norm 0.3730 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15046 | train 3.125489 | val 3.330439 | tok/s 2619000.92 | norm 0.4032 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15047 | train 3.116002 | val 3.330439 | tok/s 2625511.16 | norm 0.4350 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15048 | train 3.034697 | val 3.330439 | tok/s 2625254.06 | norm 0.3373 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15049 | train 2.922453 | val 3.330439 | tok/s 2626237.03 | norm 0.3401 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15050 | train 3.075100 | val 3.330439 | tok/s 2617676.98 | norm 0.4216 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15051 | train 2.961529 | val 3.330439 | tok/s 2622324.39 | norm 0.4333 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15052 | train 2.961271 | val 3.330439 | tok/s 2627598.18 | norm 0.3715 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15053 | train 2.852813 | val 3.330439 | tok/s 2627573.74 | norm 0.3471 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15054 | train 2.985663 | val 3.330439 | tok/s 2626578.53 | norm 0.3755 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15055 | train 3.047205 | val 3.330439 | tok/s 2619307.61 | norm 0.3868 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15056 | train 2.807451 | val 3.330439 | tok/s 2622332.16 | norm 0.3806 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15057 | train 2.990188 | val 3.330439 | tok/s 2627355.42 | norm 0.4333 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15058 | train 2.953467 | val 3.330439 | tok/s 2623813.55 | norm 0.4164 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15059 | train 3.023104 | val 3.330439 | tok/s 2621374.36 | norm 0.4014 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15060 | train 2.985275 | val 3.330439 | tok/s 2542173.53 | norm 0.3620 | dt 206.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15061 | train 2.950725 | val 3.330439 | tok/s 2626855.89 | norm 0.3333 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15062 | train 2.988705 | val 3.330439 | tok/s 2629057.18 | norm 0.3698 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15063 | train 2.970611 | val 3.330439 | tok/s 2622335.17 | norm 0.3647 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15064 | train 2.933956 | val 3.330439 | tok/s 2625608.49 | norm 0.3102 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15065 | train 2.950993 | val 3.330439 | tok/s 2614698.13 | norm 0.3240 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15066 | train 3.008028 | val 3.330439 | tok/s 2621835.54 | norm 0.3554 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15067 | train 2.984256 | val 3.330439 | tok/s 2628157.74 | norm 0.3423 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15068 | train 3.054292 | val 3.330439 | tok/s 2490442.48 | norm 0.3484 | dt 210.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15069 | train 2.979914 | val 3.330439 | tok/s 2627803.54 | norm 0.3660 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15070 | train 3.100117 | val 3.330439 | tok/s 2618433.78 | norm 0.3591 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15071 | train 3.049501 | val 3.330439 | tok/s 2623114.85 | norm 0.3817 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15072 | train 2.999458 | val 3.330439 | tok/s 2622670.35 | norm 0.3823 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15073 | train 2.980462 | val 3.330439 | tok/s 2620386.60 | norm 0.4085 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15074 | train 2.906763 | val 3.330439 | tok/s 2625114.95 | norm 0.3872 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15075 | train 3.067935 | val 3.330439 | tok/s 2620307.49 | norm 0.3826 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15076 | train 3.031780 | val 3.330439 | tok/s 2624683.02 | norm 0.4133 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15077 | train 2.851284 | val 3.330439 | tok/s 2626397.31 | norm 0.4195 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15078 | train 2.853500 | val 3.330439 | tok/s 2624630.57 | norm 0.3864 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15079 | train 2.892527 | val 3.330439 | tok/s 2622068.22 | norm 0.3902 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15080 | train 3.008566 | val 3.330439 | tok/s 2621257.40 | norm 0.3892 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15081 | train 2.868763 | val 3.330439 | tok/s 2622307.54 | norm 0.3910 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15082 | train 2.959623 | val 3.330439 | tok/s 2619186.36 | norm 0.4070 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15083 | train 2.955505 | val 3.330439 | tok/s 2621962.91 | norm 0.3970 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15084 | train 2.920712 | val 3.330439 | tok/s 2627892.58 | norm 0.3926 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15085 | train 2.764451 | val 3.330439 | tok/s 2624761.32 | norm 0.3640 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15086 | train 2.847105 | val 3.330439 | tok/s 2630446.06 | norm 0.3700 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15087 | train 2.952773 | val 3.330439 | tok/s 2627887.68 | norm 0.3952 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15088 | train 2.879073 | val 3.330439 | tok/s 2623097.02 | norm 0.3914 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15089 | train 3.021568 | val 3.330439 | tok/s 2626851.47 | norm 0.3836 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15090 | train 2.912834 | val 3.330439 | tok/s 2625425.38 | norm 0.3901 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15091 | train 3.052618 | val 3.330439 | tok/s 2627914.42 | norm 0.4102 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15092 | train 3.068676 | val 3.330439 | tok/s 2627427.61 | norm 0.4133 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15093 | train 3.018427 | val 3.330439 | tok/s 2604014.33 | norm 0.3932 | dt 201.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15094 | train 2.907568 | val 3.330439 | tok/s 2627941.32 | norm 0.3604 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15095 | train 2.846674 | val 3.330439 | tok/s 2620940.91 | norm 0.3635 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15096 | train 2.969086 | val 3.330439 | tok/s 2629411.47 | norm 0.3856 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15097 | train 2.889585 | val 3.330439 | tok/s 2623900.59 | norm 0.3715 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15098 | train 2.950167 | val 3.330439 | tok/s 2624565.81 | norm 0.3654 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15099 | train 2.962993 | val 3.330439 | tok/s 2623369.80 | norm 0.3866 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15100 | train 2.966445 | val 3.330439 | tok/s 2626638.13 | norm 0.4160 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15101 | train 2.934883 | val 3.330439 | tok/s 2626226.82 | norm 0.4151 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15102 | train 3.039456 | val 3.330439 | tok/s 2622883.22 | norm 0.4009 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15103 | train 2.921008 | val 3.330439 | tok/s 2595104.03 | norm 0.3804 | dt 202.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15104 | train 3.219786 | val 3.330439 | tok/s 2623360.86 | norm 0.4637 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15105 | train 2.964174 | val 3.330439 | tok/s 2627221.62 | norm 0.3942 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15106 | train 2.905739 | val 3.330439 | tok/s 2630956.68 | norm 0.3311 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15107 | train 2.991014 | val 3.330439 | tok/s 2608913.41 | norm 0.3671 | dt 200.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15108 | train 2.898554 | val 3.330439 | tok/s 2626918.69 | norm 0.3712 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15109 | train 2.974595 | val 3.330439 | tok/s 2629199.66 | norm 0.3693 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15110 | train 2.982672 | val 3.330439 | tok/s 2626507.64 | norm 0.3774 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15111 | train 2.976362 | val 3.330439 | tok/s 2624842.95 | norm 0.3882 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15112 | train 2.917147 | val 3.330439 | tok/s 2622843.77 | norm 0.4111 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15113 | train 2.939331 | val 3.330439 | tok/s 2619428.00 | norm 0.3988 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15114 | train 3.059106 | val 3.330439 | tok/s 2633437.49 | norm 0.4811 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15115 | train 2.932635 | val 3.330439 | tok/s 2630175.63 | norm 0.3904 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15116 | train 2.951961 | val 3.330439 | tok/s 2624651.85 | norm 0.3697 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15117 | train 2.960817 | val 3.330439 | tok/s 2626347.46 | norm 0.3589 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15118 | train 2.908835 | val 3.330439 | tok/s 2627570.48 | norm 0.3610 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15119 | train 2.954751 | val 3.330439 | tok/s 2627851.76 | norm 0.3804 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15120 | train 3.001615 | val 3.330439 | tok/s 2626775.37 | norm 0.3643 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15121 | train 2.953218 | val 3.330439 | tok/s 2632301.03 | norm 0.3673 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15122 | train 2.913446 | val 3.330439 | tok/s 2626393.14 | norm 0.3942 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15123 | train 3.230654 | val 3.330439 | tok/s 2604460.54 | norm 0.4393 | dt 201.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15124 | train 3.118178 | val 3.330439 | tok/s 2629549.12 | norm 0.4036 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15125 | train 2.987118 | val 3.330439 | tok/s 2626421.38 | norm 0.3592 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15126 | train 2.971805 | val 3.330439 | tok/s 2630512.08 | norm 0.3703 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15127 | train 3.099702 | val 3.330439 | tok/s 2624199.33 | norm 0.3987 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15128 | train 3.012622 | val 3.330439 | tok/s 2627942.52 | norm 0.4140 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15129 | train 2.958110 | val 3.330439 | tok/s 2628670.08 | norm 0.3570 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15130 | train 2.962332 | val 3.330439 | tok/s 2628481.74 | norm 0.3820 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15131 | train 2.894700 | val 3.330439 | tok/s 2630858.10 | norm 0.3995 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15132 | train 3.064066 | val 3.330439 | tok/s 2626552.11 | norm 0.3837 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15133 | train 2.958143 | val 3.330439 | tok/s 2623410.94 | norm 0.3671 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15134 | train 2.992726 | val 3.330439 | tok/s 2628282.44 | norm 0.3706 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15135 | train 2.968404 | val 3.330439 | tok/s 2625775.72 | norm 0.3782 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15136 | train 2.949936 | val 3.330439 | tok/s 2626691.03 | norm 0.3378 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15137 | train 2.955481 | val 3.330439 | tok/s 2624042.91 | norm 0.3628 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15138 | train 2.934282 | val 3.330439 | tok/s 2622425.69 | norm 0.3891 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15139 | train 2.829158 | val 3.330439 | tok/s 2627036.90 | norm 0.3891 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15140 | train 2.896955 | val 3.330439 | tok/s 2625366.30 | norm 0.4024 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15141 | train 2.981113 | val 3.330439 | tok/s 2625253.68 | norm 0.3871 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15142 | train 3.034724 | val 3.330439 | tok/s 2624767.38 | norm 0.3866 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15143 | train 2.918309 | val 3.330439 | tok/s 2623496.10 | norm 0.3696 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15144 | train 2.916424 | val 3.330439 | tok/s 2622343.25 | norm 0.3530 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15145 | train 3.062266 | val 3.330439 | tok/s 2616243.51 | norm 0.3631 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15146 | train 2.971882 | val 3.330439 | tok/s 2627464.29 | norm 0.3841 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15147 | train 2.939360 | val 3.330439 | tok/s 2624386.28 | norm 0.3848 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15148 | train 2.945333 | val 3.330439 | tok/s 2628419.47 | norm 0.4136 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15149 | train 2.859702 | val 3.330439 | tok/s 2627704.47 | norm 0.4168 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15150 | train 2.972776 | val 3.330439 | tok/s 2625178.52 | norm 0.3989 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15151 | train 3.043563 | val 3.330439 | tok/s 2629048.53 | norm 0.4398 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15152 | train 3.002185 | val 3.330439 | tok/s 2629372.95 | norm 0.4028 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15153 | train 2.929327 | val 3.330439 | tok/s 2626795.01 | norm 0.3832 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15154 | train 2.919365 | val 3.330439 | tok/s 2620736.18 | norm 0.3901 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15155 | train 2.943533 | val 3.330439 | tok/s 2624268.06 | norm 0.3744 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15156 | train 3.043931 | val 3.330439 | tok/s 2627151.81 | norm 0.3719 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15157 | train 2.945368 | val 3.330439 | tok/s 2626589.35 | norm 0.3813 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15158 | train 3.021759 | val 3.330439 | tok/s 2628429.02 | norm 0.3800 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15159 | train 2.809855 | val 3.330439 | tok/s 2625700.08 | norm 0.3567 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15160 | train 2.810124 | val 3.330439 | tok/s 2623651.58 | norm 0.3477 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15161 | train 2.904470 | val 3.330439 | tok/s 2628241.15 | norm 0.3493 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15162 | train 2.913971 | val 3.330439 | tok/s 2626273.26 | norm 0.3971 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15163 | train 2.958795 | val 3.330439 | tok/s 2627824.92 | norm 0.4059 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15164 | train 2.995301 | val 3.330439 | tok/s 2617894.93 | norm 0.3972 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15165 | train 3.048370 | val 3.330439 | tok/s 2624351.97 | norm 0.3712 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15166 | train 2.975159 | val 3.330439 | tok/s 2626536.83 | norm 0.3847 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15167 | train 2.876214 | val 3.330439 | tok/s 2624771.91 | norm 0.3645 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15168 | train 2.981453 | val 3.330439 | tok/s 2627437.80 | norm 0.3820 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15169 | train 3.086550 | val 3.330439 | tok/s 2624413.26 | norm 0.4099 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15170 | train 2.947923 | val 3.330439 | tok/s 2627156.86 | norm 0.3586 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15171 | train 3.076723 | val 3.330439 | tok/s 2630132.54 | norm 0.3502 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15172 | train 3.098261 | val 3.330439 | tok/s 2627592.98 | norm 0.3314 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15173 | train 2.942316 | val 3.330439 | tok/s 2626378.18 | norm 0.3356 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15174 | train 3.070831 | val 3.330439 | tok/s 2627150.49 | norm 0.3622 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15175 | train 3.002460 | val 3.330439 | tok/s 2624124.10 | norm 0.4014 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15176 | train 2.907931 | val 3.330439 | tok/s 2627265.82 | norm 0.3938 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15177 | train 2.973498 | val 3.330439 | tok/s 2624000.83 | norm 0.3860 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15178 | train 2.982271 | val 3.330439 | tok/s 2629064.54 | norm 0.3862 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15179 | train 2.964540 | val 3.330439 | tok/s 2628872.56 | norm 0.3523 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15180 | train 3.126727 | val 3.330439 | tok/s 2627084.17 | norm 0.3965 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15181 | train 2.907269 | val 3.330439 | tok/s 2629500.44 | norm 0.4112 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15182 | train 3.227845 | val 3.330439 | tok/s 2623194.38 | norm 0.4982 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15183 | train 2.890016 | val 3.330439 | tok/s 2626600.61 | norm 0.3565 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15184 | train 3.035053 | val 3.330439 | tok/s 2627817.34 | norm 0.3418 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15185 | train 3.026073 | val 3.330439 | tok/s 2628679.42 | norm 0.3960 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15186 | train 3.035411 | val 3.330439 | tok/s 2627934.02 | norm 0.3750 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15187 | train 2.975279 | val 3.330439 | tok/s 2629895.72 | norm 0.3286 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15188 | train 3.135491 | val 3.330439 | tok/s 2627176.45 | norm 0.3900 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15189 | train 2.984208 | val 3.330439 | tok/s 2626783.38 | norm 0.4092 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15190 | train 2.937978 | val 3.330439 | tok/s 2629055.15 | norm 0.3642 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15191 | train 2.963374 | val 3.330439 | tok/s 2631512.42 | norm 0.3750 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15192 | train 3.079112 | val 3.330439 | tok/s 2619684.46 | norm 0.4275 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15193 | train 2.931442 | val 3.330439 | tok/s 2627406.95 | norm 0.3817 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15194 | train 2.855181 | val 3.330439 | tok/s 2623417.41 | norm 0.3550 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15195 | train 2.959901 | val 3.330439 | tok/s 2628017.98 | norm 0.3783 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15196 | train 2.927647 | val 3.330439 | tok/s 2627802.38 | norm 0.4078 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15197 | train 2.891072 | val 3.330439 | tok/s 2626613.98 | norm 0.3957 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15198 | train 2.990392 | val 3.330439 | tok/s 2627026.58 | norm 0.3582 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15199 | train 2.926212 | val 3.330439 | tok/s 2622957.79 | norm 0.3573 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15200 | train 2.922246 | val 3.330439 | tok/s 2624148.11 | norm 0.4001 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15201 | train 3.094687 | val 3.330439 | tok/s 2628471.67 | norm 0.4002 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15202 | train 2.834477 | val 3.330439 | tok/s 2621836.04 | norm 0.3895 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15203 | train 2.960162 | val 3.330439 | tok/s 2627380.22 | norm 0.3961 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15204 | train 2.928879 | val 3.330439 | tok/s 2623388.89 | norm 0.4311 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15205 | train 2.859637 | val 3.330439 | tok/s 2624792.17 | norm 0.4092 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15206 | train 3.025914 | val 3.330439 | tok/s 2625042.44 | norm 0.3771 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15207 | train 3.046902 | val 3.330439 | tok/s 2623163.31 | norm 0.3593 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15208 | train 3.073736 | val 3.330439 | tok/s 2629354.86 | norm 0.4108 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15209 | train 3.009037 | val 3.330439 | tok/s 2622299.92 | norm 0.3718 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15210 | train 2.985314 | val 3.330439 | tok/s 2622903.82 | norm 0.3642 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15211 | train 2.990586 | val 3.330439 | tok/s 2621841.77 | norm 0.3700 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15212 | train 2.974246 | val 3.330439 | tok/s 2619871.30 | norm 0.3636 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15213 | train 3.036087 | val 3.330439 | tok/s 2624644.53 | norm 0.3468 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15214 | train 2.941751 | val 3.330439 | tok/s 2626067.71 | norm 0.4012 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15215 | train 3.070833 | val 3.330439 | tok/s 2625622.72 | norm 0.3909 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15216 | train 2.912026 | val 3.330439 | tok/s 2627204.43 | norm 0.3540 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15217 | train 3.159308 | val 3.330439 | tok/s 2617745.73 | norm 0.4031 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15218 | train 2.910753 | val 3.330439 | tok/s 2622838.37 | norm 0.3774 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15219 | train 3.056350 | val 3.330439 | tok/s 2623922.19 | norm 0.3556 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15220 | train 2.917319 | val 3.330439 | tok/s 2625394.71 | norm 0.3968 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15221 | train 3.014562 | val 3.330439 | tok/s 2627674.33 | norm 0.4038 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15222 | train 2.943783 | val 3.330439 | tok/s 2621075.38 | norm 0.4013 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15223 | train 3.022992 | val 3.330439 | tok/s 2629297.88 | norm 0.3785 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15224 | train 2.936480 | val 3.330439 | tok/s 2622095.41 | norm 0.3636 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15225 | train 2.941658 | val 3.330439 | tok/s 2627728.44 | norm 0.3901 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15226 | train 2.926703 | val 3.330439 | tok/s 2628443.88 | norm 0.4025 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15227 | train 2.960399 | val 3.330439 | tok/s 2626827.43 | norm 0.3949 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15228 | train 2.938903 | val 3.330439 | tok/s 2625816.11 | norm 0.3355 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15229 | train 2.937728 | val 3.330439 | tok/s 2626332.98 | norm 0.3820 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15230 | train 2.933126 | val 3.330439 | tok/s 2627407.01 | norm 0.4206 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15231 | train 2.974920 | val 3.330439 | tok/s 2616889.12 | norm 0.4032 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15232 | train 2.882966 | val 3.330439 | tok/s 2621584.33 | norm 0.3633 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15233 | train 2.996828 | val 3.330439 | tok/s 2629967.74 | norm 0.3531 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15234 | train 3.004028 | val 3.330439 | tok/s 2626469.85 | norm 0.3819 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15235 | train 2.824468 | val 3.330439 | tok/s 2628422.29 | norm 0.3847 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15236 | train 2.975735 | val 3.330439 | tok/s 2620488.28 | norm 0.4070 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15237 | train 2.909988 | val 3.330439 | tok/s 2621149.81 | norm 0.4109 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15238 | train 2.922697 | val 3.330439 | tok/s 2626934.80 | norm 0.4002 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15239 | train 2.893734 | val 3.330439 | tok/s 2628309.19 | norm 0.4326 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15240 | train 2.983737 | val 3.330439 | tok/s 2629161.18 | norm 0.4188 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15241 | train 3.031746 | val 3.330439 | tok/s 2626202.96 | norm 0.3825 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15242 | train 2.952065 | val 3.330439 | tok/s 2626325.28 | norm 0.4595 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15243 | train 2.904934 | val 3.330439 | tok/s 2626634.27 | norm 0.3874 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15244 | train 2.935122 | val 3.330439 | tok/s 2627758.44 | norm 0.3995 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15245 | train 3.039761 | val 3.330439 | tok/s 2631817.19 | norm 0.4030 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15246 | train 2.943892 | val 3.330439 | tok/s 2631002.15 | norm 0.3697 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15247 | train 3.031072 | val 3.330439 | tok/s 2628579.55 | norm 0.3305 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15248 | train 2.942098 | val 3.330439 | tok/s 2627235.72 | norm 0.3731 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15249 | train 2.966315 | val 3.330439 | tok/s 2626086.65 | norm 0.3675 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 15250 | val_loss 3.333222 step 15250 | train 2.967692 | val 3.333222 | tok/s 2620523.25 | norm 0.3718 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15251 | train 2.897556 | val 3.333222 | tok/s 2625694.56 | norm 0.3753 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15252 | train 2.989468 | val 3.333222 | tok/s 2625201.32 | norm 0.3927 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15253 | train 2.955060 | val 3.333222 | tok/s 2626133.23 | norm 0.3519 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15254 | train 2.949788 | val 3.333222 | tok/s 2629234.88 | norm 0.3554 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15255 | train 2.949824 | val 3.333222 | tok/s 2620313.17 | norm 0.3866 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15256 | train 2.949635 | val 3.333222 | tok/s 2625898.92 | norm 0.4009 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15257 | train 3.032357 | val 3.333222 | tok/s 2625616.12 | norm 0.3831 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15258 | train 2.993746 | val 3.333222 | tok/s 2626273.20 | norm 0.3905 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15259 | train 3.038725 | val 3.333222 | tok/s 2630211.22 | norm 0.3664 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15260 | train 3.011902 | val 3.333222 | tok/s 2624652.39 | norm 0.3463 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15261 | train 2.918128 | val 3.333222 | tok/s 2626656.70 | norm 0.3728 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15262 | train 3.028044 | val 3.333222 | tok/s 2624940.46 | norm 0.3999 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15263 | train 3.001888 | val 3.333222 | tok/s 2625167.19 | norm 0.4026 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15264 | train 3.100830 | val 3.333222 | tok/s 2630445.79 | norm 0.4123 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15265 | train 3.067417 | val 3.333222 | tok/s 2624325.96 | norm 0.3981 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15266 | train 3.016318 | val 3.333222 | tok/s 2629331.08 | norm 0.3515 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15267 | train 3.096590 | val 3.333222 | tok/s 2623155.46 | norm 0.3489 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15268 | train 2.901446 | val 3.333222 | tok/s 2623048.43 | norm 0.3764 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15269 | train 3.028226 | val 3.333222 | tok/s 2626324.28 | norm 0.3593 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15270 | train 2.964124 | val 3.333222 | tok/s 2624414.60 | norm 0.3778 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15271 | train 2.811355 | val 3.333222 | tok/s 2628717.27 | norm 0.3971 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15272 | train 2.886873 | val 3.333222 | tok/s 2621014.96 | norm 0.3756 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15273 | train 2.904195 | val 3.333222 | tok/s 2626407.80 | norm 0.3992 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15274 | train 2.999185 | val 3.333222 | tok/s 2629442.62 | norm 0.3875 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15275 | train 2.921621 | val 3.333222 | tok/s 2623175.82 | norm 0.3445 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15276 | train 3.007306 | val 3.333222 | tok/s 2621791.77 | norm 0.3770 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15277 | train 2.863863 | val 3.333222 | tok/s 2626474.48 | norm 0.3551 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15278 | train 2.959438 | val 3.333222 | tok/s 2629233.03 | norm 0.3618 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15279 | train 3.101725 | val 3.333222 | tok/s 2626767.60 | norm 0.4110 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15280 | train 3.031841 | val 3.333222 | tok/s 2622318.80 | norm 0.4368 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15281 | train 2.961001 | val 3.333222 | tok/s 2629218.69 | norm 0.3848 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15282 | train 2.988023 | val 3.333222 | tok/s 2624332.05 | norm 0.3657 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15283 | train 3.086535 | val 3.333222 | tok/s 2622463.88 | norm 0.4127 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15284 | train 2.908739 | val 3.333222 | tok/s 2629019.25 | norm 0.4164 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15285 | train 2.970995 | val 3.333222 | tok/s 2628616.11 | norm 0.4430 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15286 | train 3.325990 | val 3.333222 | tok/s 2624977.88 | norm 0.7129 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15287 | train 3.035119 | val 3.333222 | tok/s 2627173.28 | norm 0.3281 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15288 | train 2.989680 | val 3.333222 | tok/s 2629171.54 | norm 0.3219 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15289 | train 2.879229 | val 3.333222 | tok/s 2631048.11 | norm 0.3205 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15290 | train 2.894886 | val 3.333222 | tok/s 2629923.73 | norm 0.3882 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15291 | train 2.785620 | val 3.333222 | tok/s 2628261.85 | norm 0.3193 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15292 | train 2.904876 | val 3.333222 | tok/s 2625739.21 | norm 0.3159 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15293 | train 2.934061 | val 3.333222 | tok/s 2626804.92 | norm 0.3422 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15294 | train 3.300129 | val 3.333222 | tok/s 2623648.65 | norm 0.4163 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15295 | train 2.986635 | val 3.333222 | tok/s 2625407.38 | norm 0.3898 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15296 | train 3.019485 | val 3.333222 | tok/s 2628115.53 | norm 0.3609 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15297 | train 2.965089 | val 3.333222 | tok/s 2623293.66 | norm 0.3705 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15298 | train 3.019868 | val 3.333222 | tok/s 2625445.81 | norm 0.4099 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15299 | train 2.931598 | val 3.333222 | tok/s 2614047.83 | norm 0.3923 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15300 | train 2.979933 | val 3.333222 | tok/s 2622263.64 | norm 0.3784 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15301 | train 2.921906 | val 3.333222 | tok/s 2629434.93 | norm 0.3682 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15302 | train 3.054180 | val 3.333222 | tok/s 2629210.74 | norm 0.3851 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15303 | train 3.145000 | val 3.333222 | tok/s 2625283.06 | norm 0.4221 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15304 | train 2.953521 | val 3.333222 | tok/s 2619457.44 | norm 0.4066 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15305 | train 3.095771 | val 3.333222 | tok/s 2624700.76 | norm 0.3349 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15306 | train 2.912351 | val 3.333222 | tok/s 2619184.59 | norm 0.3478 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15307 | train 2.882255 | val 3.333222 | tok/s 2622040.70 | norm 0.3782 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15308 | train 2.951866 | val 3.333222 | tok/s 2626938.88 | norm 0.3723 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15309 | train 2.925770 | val 3.333222 | tok/s 2626249.36 | norm 0.3562 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15310 | train 3.021580 | val 3.333222 | tok/s 2610095.73 | norm 0.3773 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15311 | train 2.819526 | val 3.333222 | tok/s 2627185.26 | norm 0.3960 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15312 | train 3.001929 | val 3.333222 | tok/s 2627467.53 | norm 0.3885 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15313 | train 2.932794 | val 3.333222 | tok/s 2629100.20 | norm 0.3851 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15314 | train 3.130572 | val 3.333222 | tok/s 2623304.65 | norm 0.4090 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15315 | train 2.946467 | val 3.333222 | tok/s 2626739.82 | norm 0.3710 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15316 | train 2.949873 | val 3.333222 | tok/s 2625158.16 | norm 0.3561 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15317 | train 2.882864 | val 3.333222 | tok/s 2626843.72 | norm 0.3555 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15318 | train 2.967972 | val 3.333222 | tok/s 2627597.38 | norm 0.3434 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15319 | train 3.034084 | val 3.333222 | tok/s 2627633.50 | norm 0.3647 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15320 | train 2.948574 | val 3.333222 | tok/s 2626179.37 | norm 0.3652 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15321 | train 3.006696 | val 3.333222 | tok/s 2628487.83 | norm 0.3810 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15322 | train 3.013190 | val 3.333222 | tok/s 2623522.95 | norm 0.3633 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15323 | train 3.078509 | val 3.333222 | tok/s 2624535.04 | norm 0.4380 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15324 | train 3.022704 | val 3.333222 | tok/s 2626257.53 | norm 0.4154 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15325 | train 3.068128 | val 3.333222 | tok/s 2625256.73 | norm 0.3897 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15326 | train 2.934640 | val 3.333222 | tok/s 2626109.45 | norm 0.3574 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15327 | train 3.039035 | val 3.333222 | tok/s 2623224.25 | norm 0.3678 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15328 | train 3.058883 | val 3.333222 | tok/s 2623318.11 | norm 0.4142 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15329 | train 3.012089 | val 3.333222 | tok/s 2625725.95 | norm 0.3775 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15330 | train 2.993042 | val 3.333222 | tok/s 2435024.73 | norm 0.3492 | dt 215.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15331 | train 2.994101 | val 3.333222 | tok/s 2622300.09 | norm 0.3666 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15332 | train 2.884560 | val 3.333222 | tok/s 2625084.17 | norm 0.3877 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15333 | train 3.008562 | val 3.333222 | tok/s 2631722.34 | norm 0.4422 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15334 | train 2.991968 | val 3.333222 | tok/s 2623571.81 | norm 0.3897 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15335 | train 2.905225 | val 3.333222 | tok/s 2630323.56 | norm 0.3629 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15336 | train 2.901184 | val 3.333222 | tok/s 2631329.89 | norm 0.4202 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15337 | train 2.979404 | val 3.333222 | tok/s 2625611.10 | norm 0.3562 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15338 | train 2.978729 | val 3.333222 | tok/s 2628892.20 | norm 0.3373 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15339 | train 3.179187 | val 3.333222 | tok/s 2624820.57 | norm 0.3825 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15340 | train 2.989356 | val 3.333222 | tok/s 2624882.90 | norm 0.3910 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15341 | train 2.978019 | val 3.333222 | tok/s 2625467.86 | norm 0.3743 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15342 | train 3.092057 | val 3.333222 | tok/s 2622063.63 | norm 0.6022 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15343 | train 2.937487 | val 3.333222 | tok/s 2627418.40 | norm 0.4154 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15344 | train 2.958405 | val 3.333222 | tok/s 2626069.58 | norm 0.4011 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15345 | train 2.972741 | val 3.333222 | tok/s 2626073.90 | norm 0.4048 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15346 | train 2.973387 | val 3.333222 | tok/s 2623115.07 | norm 0.3731 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15347 | train 2.948265 | val 3.333222 | tok/s 2625044.88 | norm 0.3422 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15348 | train 2.952755 | val 3.333222 | tok/s 2626835.53 | norm 0.3885 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15349 | train 3.091321 | val 3.333222 | tok/s 2622416.60 | norm 0.4085 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15350 | train 2.915846 | val 3.333222 | tok/s 2623210.73 | norm 0.3788 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15351 | train 2.911083 | val 3.333222 | tok/s 2616228.34 | norm 0.3686 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15352 | train 3.007166 | val 3.333222 | tok/s 2616392.16 | norm 0.3773 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15353 | train 2.921802 | val 3.333222 | tok/s 2623852.45 | norm 0.3518 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15354 | train 2.961560 | val 3.333222 | tok/s 2623890.84 | norm 0.3941 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15355 | train 2.916570 | val 3.333222 | tok/s 2625718.41 | norm 0.4019 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15356 | train 2.970098 | val 3.333222 | tok/s 2623966.46 | norm 0.3919 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15357 | train 2.999367 | val 3.333222 | tok/s 2623158.31 | norm 0.3737 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15358 | train 2.985790 | val 3.333222 | tok/s 2628014.96 | norm 0.3748 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15359 | train 2.925392 | val 3.333222 | tok/s 2623878.25 | norm 0.4004 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15360 | train 2.999377 | val 3.333222 | tok/s 2624995.43 | norm 0.4050 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15361 | train 3.251495 | val 3.333222 | tok/s 2624191.94 | norm 0.5221 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15362 | train 3.064055 | val 3.333222 | tok/s 2625025.52 | norm 0.3915 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15363 | train 2.799524 | val 3.333222 | tok/s 2625340.05 | norm 0.3556 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15364 | train 3.010943 | val 3.333222 | tok/s 2622119.69 | norm 0.3605 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15365 | train 2.891283 | val 3.333222 | tok/s 2624840.26 | norm 0.3638 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15366 | train 3.048003 | val 3.333222 | tok/s 2626152.49 | norm 0.3493 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15367 | train 3.010142 | val 3.333222 | tok/s 2628983.03 | norm 0.3661 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15368 | train 2.935657 | val 3.333222 | tok/s 2628586.77 | norm 0.3655 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15369 | train 2.944357 | val 3.333222 | tok/s 2627392.90 | norm 0.3405 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15370 | train 2.957726 | val 3.333222 | tok/s 2625969.39 | norm 0.3688 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15371 | train 2.912763 | val 3.333222 | tok/s 2625337.75 | norm 0.3574 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15372 | train 2.843621 | val 3.333222 | tok/s 2625280.81 | norm 0.3802 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15373 | train 3.011760 | val 3.333222 | tok/s 2622956.40 | norm 0.4049 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15374 | train 3.039028 | val 3.333222 | tok/s 2621225.77 | norm 0.3833 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15375 | train 2.904395 | val 3.333222 | tok/s 2623896.22 | norm 0.3874 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15376 | train 2.964285 | val 3.333222 | tok/s 2618239.05 | norm 0.4046 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15377 | train 3.015940 | val 3.333222 | tok/s 2620801.07 | norm 0.4111 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15378 | train 2.889905 | val 3.333222 | tok/s 2627460.18 | norm 0.3779 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15379 | train 2.944708 | val 3.333222 | tok/s 2631872.97 | norm 0.3812 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15380 | train 3.085585 | val 3.333222 | tok/s 2629283.03 | norm 0.3919 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15381 | train 2.905571 | val 3.333222 | tok/s 2623445.81 | norm 0.3940 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15382 | train 2.954133 | val 3.333222 | tok/s 2621517.94 | norm 0.3872 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15383 | train 2.952434 | val 3.333222 | tok/s 2620771.57 | norm 0.3678 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15384 | train 3.118866 | val 3.333222 | tok/s 2623589.03 | norm 0.3659 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15385 | train 2.805293 | val 3.333222 | tok/s 2626577.39 | norm 0.3600 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15386 | train 2.989058 | val 3.333222 | tok/s 2623832.67 | norm 0.3794 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15387 | train 3.057354 | val 3.333222 | tok/s 2625357.27 | norm 0.4245 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15388 | train 3.119392 | val 3.333222 | tok/s 2627857.10 | norm 0.4077 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15389 | train 2.954007 | val 3.333222 | tok/s 2629345.22 | norm 0.3381 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15390 | train 3.006829 | val 3.333222 | tok/s 2630140.95 | norm 0.3592 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15391 | train 2.912464 | val 3.333222 | tok/s 2624459.80 | norm 0.3893 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15392 | train 2.978513 | val 3.333222 | tok/s 2629998.29 | norm 0.3952 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15393 | train 2.956388 | val 3.333222 | tok/s 2628436.64 | norm 0.3553 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15394 | train 2.994966 | val 3.333222 | tok/s 2626668.13 | norm 0.3623 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15395 | train 3.039402 | val 3.333222 | tok/s 2630297.38 | norm 0.3565 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15396 | train 3.046157 | val 3.333222 | tok/s 2627406.42 | norm 0.3984 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15397 | train 2.990070 | val 3.333222 | tok/s 2627345.69 | norm 0.3933 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15398 | train 2.870448 | val 3.333222 | tok/s 2628263.67 | norm 0.4080 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15399 | train 2.845533 | val 3.333222 | tok/s 2623709.22 | norm 0.4320 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15400 | train 3.050707 | val 3.333222 | tok/s 2629323.07 | norm 0.3988 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15401 | train 2.922865 | val 3.333222 | tok/s 2627204.45 | norm 0.3644 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15402 | train 3.065657 | val 3.333222 | tok/s 2628850.93 | norm 0.3646 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15403 | train 2.954295 | val 3.333222 | tok/s 2625575.66 | norm 0.3722 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15404 | train 2.905838 | val 3.333222 | tok/s 2620228.62 | norm 0.3414 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15405 | train 3.186276 | val 3.333222 | tok/s 2623760.05 | norm 0.4072 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15406 | train 2.953494 | val 3.333222 | tok/s 2626841.89 | norm 0.3660 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15407 | train 3.053297 | val 3.333222 | tok/s 2626543.41 | norm 0.3572 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15408 | train 3.115178 | val 3.333222 | tok/s 2622937.61 | norm 0.4541 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15409 | train 2.949064 | val 3.333222 | tok/s 2621897.05 | norm 0.3772 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15410 | train 2.898883 | val 3.333222 | tok/s 2626944.33 | norm 0.4169 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15411 | train 3.001699 | val 3.333222 | tok/s 2629808.55 | norm 0.3985 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15412 | train 2.873912 | val 3.333222 | tok/s 2629531.28 | norm 0.4038 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15413 | train 3.056382 | val 3.333222 | tok/s 2628113.19 | norm 0.4152 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15414 | train 2.955084 | val 3.333222 | tok/s 2625089.82 | norm 0.3787 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15415 | train 2.994312 | val 3.333222 | tok/s 2625448.55 | norm 0.3861 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15416 | train 2.930893 | val 3.333222 | tok/s 2628457.46 | norm 0.4036 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15417 | train 2.891406 | val 3.333222 | tok/s 2625118.66 | norm 0.3811 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15418 | train 2.847131 | val 3.333222 | tok/s 2622100.27 | norm 0.3733 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15419 | train 2.957142 | val 3.333222 | tok/s 2621746.88 | norm 0.3783 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15420 | train 2.881659 | val 3.333222 | tok/s 2626970.63 | norm 0.3705 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15421 | train 2.857651 | val 3.333222 | tok/s 2624412.40 | norm 0.3917 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15422 | train 2.911587 | val 3.333222 | tok/s 2627895.45 | norm 0.3901 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15423 | train 2.833103 | val 3.333222 | tok/s 2625605.38 | norm 0.3585 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15424 | train 3.046442 | val 3.333222 | tok/s 2630763.56 | norm 0.3737 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15425 | train 3.028607 | val 3.333222 | tok/s 2626639.87 | norm 0.3489 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15426 | train 3.000562 | val 3.333222 | tok/s 2626864.28 | norm 0.3474 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15427 | train 2.988953 | val 3.333222 | tok/s 2626633.72 | norm 0.3575 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15428 | train 2.916685 | val 3.333222 | tok/s 2616290.02 | norm 0.3527 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15429 | train 2.924017 | val 3.333222 | tok/s 2621893.29 | norm 0.3615 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15430 | train 2.973846 | val 3.333222 | tok/s 2626769.95 | norm 0.3471 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15431 | train 2.910098 | val 3.333222 | tok/s 2625978.07 | norm 0.3639 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15432 | train 3.023501 | val 3.333222 | tok/s 2627133.84 | norm 0.3623 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15433 | train 3.035633 | val 3.333222 | tok/s 2623087.54 | norm 0.3514 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15434 | train 3.072490 | val 3.333222 | tok/s 2625709.81 | norm 0.3539 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15435 | train 2.970899 | val 3.333222 | tok/s 2628551.76 | norm 0.3707 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15436 | train 2.966661 | val 3.333222 | tok/s 2625759.37 | norm 0.3483 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15437 | train 2.978054 | val 3.333222 | tok/s 2624309.20 | norm 0.4136 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15438 | train 2.983489 | val 3.333222 | tok/s 2624249.53 | norm 0.3745 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15439 | train 2.897654 | val 3.333222 | tok/s 2625339.17 | norm 0.3360 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15440 | train 2.951553 | val 3.333222 | tok/s 2626420.24 | norm 0.3592 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15441 | train 2.996535 | val 3.333222 | tok/s 2623822.59 | norm 0.4168 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15442 | train 3.088722 | val 3.333222 | tok/s 2574786.83 | norm 0.4167 | dt 203.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15443 | train 3.010107 | val 3.333222 | tok/s 2534075.37 | norm 0.4154 | dt 206.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15444 | train 2.896101 | val 3.333222 | tok/s 2625179.16 | norm 0.4098 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15445 | train 2.990463 | val 3.333222 | tok/s 2625058.58 | norm 0.3671 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15446 | train 2.754789 | val 3.333222 | tok/s 2621520.43 | norm 0.3717 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15447 | train 2.902778 | val 3.333222 | tok/s 2629261.80 | norm 0.4023 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15448 | train 2.995397 | val 3.333222 | tok/s 2626513.85 | norm 0.3962 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15449 | train 2.957436 | val 3.333222 | tok/s 2626059.25 | norm 0.3976 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15450 | train 2.967768 | val 3.333222 | tok/s 2629069.13 | norm 0.3766 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15451 | train 3.009562 | val 3.333222 | tok/s 2621712.38 | norm 0.3338 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15452 | train 2.962939 | val 3.333222 | tok/s 2628358.50 | norm 0.4025 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15453 | train 2.973016 | val 3.333222 | tok/s 2627568.32 | norm 0.4241 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15454 | train 3.017492 | val 3.333222 | tok/s 2628785.67 | norm 0.3767 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15455 | train 2.942553 | val 3.333222 | tok/s 2628674.95 | norm 0.3456 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15456 | train 2.963140 | val 3.333222 | tok/s 2619095.65 | norm 0.3650 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15457 | train 2.983800 | val 3.333222 | tok/s 2624064.21 | norm 0.3888 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15458 | train 2.988869 | val 3.333222 | tok/s 2623897.70 | norm 0.4058 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15459 | train 2.941210 | val 3.333222 | tok/s 2626399.09 | norm 0.3802 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15460 | train 2.855539 | val 3.333222 | tok/s 2626580.45 | norm 0.4314 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15461 | train 2.956693 | val 3.333222 | tok/s 2622627.13 | norm 0.3997 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15462 | train 2.975000 | val 3.333222 | tok/s 2626288.34 | norm 0.3656 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15463 | train 3.060108 | val 3.333222 | tok/s 2627094.04 | norm 0.3916 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15464 | train 2.964285 | val 3.333222 | tok/s 2622734.99 | norm 0.3966 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15465 | train 2.952111 | val 3.333222 | tok/s 2619171.65 | norm 0.3714 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15466 | train 3.007894 | val 3.333222 | tok/s 2620737.38 | norm 0.3773 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15467 | train 2.826966 | val 3.333222 | tok/s 2627705.49 | norm 0.3917 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15468 | train 2.911029 | val 3.333222 | tok/s 2619562.18 | norm 0.3895 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15469 | train 2.843071 | val 3.333222 | tok/s 2624685.24 | norm 0.9170 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15470 | train 2.919358 | val 3.333222 | tok/s 2626438.80 | norm 0.2996 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15471 | train 2.906884 | val 3.333222 | tok/s 2624433.10 | norm 0.3553 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15472 | train 2.935518 | val 3.333222 | tok/s 2627762.00 | norm 0.3775 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15473 | train 2.935900 | val 3.333222 | tok/s 2625899.95 | norm 0.3684 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15474 | train 2.896878 | val 3.333222 | tok/s 2627073.52 | norm 0.3717 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15475 | train 3.055067 | val 3.333222 | tok/s 2621652.55 | norm 0.3546 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15476 | train 2.857711 | val 3.333222 | tok/s 2620672.60 | norm 0.3466 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15477 | train 2.858143 | val 3.333222 | tok/s 2629744.43 | norm 0.3595 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15478 | train 3.038349 | val 3.333222 | tok/s 2629537.17 | norm 0.4566 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15479 | train 3.124925 | val 3.333222 | tok/s 2625418.22 | norm 0.4060 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15480 | train 2.906031 | val 3.333222 | tok/s 2619325.69 | norm 0.3558 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15481 | train 2.984428 | val 3.333222 | tok/s 2622529.03 | norm 0.3453 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15482 | train 3.023139 | val 3.333222 | tok/s 2625057.50 | norm 0.3265 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15483 | train 3.049027 | val 3.333222 | tok/s 2626566.06 | norm 0.3671 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15484 | train 2.978239 | val 3.333222 | tok/s 2599947.68 | norm 0.3777 | dt 201.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15485 | train 2.904272 | val 3.333222 | tok/s 2624524.76 | norm 0.3952 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15486 | train 2.901687 | val 3.333222 | tok/s 2625242.97 | norm 0.3373 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15487 | train 3.156728 | val 3.333222 | tok/s 2627808.76 | norm 0.3417 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15488 | train 2.971607 | val 3.333222 | tok/s 2626795.30 | norm 0.3324 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15489 | train 2.888404 | val 3.333222 | tok/s 2628244.14 | norm 0.3595 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15490 | train 2.976664 | val 3.333222 | tok/s 2622452.49 | norm 0.3833 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15491 | train 3.062986 | val 3.333222 | tok/s 2624968.32 | norm 0.4374 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15492 | train 2.892283 | val 3.333222 | tok/s 2628262.90 | norm 0.4032 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15493 | train 3.049788 | val 3.333222 | tok/s 2630347.37 | norm 0.3514 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15494 | train 3.019200 | val 3.333222 | tok/s 2629788.84 | norm 0.3648 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15495 | train 2.986292 | val 3.333222 | tok/s 2621687.29 | norm 0.3904 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15496 | train 2.916104 | val 3.333222 | tok/s 2622406.88 | norm 0.4076 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15497 | train 2.864682 | val 3.333222 | tok/s 2625571.78 | norm 0.4024 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15498 | train 3.060527 | val 3.333222 | tok/s 2621712.24 | norm 0.4001 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15499 | train 3.045664 | val 3.333222 | tok/s 2626482.20 | norm 0.3882 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 15500 | val_loss 3.333140 step 15500 | train 2.916074 | val 3.333140 | tok/s 2622730.54 | norm 0.3714 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15501 | train 2.947067 | val 3.333140 | tok/s 2623536.78 | norm 0.3781 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15502 | train 2.890029 | val 3.333140 | tok/s 2628654.33 | norm 0.3738 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15503 | train 2.969808 | val 3.333140 | tok/s 2613042.71 | norm 0.3950 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15504 | train 2.951519 | val 3.333140 | tok/s 2624347.69 | norm 0.3916 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15505 | train 2.945111 | val 3.333140 | tok/s 2628134.17 | norm 0.3504 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15506 | train 2.971304 | val 3.333140 | tok/s 2628119.29 | norm 0.3447 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15507 | train 2.905643 | val 3.333140 | tok/s 2627946.10 | norm 0.3906 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15508 | train 2.875160 | val 3.333140 | tok/s 2628911.68 | norm 0.3815 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15509 | train 2.929002 | val 3.333140 | tok/s 2625565.18 | norm 0.3571 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15510 | train 2.975169 | val 3.333140 | tok/s 2627110.88 | norm 0.3649 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15511 | train 2.950082 | val 3.333140 | tok/s 2625966.90 | norm 0.3713 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15512 | train 2.948837 | val 3.333140 | tok/s 2615201.78 | norm 0.4462 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15513 | train 2.914594 | val 3.333140 | tok/s 2626631.50 | norm 0.3778 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15514 | train 2.977179 | val 3.333140 | tok/s 2629402.99 | norm 0.3780 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15515 | train 2.879121 | val 3.333140 | tok/s 2627641.48 | norm 0.4213 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15516 | train 2.941071 | val 3.333140 | tok/s 2624987.37 | norm 0.4115 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15517 | train 2.850492 | val 3.333140 | tok/s 2623472.81 | norm 0.3753 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15518 | train 2.959510 | val 3.333140 | tok/s 2623747.57 | norm 0.3751 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15519 | train 2.934427 | val 3.333140 | tok/s 2564942.01 | norm 0.3723 | dt 204.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15520 | train 3.086363 | val 3.333140 | tok/s 2446993.14 | norm 0.4027 | dt 214.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15521 | train 2.942708 | val 3.333140 | tok/s 2623739.26 | norm 0.4128 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15522 | train 2.891852 | val 3.333140 | tok/s 2630634.21 | norm 0.3821 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15523 | train 3.053567 | val 3.333140 | tok/s 2627668.34 | norm 0.3886 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15524 | train 3.014972 | val 3.333140 | tok/s 2626595.81 | norm 0.3946 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15525 | train 3.034001 | val 3.333140 | tok/s 2628620.85 | norm 0.3929 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15526 | train 2.924536 | val 3.333140 | tok/s 2627474.97 | norm 0.4230 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15527 | train 2.926649 | val 3.333140 | tok/s 2628490.37 | norm 0.3992 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15528 | train 2.977698 | val 3.333140 | tok/s 2628052.27 | norm 0.3780 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15529 | train 3.070684 | val 3.333140 | tok/s 2625951.15 | norm 0.4110 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15530 | train 2.867227 | val 3.333140 | tok/s 2624635.01 | norm 0.3898 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15531 | train 2.925925 | val 3.333140 | tok/s 2629757.63 | norm 0.3610 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15532 | train 2.967356 | val 3.333140 | tok/s 2623655.68 | norm 0.3907 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15533 | train 2.866712 | val 3.333140 | tok/s 2619006.85 | norm 0.4274 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15534 | train 3.029837 | val 3.333140 | tok/s 2612042.60 | norm 0.4212 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15535 | train 2.965485 | val 3.333140 | tok/s 2612511.01 | norm 0.4010 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15536 | train 2.972366 | val 3.333140 | tok/s 2626807.00 | norm 0.3775 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15537 | train 2.886548 | val 3.333140 | tok/s 2624360.99 | norm 0.4049 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15538 | train 3.021694 | val 3.333140 | tok/s 2617003.04 | norm 0.4229 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15539 | train 2.974567 | val 3.333140 | tok/s 2621258.84 | norm 0.3731 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15540 | train 3.002011 | val 3.333140 | tok/s 2626521.49 | norm 0.3828 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15541 | train 2.819815 | val 3.333140 | tok/s 2627304.97 | norm 0.3771 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15542 | train 2.830409 | val 3.333140 | tok/s 2625039.20 | norm 0.3843 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15543 | train 2.999379 | val 3.333140 | tok/s 2622337.19 | norm 0.4404 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15544 | train 2.874546 | val 3.333140 | tok/s 2622623.49 | norm 0.3714 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15545 | train 3.022081 | val 3.333140 | tok/s 2628691.89 | norm 0.3614 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15546 | train 2.988132 | val 3.333140 | tok/s 2625229.32 | norm 0.3715 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15547 | train 2.846270 | val 3.333140 | tok/s 2627412.49 | norm 0.3407 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15548 | train 3.010053 | val 3.333140 | tok/s 2627677.34 | norm 0.3553 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15549 | train 2.989621 | val 3.333140 | tok/s 2626930.40 | norm 0.3605 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15550 | train 2.975445 | val 3.333140 | tok/s 2628761.90 | norm 0.3612 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15551 | train 2.956048 | val 3.333140 | tok/s 2625022.33 | norm 0.3840 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15552 | train 3.040705 | val 3.333140 | tok/s 2624853.64 | norm 0.4341 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15553 | train 2.968785 | val 3.333140 | tok/s 2621533.44 | norm 0.3997 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15554 | train 2.939364 | val 3.333140 | tok/s 2622332.89 | norm 0.3702 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15555 | train 2.973917 | val 3.333140 | tok/s 2628143.69 | norm 0.3601 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15556 | train 2.933169 | val 3.333140 | tok/s 2627374.94 | norm 0.3694 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15557 | train 3.040626 | val 3.333140 | tok/s 2625039.94 | norm 0.3914 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15558 | train 2.944643 | val 3.333140 | tok/s 2624149.16 | norm 0.3663 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15559 | train 2.943713 | val 3.333140 | tok/s 2623288.95 | norm 0.3217 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15560 | train 2.955496 | val 3.333140 | tok/s 2629312.85 | norm 0.3490 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15561 | train 2.970289 | val 3.333140 | tok/s 2628926.75 | norm 0.3644 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15562 | train 2.940451 | val 3.333140 | tok/s 2627145.77 | norm 0.4001 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15563 | train 3.013230 | val 3.333140 | tok/s 2620530.47 | norm 0.4148 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15564 | train 2.952454 | val 3.333140 | tok/s 2620480.42 | norm 0.4275 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15565 | train 2.828982 | val 3.333140 | tok/s 2630305.18 | norm 0.4154 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15566 | train 2.955942 | val 3.333140 | tok/s 2624824.09 | norm 0.4095 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15567 | train 2.977386 | val 3.333140 | tok/s 2628942.37 | norm 0.3863 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15568 | train 2.973394 | val 3.333140 | tok/s 2628187.56 | norm 0.4010 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15569 | train 2.951946 | val 3.333140 | tok/s 2626789.76 | norm 0.4206 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15570 | train 2.923627 | val 3.333140 | tok/s 2589894.69 | norm 0.3976 | dt 202.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15571 | train 2.850543 | val 3.333140 | tok/s 2625529.16 | norm 0.3819 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15572 | train 2.915455 | val 3.333140 | tok/s 2626572.95 | norm 0.3810 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15573 | train 3.003883 | val 3.333140 | tok/s 2626023.54 | norm 0.4223 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15574 | train 3.074715 | val 3.333140 | tok/s 2627509.03 | norm 0.4077 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15575 | train 2.894003 | val 3.333140 | tok/s 2626565.48 | norm 0.3461 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15576 | train 3.052902 | val 3.333140 | tok/s 2624884.48 | norm 0.3808 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15577 | train 2.805231 | val 3.333140 | tok/s 2625515.77 | norm 0.3423 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15578 | train 2.940500 | val 3.333140 | tok/s 2624449.99 | norm 0.3427 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15579 | train 2.987819 | val 3.333140 | tok/s 2627165.38 | norm 0.3779 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15580 | train 2.942827 | val 3.333140 | tok/s 2626003.63 | norm 0.3823 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15581 | train 3.310745 | val 3.333140 | tok/s 2624336.64 | norm 0.9738 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15582 | train 2.926237 | val 3.333140 | tok/s 2629559.36 | norm 0.4188 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15583 | train 2.869798 | val 3.333140 | tok/s 2628005.81 | norm 0.3532 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15584 | train 2.963851 | val 3.333140 | tok/s 2628024.57 | norm 0.3377 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15585 | train 2.918316 | val 3.333140 | tok/s 2624684.86 | norm 0.3637 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15586 | train 3.002489 | val 3.333140 | tok/s 2620087.94 | norm 0.3841 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15587 | train 2.949459 | val 3.333140 | tok/s 2625138.28 | norm 0.3511 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15588 | train 2.996016 | val 3.333140 | tok/s 2628885.58 | norm 0.3312 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15589 | train 2.872358 | val 3.333140 | tok/s 2628546.10 | norm 0.3583 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15590 | train 3.053523 | val 3.333140 | tok/s 2614647.47 | norm 0.3912 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15591 | train 2.902255 | val 3.333140 | tok/s 2622551.79 | norm 0.3695 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15592 | train 3.020362 | val 3.333140 | tok/s 2627503.00 | norm 0.3701 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15593 | train 2.901515 | val 3.333140 | tok/s 2620253.20 | norm 0.3705 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15594 | train 2.962349 | val 3.333140 | tok/s 2622929.61 | norm 0.3679 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15595 | train 2.947375 | val 3.333140 | tok/s 2620898.84 | norm 0.3811 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15596 | train 2.967027 | val 3.333140 | tok/s 2625731.11 | norm 0.3777 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15597 | train 2.996526 | val 3.333140 | tok/s 2603494.79 | norm 0.4172 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15598 | train 3.007304 | val 3.333140 | tok/s 2623925.26 | norm 0.4294 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15599 | train 2.831297 | val 3.333140 | tok/s 2625791.00 | norm 0.4958 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15600 | train 2.967973 | val 3.333140 | tok/s 2624864.58 | norm 0.4371 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15601 | train 3.001877 | val 3.333140 | tok/s 2619620.53 | norm 0.3423 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15602 | train 3.087981 | val 3.333140 | tok/s 2630623.59 | norm 0.3643 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15603 | train 3.019895 | val 3.333140 | tok/s 2630508.48 | norm 0.3537 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15604 | train 2.961246 | val 3.333140 | tok/s 2629393.39 | norm 0.3148 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15605 | train 2.829606 | val 3.333140 | tok/s 2617729.80 | norm 0.3657 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15606 | train 2.981512 | val 3.333140 | tok/s 2624284.30 | norm 0.3897 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15607 | train 2.939746 | val 3.333140 | tok/s 2628242.04 | norm 0.3725 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15608 | train 3.006276 | val 3.333140 | tok/s 2628672.58 | norm 0.3999 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15609 | train 2.939458 | val 3.333140 | tok/s 2622843.33 | norm 0.3676 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15610 | train 2.825413 | val 3.333140 | tok/s 2623787.46 | norm 0.3760 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15611 | train 2.973234 | val 3.333140 | tok/s 2625922.99 | norm 0.4245 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15612 | train 3.033947 | val 3.333140 | tok/s 2627528.19 | norm 0.4024 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15613 | train 2.942868 | val 3.333140 | tok/s 2624024.19 | norm 0.3649 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15614 | train 2.964772 | val 3.333140 | tok/s 2629684.39 | norm 0.3696 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15615 | train 2.937244 | val 3.333140 | tok/s 2623591.10 | norm 0.3770 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15616 | train 2.946165 | val 3.333140 | tok/s 2613502.41 | norm 0.3805 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15617 | train 2.971723 | val 3.333140 | tok/s 2626640.84 | norm 0.3657 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15618 | train 2.960375 | val 3.333140 | tok/s 2624638.52 | norm 0.3772 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15619 | train 2.896560 | val 3.333140 | tok/s 2624001.90 | norm 0.3933 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15620 | train 2.854855 | val 3.333140 | tok/s 2624064.75 | norm 0.4118 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15621 | train 3.178022 | val 3.333140 | tok/s 2623573.57 | norm 0.4081 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15622 | train 2.819985 | val 3.333140 | tok/s 2629029.31 | norm 0.3792 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15623 | train 3.129241 | val 3.333140 | tok/s 2626779.03 | norm 0.3632 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15624 | train 3.129248 | val 3.333140 | tok/s 2626088.62 | norm 0.4110 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15625 | train 3.010981 | val 3.333140 | tok/s 2624894.94 | norm 0.3990 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15626 | train 2.931976 | val 3.333140 | tok/s 2625904.67 | norm 0.3865 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15627 | train 3.030552 | val 3.333140 | tok/s 2624812.74 | norm 0.3507 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15628 | train 2.979203 | val 3.333140 | tok/s 2628634.12 | norm 0.4106 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15629 | train 2.876544 | val 3.333140 | tok/s 2628270.18 | norm 0.3793 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15630 | train 3.302609 | val 3.333140 | tok/s 2628250.82 | norm 0.4209 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15631 | train 3.037357 | val 3.333140 | tok/s 2630355.92 | norm 0.3841 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15632 | train 2.956267 | val 3.333140 | tok/s 2625384.46 | norm 0.3659 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15633 | train 2.982842 | val 3.333140 | tok/s 2621979.35 | norm 0.4203 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15634 | train 2.984372 | val 3.333140 | tok/s 2627829.19 | norm 0.3536 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15635 | train 2.915738 | val 3.333140 | tok/s 2441914.01 | norm 0.3494 | dt 214.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15636 | train 2.974824 | val 3.333140 | tok/s 2618476.83 | norm 0.3571 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15637 | train 2.919211 | val 3.333140 | tok/s 2628331.46 | norm 0.3433 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15638 | train 2.983139 | val 3.333140 | tok/s 2624163.45 | norm 0.3803 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15639 | train 2.802184 | val 3.333140 | tok/s 2628893.28 | norm 0.3691 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15640 | train 3.181476 | val 3.333140 | tok/s 2628229.15 | norm 0.4102 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15641 | train 2.919991 | val 3.333140 | tok/s 2627664.78 | norm 0.3905 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15642 | train 2.960675 | val 3.333140 | tok/s 2623568.44 | norm 0.3686 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15643 | train 2.809438 | val 3.333140 | tok/s 2624835.17 | norm 0.3714 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15644 | train 3.142309 | val 3.333140 | tok/s 2627866.42 | norm 0.3954 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15645 | train 2.943372 | val 3.333140 | tok/s 2626068.51 | norm 0.3829 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15646 | train 3.157278 | val 3.333140 | tok/s 2625575.03 | norm 0.4038 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15647 | train 2.921228 | val 3.333140 | tok/s 2624879.33 | norm 0.3915 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15648 | train 3.093605 | val 3.333140 | tok/s 2625231.27 | norm 0.3821 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15649 | train 3.123091 | val 3.333140 | tok/s 2627774.36 | norm 0.4014 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15650 | train 2.985720 | val 3.333140 | tok/s 2603255.00 | norm 0.3918 | dt 201.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15651 | train 2.833357 | val 3.333140 | tok/s 2625977.23 | norm 0.3793 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15652 | train 2.931686 | val 3.333140 | tok/s 2622169.21 | norm 0.4271 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15653 | train 2.957886 | val 3.333140 | tok/s 2625828.98 | norm 0.4539 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15654 | train 2.895113 | val 3.333140 | tok/s 2628529.78 | norm 0.3508 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15655 | train 2.935514 | val 3.333140 | tok/s 2622472.85 | norm 0.3237 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15656 | train 2.915428 | val 3.333140 | tok/s 2622511.93 | norm 0.3240 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15657 | train 2.954603 | val 3.333140 | tok/s 2614374.98 | norm 0.3986 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15658 | train 2.946575 | val 3.333140 | tok/s 2628771.88 | norm 0.3923 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15659 | train 2.991889 | val 3.333140 | tok/s 2625916.72 | norm 0.3757 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15660 | train 2.965022 | val 3.333140 | tok/s 2626103.59 | norm 0.3615 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15661 | train 2.967284 | val 3.333140 | tok/s 2626395.69 | norm 0.3851 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15662 | train 3.071988 | val 3.333140 | tok/s 2627677.23 | norm 0.3986 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15663 | train 2.954649 | val 3.333140 | tok/s 2626843.25 | norm 0.3845 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15664 | train 2.891585 | val 3.333140 | tok/s 2630077.67 | norm 0.3464 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15665 | train 2.945578 | val 3.333140 | tok/s 2626841.50 | norm 0.3657 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15666 | train 2.809727 | val 3.333140 | tok/s 2627480.59 | norm 0.4271 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15667 | train 2.857082 | val 3.333140 | tok/s 2625536.77 | norm 0.4499 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15668 | train 3.001048 | val 3.333140 | tok/s 2626275.66 | norm 0.4210 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15669 | train 2.934792 | val 3.333140 | tok/s 2625802.53 | norm 0.3854 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15670 | train 2.938397 | val 3.333140 | tok/s 2623413.97 | norm 0.3415 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15671 | train 3.027910 | val 3.333140 | tok/s 2627027.35 | norm 0.3444 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15672 | train 2.921937 | val 3.333140 | tok/s 2628645.35 | norm 0.3448 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15673 | train 2.998027 | val 3.333140 | tok/s 2622604.75 | norm 0.3548 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15674 | train 2.992096 | val 3.333140 | tok/s 2627679.53 | norm 0.3946 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15675 | train 2.984735 | val 3.333140 | tok/s 2627353.71 | norm 0.3544 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15676 | train 2.975374 | val 3.333140 | tok/s 2631260.36 | norm 0.3363 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15677 | train 3.054353 | val 3.333140 | tok/s 2624324.68 | norm 0.3628 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15678 | train 2.930094 | val 3.333140 | tok/s 2621672.10 | norm 0.3751 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15679 | train 2.951213 | val 3.333140 | tok/s 2625917.37 | norm 0.3800 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15680 | train 3.061878 | val 3.333140 | tok/s 2622256.48 | norm 0.3392 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15681 | train 2.944343 | val 3.333140 | tok/s 2624577.89 | norm 0.3625 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15682 | train 3.060519 | val 3.333140 | tok/s 2629135.99 | norm 0.3969 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15683 | train 3.090674 | val 3.333140 | tok/s 2628264.26 | norm 0.3854 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15684 | train 3.036463 | val 3.333140 | tok/s 2628923.27 | norm 0.3661 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15685 | train 2.805730 | val 3.333140 | tok/s 2627031.55 | norm 0.3764 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15686 | train 2.934294 | val 3.333140 | tok/s 2628169.92 | norm 0.3817 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15687 | train 2.808305 | val 3.333140 | tok/s 2623521.58 | norm 0.3711 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15688 | train 2.816655 | val 3.333140 | tok/s 2625434.76 | norm 0.7976 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15689 | train 2.913529 | val 3.333140 | tok/s 2628543.90 | norm 0.3298 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15690 | train 3.002325 | val 3.333140 | tok/s 2620844.99 | norm 0.3102 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15691 | train 2.892925 | val 3.333140 | tok/s 2626405.19 | norm 0.3262 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15692 | train 3.035067 | val 3.333140 | tok/s 2625867.04 | norm 0.3681 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15693 | train 2.995888 | val 3.333140 | tok/s 2627446.18 | norm 0.3923 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15694 | train 3.002281 | val 3.333140 | tok/s 2623479.32 | norm 0.3717 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15695 | train 2.961253 | val 3.333140 | tok/s 2623683.25 | norm 0.4354 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15696 | train 3.095235 | val 3.333140 | tok/s 2624471.86 | norm 0.4064 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15697 | train 3.060767 | val 3.333140 | tok/s 2492789.28 | norm 0.3731 | dt 210.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15698 | train 2.960590 | val 3.333140 | tok/s 2626878.85 | norm 0.4099 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15699 | train 2.937708 | val 3.333140 | tok/s 2625592.18 | norm 0.3642 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15700 | train 2.874584 | val 3.333140 | tok/s 2627617.21 | norm 0.3549 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15701 | train 2.846951 | val 3.333140 | tok/s 2629183.56 | norm 0.3446 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15702 | train 3.039704 | val 3.333140 | tok/s 2625047.91 | norm 0.3579 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15703 | train 3.022928 | val 3.333140 | tok/s 2629237.19 | norm 0.3480 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15704 | train 2.944623 | val 3.333140 | tok/s 2619477.58 | norm 0.3210 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15705 | train 3.008885 | val 3.333140 | tok/s 2621555.61 | norm 0.3651 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15706 | train 2.956136 | val 3.333140 | tok/s 2629367.50 | norm 0.3650 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15707 | train 2.951781 | val 3.333140 | tok/s 2573931.23 | norm 0.3770 | dt 203.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15708 | train 3.065137 | val 3.333140 | tok/s 2626313.73 | norm 0.4008 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15709 | train 2.953782 | val 3.333140 | tok/s 2621861.39 | norm 0.3793 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15710 | train 2.918742 | val 3.333140 | tok/s 2587143.79 | norm 0.3484 | dt 202.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15711 | train 2.981055 | val 3.333140 | tok/s 2630918.70 | norm 0.3839 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15712 | train 2.905977 | val 3.333140 | tok/s 2625407.24 | norm 0.3759 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15713 | train 2.872906 | val 3.333140 | tok/s 2622928.56 | norm 0.3790 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15714 | train 2.988630 | val 3.333140 | tok/s 2624238.61 | norm 0.3752 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15715 | train 2.996356 | val 3.333140 | tok/s 2624531.69 | norm 0.3601 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15716 | train 3.129874 | val 3.333140 | tok/s 2624272.21 | norm 0.3939 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15717 | train 2.953293 | val 3.333140 | tok/s 2626534.07 | norm 0.4013 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15718 | train 3.014897 | val 3.333140 | tok/s 2626079.00 | norm 0.3562 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15719 | train 2.919529 | val 3.333140 | tok/s 2622161.11 | norm 0.4257 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15720 | train 2.895685 | val 3.333140 | tok/s 2622558.85 | norm 0.3747 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15721 | train 2.877589 | val 3.333140 | tok/s 2621632.66 | norm 0.3521 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15722 | train 3.014204 | val 3.333140 | tok/s 2623610.40 | norm 0.3959 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15723 | train 3.057846 | val 3.333140 | tok/s 2626888.64 | norm 0.3929 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15724 | train 2.993654 | val 3.333140 | tok/s 2621523.02 | norm 0.3355 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15725 | train 2.924186 | val 3.333140 | tok/s 2622911.57 | norm 0.3745 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15726 | train 2.908713 | val 3.333140 | tok/s 2628636.64 | norm 0.3952 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15727 | train 3.040556 | val 3.333140 | tok/s 2626207.18 | norm 0.3743 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15728 | train 2.926797 | val 3.333140 | tok/s 2626230.68 | norm 0.3712 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15729 | train 3.071977 | val 3.333140 | tok/s 2625926.26 | norm 0.4241 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15730 | train 3.133637 | val 3.333140 | tok/s 2623556.27 | norm 0.4312 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15731 | train 2.999656 | val 3.333140 | tok/s 2627866.40 | norm 0.3773 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15732 | train 3.174132 | val 3.333140 | tok/s 2620917.94 | norm 0.4032 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15733 | train 2.956385 | val 3.333140 | tok/s 2629927.05 | norm 0.3773 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15734 | train 2.967357 | val 3.333140 | tok/s 2623782.16 | norm 0.3600 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15735 | train 2.972571 | val 3.333140 | tok/s 2628617.33 | norm 0.3844 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15736 | train 2.900819 | val 3.333140 | tok/s 2628524.61 | norm 0.3921 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15737 | train 2.993004 | val 3.333140 | tok/s 2592599.12 | norm 0.4055 | dt 202.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15738 | train 3.059230 | val 3.333140 | tok/s 2631533.17 | norm 0.4145 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15739 | train 2.871412 | val 3.333140 | tok/s 2627092.45 | norm 0.3647 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15740 | train 2.895277 | val 3.333140 | tok/s 2627432.06 | norm 0.3763 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15741 | train 3.004865 | val 3.333140 | tok/s 2627463.79 | norm 0.3926 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15742 | train 2.965877 | val 3.333140 | tok/s 2628932.91 | norm 0.3757 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15743 | train 3.021351 | val 3.333140 | tok/s 2630298.31 | norm 0.4146 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15744 | train 2.907583 | val 3.333140 | tok/s 2631600.61 | norm 0.3679 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15745 | train 2.958279 | val 3.333140 | tok/s 2630613.03 | norm 0.3791 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15746 | train 3.007750 | val 3.333140 | tok/s 2602122.38 | norm 0.4183 | dt 201.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15747 | train 3.004283 | val 3.333140 | tok/s 2622547.00 | norm 0.3948 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15748 | train 3.024994 | val 3.333140 | tok/s 2628583.60 | norm 0.3407 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15749 | train 2.889834 | val 3.333140 | tok/s 2627557.86 | norm 0.3387 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 15750 | val_loss 3.333155 step 15750 | train 2.883380 | val 3.333155 | tok/s 2600754.86 | norm 0.3669 | dt 201.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15751 | train 2.916644 | val 3.333155 | tok/s 2625269.73 | norm 0.4024 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15752 | train 2.921404 | val 3.333155 | tok/s 2625025.82 | norm 0.3780 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15753 | train 3.017136 | val 3.333155 | tok/s 2616891.75 | norm 0.4173 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15754 | train 3.020598 | val 3.333155 | tok/s 2629728.04 | norm 0.4207 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15755 | train 2.926154 | val 3.333155 | tok/s 2498493.41 | norm 0.3502 | dt 209.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15756 | train 2.909364 | val 3.333155 | tok/s 2496282.81 | norm 0.3718 | dt 210.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15757 | train 3.004504 | val 3.333155 | tok/s 2622472.03 | norm 0.4086 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15758 | train 2.883088 | val 3.333155 | tok/s 2627325.79 | norm 0.3906 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15759 | train 2.978988 | val 3.333155 | tok/s 2623015.99 | norm 0.4211 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15760 | train 2.861329 | val 3.333155 | tok/s 2624266.85 | norm 0.3868 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15761 | train 2.909136 | val 3.333155 | tok/s 2626724.27 | norm 0.3802 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15762 | train 2.974775 | val 3.333155 | tok/s 2619670.06 | norm 0.4049 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15763 | train 2.990405 | val 3.333155 | tok/s 2610053.74 | norm 0.3928 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15764 | train 3.141824 | val 3.333155 | tok/s 2628509.43 | norm 0.4158 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15765 | train 3.045017 | val 3.333155 | tok/s 2624568.13 | norm 0.4117 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15766 | train 3.057511 | val 3.333155 | tok/s 2631121.37 | norm 0.4181 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15767 | train 2.940332 | val 3.333155 | tok/s 2627077.21 | norm 0.4108 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15768 | train 3.094435 | val 3.333155 | tok/s 2615683.61 | norm 0.3943 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15769 | train 2.923105 | val 3.333155 | tok/s 2620703.32 | norm 0.3806 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15770 | train 2.943594 | val 3.333155 | tok/s 2623377.47 | norm 0.3634 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15771 | train 2.931075 | val 3.333155 | tok/s 2622864.31 | norm 0.3710 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15772 | train 2.971310 | val 3.333155 | tok/s 2619823.02 | norm 0.3634 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15773 | train 2.982940 | val 3.333155 | tok/s 2622008.18 | norm 0.3863 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15774 | train 2.971713 | val 3.333155 | tok/s 2627726.88 | norm 0.3796 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15775 | train 2.908652 | val 3.333155 | tok/s 2579131.27 | norm 0.3908 | dt 203.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15776 | train 2.989037 | val 3.333155 | tok/s 2622852.60 | norm 0.3935 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15777 | train 2.911166 | val 3.333155 | tok/s 2626885.53 | norm 0.3808 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15778 | train 2.990889 | val 3.333155 | tok/s 2629342.01 | norm 0.3919 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15779 | train 3.055690 | val 3.333155 | tok/s 2629691.49 | norm 0.3915 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15780 | train 2.913186 | val 3.333155 | tok/s 2626084.97 | norm 0.3835 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15781 | train 2.972698 | val 3.333155 | tok/s 2629178.16 | norm 0.3689 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15782 | train 2.930776 | val 3.333155 | tok/s 2629445.43 | norm 0.3650 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15783 | train 2.894789 | val 3.333155 | tok/s 2627179.55 | norm 0.3646 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15784 | train 2.977142 | val 3.333155 | tok/s 2605263.39 | norm 0.3679 | dt 201.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15785 | train 3.024874 | val 3.333155 | tok/s 2629161.39 | norm 0.3716 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15786 | train 3.048270 | val 3.333155 | tok/s 2631046.48 | norm 0.3916 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15787 | train 2.908029 | val 3.333155 | tok/s 2631839.13 | norm 0.3845 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15788 | train 2.886169 | val 3.333155 | tok/s 2631254.42 | norm 0.3562 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15789 | train 2.984578 | val 3.333155 | tok/s 2629129.86 | norm 0.3609 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15790 | train 3.041948 | val 3.333155 | tok/s 2623201.87 | norm 0.3585 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15791 | train 2.910081 | val 3.333155 | tok/s 2630235.75 | norm 0.3745 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15792 | train 2.906467 | val 3.333155 | tok/s 2627448.65 | norm 0.3586 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15793 | train 2.950843 | val 3.333155 | tok/s 2625619.87 | norm 0.3784 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15794 | train 2.914400 | val 3.333155 | tok/s 2621486.05 | norm 0.3881 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15795 | train 2.924883 | val 3.333155 | tok/s 2617228.89 | norm 0.3927 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15796 | train 2.927978 | val 3.333155 | tok/s 2627418.05 | norm 0.3919 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15797 | train 2.956517 | val 3.333155 | tok/s 2511208.43 | norm 0.3915 | dt 208.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15798 | train 2.897308 | val 3.333155 | tok/s 2613951.26 | norm 0.3739 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15799 | train 3.045686 | val 3.333155 | tok/s 2620897.00 | norm 0.3701 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15800 | train 2.933054 | val 3.333155 | tok/s 2617457.09 | norm 0.3574 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15801 | train 2.965065 | val 3.333155 | tok/s 2623766.09 | norm 0.3565 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15802 | train 2.937038 | val 3.333155 | tok/s 2622683.53 | norm 0.3992 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15803 | train 2.883451 | val 3.333155 | tok/s 2626780.18 | norm 0.4008 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15804 | train 2.897812 | val 3.333155 | tok/s 2625480.42 | norm 0.3711 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15805 | train 2.885448 | val 3.333155 | tok/s 2623533.25 | norm 0.3827 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15806 | train 3.015403 | val 3.333155 | tok/s 2624524.95 | norm 0.3697 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15807 | train 2.949479 | val 3.333155 | tok/s 2621850.56 | norm 0.3691 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15808 | train 2.954285 | val 3.333155 | tok/s 2626886.31 | norm 0.3823 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15809 | train 3.112232 | val 3.333155 | tok/s 2622343.59 | norm 0.3827 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15810 | train 2.933831 | val 3.333155 | tok/s 2619164.27 | norm 0.3684 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15811 | train 2.955909 | val 3.333155 | tok/s 2617239.06 | norm 0.3967 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15812 | train 2.991996 | val 3.333155 | tok/s 2598713.02 | norm 0.4115 | dt 201.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15813 | train 3.013394 | val 3.333155 | tok/s 2585335.07 | norm 0.3737 | dt 202.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15814 | train 3.055739 | val 3.333155 | tok/s 2622768.88 | norm 0.3693 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15815 | train 2.976042 | val 3.333155 | tok/s 2622466.88 | norm 0.3833 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15816 | train 2.953723 | val 3.333155 | tok/s 2621441.97 | norm 0.3664 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15817 | train 2.926762 | val 3.333155 | tok/s 2625307.55 | norm 0.3716 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15818 | train 2.840444 | val 3.333155 | tok/s 2625649.90 | norm 0.4148 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15819 | train 3.025715 | val 3.333155 | tok/s 2620944.81 | norm 0.4088 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15820 | train 3.009004 | val 3.333155 | tok/s 2621372.39 | norm 0.3960 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15821 | train 3.022507 | val 3.333155 | tok/s 2628251.60 | norm 0.3795 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15822 | train 2.923008 | val 3.333155 | tok/s 2624575.69 | norm 0.3822 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15823 | train 3.020244 | val 3.333155 | tok/s 2619968.74 | norm 0.4187 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15824 | train 3.148684 | val 3.333155 | tok/s 2621112.20 | norm 0.3904 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15825 | train 2.993469 | val 3.333155 | tok/s 2625393.95 | norm 0.3907 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15826 | train 2.962584 | val 3.333155 | tok/s 2625415.67 | norm 0.3807 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15827 | train 2.997354 | val 3.333155 | tok/s 2623065.04 | norm 0.3756 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15828 | train 3.026839 | val 3.333155 | tok/s 2626279.84 | norm 0.3896 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15829 | train 2.942896 | val 3.333155 | tok/s 2623400.74 | norm 0.3492 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15830 | train 3.000642 | val 3.333155 | tok/s 2621530.34 | norm 0.3636 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15831 | train 3.047446 | val 3.333155 | tok/s 2627303.37 | norm 0.3708 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15832 | train 2.993556 | val 3.333155 | tok/s 2625734.13 | norm 0.3623 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15833 | train 2.976749 | val 3.333155 | tok/s 2629428.69 | norm 0.3498 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15834 | train 2.926947 | val 3.333155 | tok/s 2623811.53 | norm 0.3542 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15835 | train 2.935722 | val 3.333155 | tok/s 2628335.23 | norm 0.3650 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15836 | train 2.949888 | val 3.333155 | tok/s 2627143.31 | norm 0.3727 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15837 | train 2.894057 | val 3.333155 | tok/s 2623013.60 | norm 0.3624 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15838 | train 2.877153 | val 3.333155 | tok/s 2619947.01 | norm 0.3886 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15839 | train 2.982933 | val 3.333155 | tok/s 2626021.71 | norm 0.3908 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15840 | train 2.960791 | val 3.333155 | tok/s 2625106.97 | norm 0.3910 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15841 | train 2.803257 | val 3.333155 | tok/s 2624830.52 | norm 0.3645 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15842 | train 2.975612 | val 3.333155 | tok/s 2618475.74 | norm 0.3686 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15843 | train 2.893625 | val 3.333155 | tok/s 2623098.03 | norm 0.3605 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15844 | train 2.903777 | val 3.333155 | tok/s 2617536.51 | norm 0.3685 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15845 | train 2.889910 | val 3.333155 | tok/s 2622764.84 | norm 0.4044 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15846 | train 2.927109 | val 3.333155 | tok/s 2621485.88 | norm 0.3838 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15847 | train 2.946917 | val 3.333155 | tok/s 2619432.59 | norm 0.3732 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15848 | train 2.988071 | val 3.333155 | tok/s 2624279.77 | norm 0.3727 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15849 | train 2.979419 | val 3.333155 | tok/s 2623014.85 | norm 0.3743 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15850 | train 3.077547 | val 3.333155 | tok/s 2608799.79 | norm 0.4575 | dt 200.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15851 | train 2.942617 | val 3.333155 | tok/s 2624964.99 | norm 0.3802 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15852 | train 2.889050 | val 3.333155 | tok/s 2627295.51 | norm 0.3830 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15853 | train 2.918228 | val 3.333155 | tok/s 2625654.66 | norm 0.4028 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15854 | train 2.852637 | val 3.333155 | tok/s 2622972.55 | norm 0.3735 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15855 | train 3.000679 | val 3.333155 | tok/s 2627269.25 | norm 0.3757 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15856 | train 2.930450 | val 3.333155 | tok/s 2621565.52 | norm 0.3721 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15857 | train 2.939514 | val 3.333155 | tok/s 2625860.93 | norm 0.3765 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15858 | train 2.843684 | val 3.333155 | tok/s 2542763.38 | norm 0.4332 | dt 206.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15859 | train 2.833347 | val 3.333155 | tok/s 2626994.43 | norm 0.4136 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15860 | train 3.013964 | val 3.333155 | tok/s 2626797.46 | norm 0.3691 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15861 | train 3.035939 | val 3.333155 | tok/s 2616030.39 | norm 0.4006 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15862 | train 2.940719 | val 3.333155 | tok/s 2622423.28 | norm 0.3899 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15863 | train 2.895440 | val 3.333155 | tok/s 2629645.86 | norm 0.3650 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15864 | train 2.929571 | val 3.333155 | tok/s 2625775.58 | norm 0.3729 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15865 | train 2.871444 | val 3.333155 | tok/s 2625934.84 | norm 0.3859 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15866 | train 2.971378 | val 3.333155 | tok/s 2620313.15 | norm 0.3917 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15867 | train 3.005667 | val 3.333155 | tok/s 2611208.79 | norm 0.3905 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15868 | train 2.992024 | val 3.333155 | tok/s 2617293.36 | norm 0.3646 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15869 | train 2.914587 | val 3.333155 | tok/s 2626801.67 | norm 0.3638 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15870 | train 2.922515 | val 3.333155 | tok/s 2631908.89 | norm 0.3862 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15871 | train 2.976539 | val 3.333155 | tok/s 2622757.71 | norm 0.3909 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15872 | train 2.968115 | val 3.333155 | tok/s 2626138.89 | norm 0.3637 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15873 | train 2.853015 | val 3.333155 | tok/s 2627646.11 | norm 0.3867 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15874 | train 2.842412 | val 3.333155 | tok/s 2478117.14 | norm 0.4061 | dt 211.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15875 | train 2.877208 | val 3.333155 | tok/s 2607640.50 | norm 0.3931 | dt 201.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15876 | train 3.014472 | val 3.333155 | tok/s 2627032.99 | norm 0.3905 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15877 | train 3.034114 | val 3.333155 | tok/s 2625635.60 | norm 0.3662 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15878 | train 3.036073 | val 3.333155 | tok/s 2631476.58 | norm 0.3440 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15879 | train 3.000226 | val 3.333155 | tok/s 2628636.93 | norm 0.3620 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15880 | train 2.969742 | val 3.333155 | tok/s 2629043.48 | norm 0.3448 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15881 | train 2.989578 | val 3.333155 | tok/s 2628482.76 | norm 0.4069 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15882 | train 2.966830 | val 3.333155 | tok/s 2624429.55 | norm 0.4178 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15883 | train 2.924869 | val 3.333155 | tok/s 2629572.44 | norm 0.3633 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15884 | train 3.075184 | val 3.333155 | tok/s 2625131.45 | norm 0.3660 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15885 | train 2.971884 | val 3.333155 | tok/s 2625855.88 | norm 0.3761 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15886 | train 3.124952 | val 3.333155 | tok/s 2629523.42 | norm 0.4118 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15887 | train 2.948435 | val 3.333155 | tok/s 2631930.65 | norm 0.3707 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15888 | train 2.971742 | val 3.333155 | tok/s 2630952.64 | norm 0.3486 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15889 | train 2.916386 | val 3.333155 | tok/s 2626099.59 | norm 0.4038 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15890 | train 3.072587 | val 3.333155 | tok/s 2629361.90 | norm 0.3747 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15891 | train 3.180743 | val 3.333155 | tok/s 2627306.37 | norm 0.4268 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15892 | train 3.057102 | val 3.333155 | tok/s 2626872.16 | norm 0.3828 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15893 | train 2.970718 | val 3.333155 | tok/s 2628712.44 | norm 0.3288 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15894 | train 2.951300 | val 3.333155 | tok/s 2626604.52 | norm 0.3896 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15895 | train 2.979214 | val 3.333155 | tok/s 2628649.37 | norm 0.4357 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15896 | train 2.958521 | val 3.333155 | tok/s 2631751.36 | norm 0.3981 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15897 | train 2.887038 | val 3.333155 | tok/s 2630867.28 | norm 0.4197 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15898 | train 2.987141 | val 3.333155 | tok/s 2631692.42 | norm 0.4179 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15899 | train 3.075408 | val 3.333155 | tok/s 2624190.89 | norm 0.3689 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15900 | train 2.906426 | val 3.333155 | tok/s 2628950.48 | norm 0.4132 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15901 | train 2.998988 | val 3.333155 | tok/s 2626862.09 | norm 0.3622 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15902 | train 2.957630 | val 3.333155 | tok/s 2630799.66 | norm 0.3315 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15903 | train 2.957444 | val 3.333155 | tok/s 2629526.95 | norm 0.3505 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15904 | train 2.885499 | val 3.333155 | tok/s 2624993.90 | norm 0.3706 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15905 | train 2.830079 | val 3.333155 | tok/s 2625989.42 | norm 0.3960 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15906 | train 2.837243 | val 3.333155 | tok/s 2622887.76 | norm 0.3691 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15907 | train 3.047165 | val 3.333155 | tok/s 2630286.69 | norm 0.3507 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15908 | train 2.994238 | val 3.333155 | tok/s 2627617.23 | norm 0.3780 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15909 | train 3.043674 | val 3.333155 | tok/s 2625906.07 | norm 0.3897 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15910 | train 2.882340 | val 3.333155 | tok/s 2625985.16 | norm 0.3677 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15911 | train 3.089485 | val 3.333155 | tok/s 2482082.45 | norm 0.4058 | dt 211.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15912 | train 3.027851 | val 3.333155 | tok/s 2626640.39 | norm 0.3803 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15913 | train 2.976211 | val 3.333155 | tok/s 2625690.52 | norm 0.3708 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15914 | train 3.054018 | val 3.333155 | tok/s 2626240.31 | norm 0.4079 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15915 | train 2.890843 | val 3.333155 | tok/s 2633909.22 | norm 0.3982 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15916 | train 2.980817 | val 3.333155 | tok/s 2630795.29 | norm 0.3721 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15917 | train 2.992375 | val 3.333155 | tok/s 2630995.82 | norm 0.3536 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15918 | train 2.976309 | val 3.333155 | tok/s 2627944.48 | norm 0.3935 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15919 | train 2.964334 | val 3.333155 | tok/s 2629022.11 | norm 0.3602 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15920 | train 3.100135 | val 3.333155 | tok/s 2629981.64 | norm 0.3675 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15921 | train 3.134368 | val 3.333155 | tok/s 2628084.82 | norm 0.4128 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15922 | train 2.993907 | val 3.333155 | tok/s 2630598.05 | norm 0.3531 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15923 | train 2.867255 | val 3.333155 | tok/s 2627480.18 | norm 0.3580 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15924 | train 2.899703 | val 3.333155 | tok/s 2625730.67 | norm 0.4033 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15925 | train 2.994370 | val 3.333155 | tok/s 2628766.91 | norm 0.4243 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15926 | train 2.920762 | val 3.333155 | tok/s 2627709.29 | norm 0.3767 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15927 | train 2.944620 | val 3.333155 | tok/s 2628615.08 | norm 0.4060 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15928 | train 2.930294 | val 3.333155 | tok/s 2625531.26 | norm 0.4017 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15929 | train 2.971482 | val 3.333155 | tok/s 2626790.11 | norm 0.3567 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15930 | train 2.861304 | val 3.333155 | tok/s 2628591.03 | norm 0.3870 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15931 | train 2.854611 | val 3.333155 | tok/s 2630179.18 | norm 0.3911 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15932 | train 2.900556 | val 3.333155 | tok/s 2628125.40 | norm 0.3646 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15933 | train 2.883824 | val 3.333155 | tok/s 2626096.08 | norm 0.4192 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15934 | train 2.865288 | val 3.333155 | tok/s 2626149.47 | norm 0.4160 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15935 | train 2.989814 | val 3.333155 | tok/s 2631469.82 | norm 0.3780 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15936 | train 2.875389 | val 3.333155 | tok/s 2632578.64 | norm 0.3921 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15937 | train 2.974606 | val 3.333155 | tok/s 2626547.08 | norm 0.3716 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15938 | train 2.911672 | val 3.333155 | tok/s 2625621.60 | norm 0.3450 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15939 | train 3.002563 | val 3.333155 | tok/s 2627282.62 | norm 0.3328 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15940 | train 2.811721 | val 3.333155 | tok/s 2629744.56 | norm 0.3720 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15941 | train 3.042089 | val 3.333155 | tok/s 2625293.88 | norm 0.3637 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15942 | train 2.968584 | val 3.333155 | tok/s 2632663.14 | norm 0.3780 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15943 | train 2.906780 | val 3.333155 | tok/s 2625832.59 | norm 0.3951 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15944 | train 2.877993 | val 3.333155 | tok/s 2634153.00 | norm 0.4047 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15945 | train 3.036808 | val 3.333155 | tok/s 2630590.65 | norm 0.3786 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15946 | train 2.949896 | val 3.333155 | tok/s 2623858.44 | norm 0.3861 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15947 | train 2.844191 | val 3.333155 | tok/s 2626612.76 | norm 0.5229 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15948 | train 2.908530 | val 3.333155 | tok/s 2629628.39 | norm 0.4252 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15949 | train 2.883785 | val 3.333155 | tok/s 2633822.34 | norm 0.3612 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15950 | train 2.967730 | val 3.333155 | tok/s 2620782.98 | norm 0.4010 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15951 | train 2.910634 | val 3.333155 | tok/s 2600745.99 | norm 0.3891 | dt 201.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15952 | train 2.938049 | val 3.333155 | tok/s 2589977.45 | norm 0.3787 | dt 202.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15953 | train 2.928238 | val 3.333155 | tok/s 2626530.90 | norm 0.3777 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15954 | train 2.946242 | val 3.333155 | tok/s 2630332.17 | norm 0.3792 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15955 | train 2.857768 | val 3.333155 | tok/s 2618876.84 | norm 0.3794 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15956 | train 2.920625 | val 3.333155 | tok/s 2621343.14 | norm 0.3823 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15957 | train 3.023386 | val 3.333155 | tok/s 2628385.93 | norm 0.3722 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15958 | train 2.952267 | val 3.333155 | tok/s 2627404.31 | norm 0.3548 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15959 | train 3.002138 | val 3.333155 | tok/s 2629135.94 | norm 0.3754 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15960 | train 2.984022 | val 3.333155 | tok/s 2617396.78 | norm 0.3801 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15961 | train 2.836878 | val 3.333155 | tok/s 2625562.58 | norm 0.3716 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15962 | train 3.031090 | val 3.333155 | tok/s 2627950.14 | norm 0.3912 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15963 | train 2.937232 | val 3.333155 | tok/s 2624573.86 | norm 0.4106 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15964 | train 2.914971 | val 3.333155 | tok/s 2630621.32 | norm 0.3973 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15965 | train 2.816882 | val 3.333155 | tok/s 2460486.23 | norm 0.3961 | dt 213.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15966 | train 2.817854 | val 3.333155 | tok/s 2626452.47 | norm 0.3798 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15967 | train 3.028717 | val 3.333155 | tok/s 2607272.15 | norm 0.3911 | dt 201.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15968 | train 2.988622 | val 3.333155 | tok/s 2633424.72 | norm 0.3855 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15969 | train 2.898585 | val 3.333155 | tok/s 2636016.29 | norm 0.4005 | dt 198.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15970 | train 2.904042 | val 3.333155 | tok/s 2629378.71 | norm 0.3868 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15971 | train 2.939812 | val 3.333155 | tok/s 2629607.92 | norm 0.4084 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15972 | train 2.956267 | val 3.333155 | tok/s 2634022.91 | norm 0.3804 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15973 | train 3.199967 | val 3.333155 | tok/s 2627270.66 | norm 0.4078 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15974 | train 2.854539 | val 3.333155 | tok/s 2630042.92 | norm 0.3576 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15975 | train 2.927548 | val 3.333155 | tok/s 2623367.01 | norm 0.3668 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15976 | train 2.956694 | val 3.333155 | tok/s 2625598.36 | norm 0.3866 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15977 | train 2.879427 | val 3.333155 | tok/s 2631353.28 | norm 0.3659 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15978 | train 2.968255 | val 3.333155 | tok/s 2627236.97 | norm 0.3821 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15979 | train 3.079451 | val 3.333155 | tok/s 2628857.94 | norm 0.3724 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15980 | train 2.902122 | val 3.333155 | tok/s 2626561.19 | norm 0.3648 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15981 | train 2.859329 | val 3.333155 | tok/s 2629934.31 | norm 0.3597 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15982 | train 2.987356 | val 3.333155 | tok/s 2628361.11 | norm 0.3823 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15983 | train 2.990515 | val 3.333155 | tok/s 2626113.67 | norm 0.3933 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15984 | train 2.906842 | val 3.333155 | tok/s 2618596.92 | norm 0.4063 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15985 | train 2.951377 | val 3.333155 | tok/s 2620575.17 | norm 0.3694 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15986 | train 2.951046 | val 3.333155 | tok/s 2628213.34 | norm 0.3358 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15987 | train 2.931405 | val 3.333155 | tok/s 2630004.23 | norm 0.3741 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15988 | train 2.937507 | val 3.333155 | tok/s 2624178.70 | norm 0.3511 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15989 | train 3.015158 | val 3.333155 | tok/s 2629638.71 | norm 0.3208 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15990 | train 2.918031 | val 3.333155 | tok/s 2629055.97 | norm 0.3545 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15991 | train 2.989714 | val 3.333155 | tok/s 2632369.88 | norm 0.4031 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15992 | train 3.059291 | val 3.333155 | tok/s 2623712.87 | norm 0.3704 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15993 | train 2.812155 | val 3.333155 | tok/s 2621903.49 | norm 0.3746 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15994 | train 3.047269 | val 3.333155 | tok/s 2626530.94 | norm 0.4066 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15995 | train 3.014069 | val 3.333155 | tok/s 2629869.84 | norm 0.3728 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15996 | train 2.959393 | val 3.333155 | tok/s 2627395.16 | norm 0.3807 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15997 | train 2.910596 | val 3.333155 | tok/s 2630098.01 | norm 0.3789 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15998 | train 2.965829 | val 3.333155 | tok/s 2631013.56 | norm 0.3361 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 15999 | train 2.953630 | val 3.333155 | tok/s 2606765.80 | norm 0.3784 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 16000 | val_loss 3.323671 step 16000 | train 2.951974 | val 3.323671 | tok/s 2611139.34 | norm 0.4006 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16001 | train 2.927798 | val 3.323671 | tok/s 2627703.05 | norm 0.3895 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16002 | train 3.074480 | val 3.323671 | tok/s 2629565.63 | norm 0.4329 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16003 | train 2.944996 | val 3.323671 | tok/s 2629345.28 | norm 0.3577 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16004 | train 3.089486 | val 3.323671 | tok/s 2619537.44 | norm 0.3654 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16005 | train 2.931568 | val 3.323671 | tok/s 2627971.22 | norm 0.4036 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16006 | train 3.010032 | val 3.323671 | tok/s 2627534.01 | norm 0.3874 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16007 | train 2.910446 | val 3.323671 | tok/s 2630170.79 | norm 0.4426 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16008 | train 2.834379 | val 3.323671 | tok/s 2627340.36 | norm 0.3814 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16009 | train 2.983853 | val 3.323671 | tok/s 2624541.11 | norm 0.3668 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16010 | train 2.840337 | val 3.323671 | tok/s 2629928.96 | norm 0.4127 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16011 | train 2.937476 | val 3.323671 | tok/s 2626475.78 | norm 0.3509 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16012 | train 2.934287 | val 3.323671 | tok/s 2625566.75 | norm 0.3630 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16013 | train 3.096266 | val 3.323671 | tok/s 2627004.11 | norm 0.4046 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16014 | train 2.941589 | val 3.323671 | tok/s 2627775.28 | norm 0.3611 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16015 | train 3.006152 | val 3.323671 | tok/s 2629288.19 | norm 0.3805 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16016 | train 3.085222 | val 3.323671 | tok/s 2624941.61 | norm 0.3798 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16017 | train 2.932322 | val 3.323671 | tok/s 2628437.70 | norm 0.3950 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16018 | train 3.015180 | val 3.323671 | tok/s 2617383.34 | norm 0.3979 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16019 | train 3.016983 | val 3.323671 | tok/s 2623042.47 | norm 0.3429 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16020 | train 3.031456 | val 3.323671 | tok/s 2627483.30 | norm 0.3523 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16021 | train 2.960655 | val 3.323671 | tok/s 2626926.39 | norm 0.3792 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16022 | train 3.057192 | val 3.323671 | tok/s 2630154.11 | norm 0.4121 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16023 | train 2.910334 | val 3.323671 | tok/s 2488121.48 | norm 0.4016 | dt 210.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16024 | train 2.948956 | val 3.323671 | tok/s 2625131.58 | norm 0.3706 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16025 | train 2.903926 | val 3.323671 | tok/s 2630719.52 | norm 0.3783 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16026 | train 2.892108 | val 3.323671 | tok/s 2627624.01 | norm 0.3918 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16027 | train 3.031640 | val 3.323671 | tok/s 2633189.79 | norm 0.5243 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16028 | train 2.860137 | val 3.323671 | tok/s 2623016.55 | norm 0.3959 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16029 | train 2.999420 | val 3.323671 | tok/s 2629682.03 | norm 0.3482 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16030 | train 2.859481 | val 3.323671 | tok/s 2631106.36 | norm 0.4215 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16031 | train 2.979926 | val 3.323671 | tok/s 2629373.29 | norm 0.3938 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16032 | train 2.881123 | val 3.323671 | tok/s 2610365.39 | norm 0.3634 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16033 | train 2.939451 | val 3.323671 | tok/s 2621533.20 | norm 0.4043 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16034 | train 2.820164 | val 3.323671 | tok/s 2629956.98 | norm 0.4102 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16035 | train 2.970333 | val 3.323671 | tok/s 2628454.78 | norm 0.3706 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16036 | train 3.004867 | val 3.323671 | tok/s 2554907.01 | norm 0.3734 | dt 205.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16037 | train 2.808219 | val 3.323671 | tok/s 2629260.58 | norm 0.3782 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16038 | train 3.009507 | val 3.323671 | tok/s 2632423.71 | norm 0.4286 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16039 | train 2.938400 | val 3.323671 | tok/s 2629708.32 | norm 0.3766 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16040 | train 2.855506 | val 3.323671 | tok/s 2628421.45 | norm 0.3728 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16041 | train 3.019052 | val 3.323671 | tok/s 2629375.14 | norm 0.4053 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16042 | train 2.966379 | val 3.323671 | tok/s 2628528.18 | norm 0.4044 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16043 | train 2.922850 | val 3.323671 | tok/s 2625134.67 | norm 0.3521 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16044 | train 2.902434 | val 3.323671 | tok/s 2626769.62 | norm 0.3601 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16045 | train 3.071769 | val 3.323671 | tok/s 2625042.20 | norm 0.3739 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16046 | train 2.911504 | val 3.323671 | tok/s 2624497.22 | norm 0.3726 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16047 | train 2.986339 | val 3.323671 | tok/s 2611767.28 | norm 0.3776 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16048 | train 2.912701 | val 3.323671 | tok/s 2623606.52 | norm 0.3657 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16049 | train 2.860713 | val 3.323671 | tok/s 2629692.55 | norm 0.3634 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16050 | train 2.912897 | val 3.323671 | tok/s 2475925.84 | norm 0.3552 | dt 211.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16051 | train 3.198718 | val 3.323671 | tok/s 2627845.21 | norm 0.4241 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16052 | train 2.910490 | val 3.323671 | tok/s 2618771.51 | norm 0.5749 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16053 | train 3.083479 | val 3.323671 | tok/s 2623648.90 | norm 0.3746 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16054 | train 3.010165 | val 3.323671 | tok/s 2630453.79 | norm 0.3808 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16055 | train 3.062239 | val 3.323671 | tok/s 2631141.16 | norm 0.3919 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16056 | train 3.048526 | val 3.323671 | tok/s 2629434.33 | norm 0.3867 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16057 | train 2.955527 | val 3.323671 | tok/s 2626148.76 | norm 0.3984 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16058 | train 2.943408 | val 3.323671 | tok/s 2624974.16 | norm 0.3592 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16059 | train 2.885123 | val 3.323671 | tok/s 2628512.50 | norm 0.3568 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16060 | train 2.900758 | val 3.323671 | tok/s 2624574.31 | norm 0.3619 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16061 | train 2.931568 | val 3.323671 | tok/s 2628973.02 | norm 0.3754 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16062 | train 2.966917 | val 3.323671 | tok/s 2626317.33 | norm 0.3493 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16063 | train 3.174384 | val 3.323671 | tok/s 2626379.18 | norm 0.4448 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16064 | train 2.927786 | val 3.323671 | tok/s 2630368.96 | norm 0.4056 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16065 | train 2.944488 | val 3.323671 | tok/s 2625297.18 | norm 0.3577 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16066 | train 3.163413 | val 3.323671 | tok/s 2624762.29 | norm 0.3907 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16067 | train 2.988945 | val 3.323671 | tok/s 2625902.13 | norm 0.3935 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16068 | train 3.101946 | val 3.323671 | tok/s 2624086.22 | norm 0.3495 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16069 | train 2.983629 | val 3.323671 | tok/s 2627813.54 | norm 0.3683 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16070 | train 3.052689 | val 3.323671 | tok/s 2626925.49 | norm 0.3679 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16071 | train 2.939670 | val 3.323671 | tok/s 2628118.47 | norm 0.3782 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16072 | train 2.922926 | val 3.323671 | tok/s 2627172.64 | norm 0.3856 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16073 | train 3.097755 | val 3.323671 | tok/s 2630418.53 | norm 0.3841 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16074 | train 2.961123 | val 3.323671 | tok/s 2629815.57 | norm 0.3592 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16075 | train 2.884869 | val 3.323671 | tok/s 2624980.19 | norm 0.3616 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16076 | train 2.880164 | val 3.323671 | tok/s 2627351.49 | norm 0.3840 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16077 | train 2.975552 | val 3.323671 | tok/s 2599293.33 | norm 0.3907 | dt 201.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16078 | train 2.988271 | val 3.323671 | tok/s 2600397.89 | norm 0.3912 | dt 201.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16079 | train 2.875283 | val 3.323671 | tok/s 2625306.42 | norm 0.4016 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16080 | train 2.949978 | val 3.323671 | tok/s 2629066.61 | norm 0.4059 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16081 | train 3.071316 | val 3.323671 | tok/s 2628744.73 | norm 0.3932 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16082 | train 2.997326 | val 3.323671 | tok/s 2624707.30 | norm 0.3678 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16083 | train 3.077560 | val 3.323671 | tok/s 2628674.12 | norm 0.3199 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16084 | train 2.889577 | val 3.323671 | tok/s 2625156.79 | norm 0.3430 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16085 | train 2.925756 | val 3.323671 | tok/s 2627128.48 | norm 0.3606 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16086 | train 2.944571 | val 3.323671 | tok/s 2630018.38 | norm 0.3640 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16087 | train 2.927252 | val 3.323671 | tok/s 2631365.26 | norm 0.4026 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16088 | train 2.939091 | val 3.323671 | tok/s 2629756.64 | norm 0.3750 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16089 | train 2.936026 | val 3.323671 | tok/s 2630635.99 | norm 0.3707 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16090 | train 2.952474 | val 3.323671 | tok/s 2600481.68 | norm 0.3951 | dt 201.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16091 | train 2.989542 | val 3.323671 | tok/s 2631716.11 | norm 0.3598 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16092 | train 2.890190 | val 3.323671 | tok/s 2626290.93 | norm 0.3586 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16093 | train 3.015292 | val 3.323671 | tok/s 2628579.68 | norm 0.3987 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16094 | train 3.075911 | val 3.323671 | tok/s 2625096.39 | norm 0.3918 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16095 | train 2.888414 | val 3.323671 | tok/s 2620327.09 | norm 0.3941 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16096 | train 2.904819 | val 3.323671 | tok/s 2631481.07 | norm 0.3595 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16097 | train 2.978495 | val 3.323671 | tok/s 2627042.26 | norm 0.3575 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16098 | train 2.922186 | val 3.323671 | tok/s 2630174.32 | norm 0.3606 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16099 | train 3.003041 | val 3.323671 | tok/s 2622155.94 | norm 0.3544 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16100 | train 2.930803 | val 3.323671 | tok/s 2625274.87 | norm 0.3655 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16101 | train 2.968805 | val 3.323671 | tok/s 2626594.67 | norm 0.3706 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16102 | train 3.036876 | val 3.323671 | tok/s 2628132.88 | norm 0.3741 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16103 | train 2.996388 | val 3.323671 | tok/s 2628248.12 | norm 0.3708 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16104 | train 2.928212 | val 3.323671 | tok/s 2627268.45 | norm 0.4087 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16105 | train 2.953267 | val 3.323671 | tok/s 2629464.03 | norm 0.3856 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16106 | train 3.011653 | val 3.323671 | tok/s 2629093.67 | norm 0.3706 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16107 | train 3.022461 | val 3.323671 | tok/s 2627427.87 | norm 0.3689 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16108 | train 2.891023 | val 3.323671 | tok/s 2630919.47 | norm 0.3645 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16109 | train 2.912395 | val 3.323671 | tok/s 2626022.36 | norm 0.3616 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16110 | train 2.949233 | val 3.323671 | tok/s 2628083.28 | norm 0.3739 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16111 | train 3.015833 | val 3.323671 | tok/s 2631661.10 | norm 0.3916 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16112 | train 2.965988 | val 3.323671 | tok/s 2629213.19 | norm 0.3873 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16113 | train 2.932078 | val 3.323671 | tok/s 2630066.99 | norm 0.3491 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16114 | train 2.878631 | val 3.323671 | tok/s 2626598.77 | norm 0.3314 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16115 | train 2.884890 | val 3.323671 | tok/s 2627913.68 | norm 0.3606 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16116 | train 2.941097 | val 3.323671 | tok/s 2628314.33 | norm 0.4246 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16117 | train 2.977568 | val 3.323671 | tok/s 2627450.65 | norm 0.3876 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16118 | train 2.950654 | val 3.323671 | tok/s 2627437.27 | norm 0.4071 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16119 | train 2.913791 | val 3.323671 | tok/s 2610167.22 | norm 0.4222 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16120 | train 2.851340 | val 3.323671 | tok/s 2625755.03 | norm 0.3786 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16121 | train 2.920766 | val 3.323671 | tok/s 2631316.87 | norm 0.3390 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16122 | train 2.999202 | val 3.323671 | tok/s 2627441.21 | norm 0.3792 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16123 | train 2.922767 | val 3.323671 | tok/s 2630202.04 | norm 0.3920 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16124 | train 2.887303 | val 3.323671 | tok/s 2630560.34 | norm 0.4131 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16125 | train 2.876305 | val 3.323671 | tok/s 2627838.00 | norm 0.3899 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16126 | train 2.891720 | val 3.323671 | tok/s 2627115.70 | norm 0.4042 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16127 | train 2.936576 | val 3.323671 | tok/s 2626299.71 | norm 0.4014 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16128 | train 2.933315 | val 3.323671 | tok/s 2630006.03 | norm 0.4587 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16129 | train 3.038480 | val 3.323671 | tok/s 2626788.22 | norm 0.3833 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16130 | train 3.040897 | val 3.323671 | tok/s 2629058.17 | norm 0.3513 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16131 | train 3.027622 | val 3.323671 | tok/s 2628171.63 | norm 0.3815 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16132 | train 2.965925 | val 3.323671 | tok/s 2626347.37 | norm 0.3926 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16133 | train 2.837117 | val 3.323671 | tok/s 2626863.17 | norm 0.3549 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16134 | train 3.006179 | val 3.323671 | tok/s 2626868.37 | norm 0.3829 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16135 | train 2.963512 | val 3.323671 | tok/s 2629557.00 | norm 0.3917 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16136 | train 2.801569 | val 3.323671 | tok/s 2627224.57 | norm 0.4081 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16137 | train 2.954542 | val 3.323671 | tok/s 2627615.08 | norm 0.4035 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16138 | train 2.943282 | val 3.323671 | tok/s 2539564.70 | norm 0.3666 | dt 206.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16139 | train 2.967219 | val 3.323671 | tok/s 2629893.82 | norm 0.3863 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16140 | train 2.932917 | val 3.323671 | tok/s 2631202.96 | norm 0.4011 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16141 | train 2.966279 | val 3.323671 | tok/s 2620958.24 | norm 0.4529 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16142 | train 2.899632 | val 3.323671 | tok/s 2624759.88 | norm 0.3827 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16143 | train 2.959611 | val 3.323671 | tok/s 2629799.92 | norm 0.3695 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16144 | train 2.867287 | val 3.323671 | tok/s 2628677.88 | norm 0.4067 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16145 | train 2.923489 | val 3.323671 | tok/s 2630336.80 | norm 0.3526 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16146 | train 2.939778 | val 3.323671 | tok/s 2478645.12 | norm 0.3657 | dt 211.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16147 | train 2.848566 | val 3.323671 | tok/s 2630227.96 | norm 0.3993 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16148 | train 2.982076 | val 3.323671 | tok/s 2633007.05 | norm 0.4254 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16149 | train 2.973244 | val 3.323671 | tok/s 2631325.90 | norm 0.3946 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16150 | train 2.970358 | val 3.323671 | tok/s 2630926.11 | norm 0.4019 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16151 | train 2.892552 | val 3.323671 | tok/s 2631682.81 | norm 0.3730 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16152 | train 2.954267 | val 3.323671 | tok/s 2632351.31 | norm 0.3555 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16153 | train 2.901985 | val 3.323671 | tok/s 2628605.04 | norm 0.3422 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16154 | train 3.004291 | val 3.323671 | tok/s 2626978.38 | norm 0.3430 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16155 | train 2.960876 | val 3.323671 | tok/s 2630070.63 | norm 0.3559 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16156 | train 2.933628 | val 3.323671 | tok/s 2629162.52 | norm 0.3776 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16157 | train 2.892704 | val 3.323671 | tok/s 2623560.38 | norm 0.3742 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16158 | train 2.933214 | val 3.323671 | tok/s 2630095.34 | norm 0.3630 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16159 | train 2.927050 | val 3.323671 | tok/s 2632744.31 | norm 0.3676 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16160 | train 3.012554 | val 3.323671 | tok/s 2629266.31 | norm 0.3800 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16161 | train 3.030121 | val 3.323671 | tok/s 2624558.33 | norm 0.3995 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16162 | train 2.989702 | val 3.323671 | tok/s 2629347.92 | norm 0.4365 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16163 | train 2.872724 | val 3.323671 | tok/s 2627959.35 | norm 0.4038 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16164 | train 2.887503 | val 3.323671 | tok/s 2627102.73 | norm 0.3686 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16165 | train 2.928947 | val 3.323671 | tok/s 2628862.15 | norm 0.4572 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16166 | train 3.057342 | val 3.323671 | tok/s 2627641.97 | norm 0.3373 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16167 | train 2.890441 | val 3.323671 | tok/s 2628362.74 | norm 0.3477 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16168 | train 3.121304 | val 3.323671 | tok/s 2628879.26 | norm 0.4030 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16169 | train 2.858920 | val 3.323671 | tok/s 2626096.75 | norm 0.3470 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16170 | train 2.926519 | val 3.323671 | tok/s 2623028.77 | norm 0.3176 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16171 | train 2.943140 | val 3.323671 | tok/s 2631750.87 | norm 0.3721 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16172 | train 3.020514 | val 3.323671 | tok/s 2630653.01 | norm 0.3811 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16173 | train 3.113267 | val 3.323671 | tok/s 2632298.48 | norm 0.3833 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16174 | train 2.836149 | val 3.323671 | tok/s 2622262.34 | norm 0.3665 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16175 | train 2.963994 | val 3.323671 | tok/s 2631284.33 | norm 0.3748 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16176 | train 3.019579 | val 3.323671 | tok/s 2628578.27 | norm 0.3950 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16177 | train 2.901374 | val 3.323671 | tok/s 2630393.00 | norm 0.3885 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16178 | train 2.977767 | val 3.323671 | tok/s 2631452.55 | norm 0.4128 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16179 | train 2.959595 | val 3.323671 | tok/s 2626906.51 | norm 0.3999 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16180 | train 3.074482 | val 3.323671 | tok/s 2630808.24 | norm 0.4248 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16181 | train 2.939943 | val 3.323671 | tok/s 2626458.76 | norm 0.3696 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16182 | train 2.896477 | val 3.323671 | tok/s 2626732.52 | norm 0.3664 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16183 | train 2.983059 | val 3.323671 | tok/s 2630996.20 | norm 0.4296 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16184 | train 2.874520 | val 3.323671 | tok/s 2629198.69 | norm 0.3843 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16185 | train 2.902467 | val 3.323671 | tok/s 2632161.33 | norm 0.3734 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16186 | train 3.035337 | val 3.323671 | tok/s 2627562.40 | norm 0.3662 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16187 | train 3.034671 | val 3.323671 | tok/s 2631350.43 | norm 0.3446 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16188 | train 3.003070 | val 3.323671 | tok/s 2434081.10 | norm 0.3565 | dt 215.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16189 | train 3.090671 | val 3.323671 | tok/s 2618092.91 | norm 0.3983 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16190 | train 2.968583 | val 3.323671 | tok/s 2629642.34 | norm 0.3662 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16191 | train 2.998853 | val 3.323671 | tok/s 2629730.48 | norm 0.3617 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16192 | train 2.939140 | val 3.323671 | tok/s 2630509.35 | norm 0.3773 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16193 | train 3.030583 | val 3.323671 | tok/s 2623486.02 | norm 0.4020 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16194 | train 2.993449 | val 3.323671 | tok/s 2616745.22 | norm 0.3965 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16195 | train 2.994088 | val 3.323671 | tok/s 2628164.21 | norm 0.4003 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16196 | train 2.914159 | val 3.323671 | tok/s 2628444.78 | norm 0.3816 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16197 | train 2.925838 | val 3.323671 | tok/s 2633254.45 | norm 0.3889 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16198 | train 2.986631 | val 3.323671 | tok/s 2624760.30 | norm 0.3826 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16199 | train 2.908324 | val 3.323671 | tok/s 2621876.90 | norm 0.3877 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16200 | train 2.928767 | val 3.323671 | tok/s 2632589.42 | norm 0.3848 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16201 | train 2.912815 | val 3.323671 | tok/s 2629420.04 | norm 0.3701 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16202 | train 3.147141 | val 3.323671 | tok/s 2630692.66 | norm 0.4352 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16203 | train 2.958089 | val 3.323671 | tok/s 2627226.29 | norm 0.4491 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16204 | train 3.086587 | val 3.323671 | tok/s 2628902.01 | norm 0.3680 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16205 | train 2.901078 | val 3.323671 | tok/s 2629447.96 | norm 0.3728 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16206 | train 3.053607 | val 3.323671 | tok/s 2626667.84 | norm 0.3541 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16207 | train 2.911976 | val 3.323671 | tok/s 2631016.47 | norm 0.4007 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16208 | train 2.915660 | val 3.323671 | tok/s 2627063.25 | norm 0.3628 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16209 | train 2.871222 | val 3.323671 | tok/s 2628654.91 | norm 0.3393 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16210 | train 2.930133 | val 3.323671 | tok/s 2630560.06 | norm 0.3658 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16211 | train 2.957674 | val 3.323671 | tok/s 2621798.74 | norm 0.3707 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16212 | train 2.856332 | val 3.323671 | tok/s 2627586.78 | norm 0.3886 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16213 | train 2.937146 | val 3.323671 | tok/s 2631091.59 | norm 0.3836 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16214 | train 2.917063 | val 3.323671 | tok/s 2629717.69 | norm 0.3582 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16215 | train 2.903371 | val 3.323671 | tok/s 2630051.37 | norm 0.3843 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16216 | train 3.022309 | val 3.323671 | tok/s 2627029.15 | norm 0.3998 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16217 | train 2.961771 | val 3.323671 | tok/s 2630403.42 | norm 0.4104 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16218 | train 3.048804 | val 3.323671 | tok/s 2632155.51 | norm 0.3969 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16219 | train 2.928983 | val 3.323671 | tok/s 2632138.81 | norm 0.4065 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16220 | train 2.937659 | val 3.323671 | tok/s 2631101.43 | norm 0.4183 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16221 | train 3.030313 | val 3.323671 | tok/s 2623349.52 | norm 0.3917 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16222 | train 2.858564 | val 3.323671 | tok/s 2629534.97 | norm 0.3488 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16223 | train 2.959457 | val 3.323671 | tok/s 2629815.04 | norm 0.3865 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16224 | train 2.965254 | val 3.323671 | tok/s 2634003.34 | norm 0.3899 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16225 | train 2.935025 | val 3.323671 | tok/s 2630908.43 | norm 0.3553 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16226 | train 2.930585 | val 3.323671 | tok/s 2622508.19 | norm 0.4008 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16227 | train 2.883377 | val 3.323671 | tok/s 2631662.36 | norm 0.4033 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16228 | train 3.013651 | val 3.323671 | tok/s 2630913.09 | norm 0.3693 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16229 | train 3.026516 | val 3.323671 | tok/s 2629053.83 | norm 0.3821 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16230 | train 3.025934 | val 3.323671 | tok/s 2627234.73 | norm 0.3732 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16231 | train 3.105343 | val 3.323671 | tok/s 2622561.67 | norm 0.3830 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16232 | train 3.174256 | val 3.323671 | tok/s 2628898.71 | norm 0.3931 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16233 | train 2.985044 | val 3.323671 | tok/s 2629594.95 | norm 0.3477 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16234 | train 3.080811 | val 3.323671 | tok/s 2631091.79 | norm 0.3276 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16235 | train 2.981564 | val 3.323671 | tok/s 2628718.29 | norm 0.3509 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16236 | train 2.849812 | val 3.323671 | tok/s 2629613.23 | norm 0.3715 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16237 | train 2.914085 | val 3.323671 | tok/s 2631385.20 | norm 0.3781 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16238 | train 3.031581 | val 3.323671 | tok/s 2626200.78 | norm 0.3890 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16239 | train 2.966132 | val 3.323671 | tok/s 2630101.70 | norm 0.3728 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16240 | train 2.955241 | val 3.323671 | tok/s 2623364.38 | norm 0.3707 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16241 | train 3.155555 | val 3.323671 | tok/s 2597622.62 | norm 0.4124 | dt 201.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16242 | train 2.868170 | val 3.323671 | tok/s 2628229.79 | norm 0.3669 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16243 | train 3.005524 | val 3.323671 | tok/s 2625423.48 | norm 0.3519 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16244 | train 2.877589 | val 3.323671 | tok/s 2630842.04 | norm 0.3700 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16245 | train 2.978226 | val 3.323671 | tok/s 2539771.54 | norm 0.4223 | dt 206.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16246 | train 2.942883 | val 3.323671 | tok/s 2633194.87 | norm 0.4069 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16247 | train 2.887648 | val 3.323671 | tok/s 2631336.07 | norm 0.3866 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16248 | train 2.971699 | val 3.323671 | tok/s 2628108.05 | norm 0.3791 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16249 | train 2.929805 | val 3.323671 | tok/s 2628368.38 | norm 0.3749 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 16250 | val_loss 3.332702 step 16250 | train 2.957690 | val 3.332702 | tok/s 2615105.74 | norm 0.3786 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16251 | train 2.875141 | val 3.332702 | tok/s 2627145.27 | norm 0.3909 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16252 | train 2.924700 | val 3.332702 | tok/s 2571504.60 | norm 0.3649 | dt 203.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16253 | train 2.908974 | val 3.332702 | tok/s 2631960.62 | norm 0.3881 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16254 | train 2.899339 | val 3.332702 | tok/s 2626919.18 | norm 0.3828 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16255 | train 3.026763 | val 3.332702 | tok/s 2632612.07 | norm 0.3596 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16256 | train 2.935708 | val 3.332702 | tok/s 2626893.15 | norm 0.3539 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16257 | train 2.896930 | val 3.332702 | tok/s 2626944.20 | norm 0.3569 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16258 | train 2.987968 | val 3.332702 | tok/s 2633194.61 | norm 0.3738 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16259 | train 3.197257 | val 3.332702 | tok/s 2626876.33 | norm 0.3854 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16260 | train 3.031829 | val 3.332702 | tok/s 2628944.36 | norm 0.3852 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16261 | train 2.984472 | val 3.332702 | tok/s 2619703.52 | norm 0.3922 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16262 | train 3.086595 | val 3.332702 | tok/s 2628975.74 | norm 0.4194 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16263 | train 2.896003 | val 3.332702 | tok/s 2630054.82 | norm 0.4223 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16264 | train 3.003253 | val 3.332702 | tok/s 2624956.15 | norm 0.4026 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16265 | train 3.092580 | val 3.332702 | tok/s 2630110.75 | norm 0.4040 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16266 | train 2.965634 | val 3.332702 | tok/s 2621687.62 | norm 0.3726 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16267 | train 3.046662 | val 3.332702 | tok/s 2628353.85 | norm 0.3454 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16268 | train 2.817218 | val 3.332702 | tok/s 2631485.06 | norm 0.3447 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16269 | train 3.032429 | val 3.332702 | tok/s 2629145.86 | norm 0.3707 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16270 | train 2.941604 | val 3.332702 | tok/s 2631467.46 | norm 0.3741 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16271 | train 2.889349 | val 3.332702 | tok/s 2540776.19 | norm 0.3882 | dt 206.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16272 | train 2.929776 | val 3.332702 | tok/s 2595204.57 | norm 0.3888 | dt 202.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16273 | train 3.032553 | val 3.332702 | tok/s 2629055.01 | norm 0.4187 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16274 | train 2.928529 | val 3.332702 | tok/s 2631223.63 | norm 0.3777 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16275 | train 2.936873 | val 3.332702 | tok/s 2629844.32 | norm 0.3186 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16276 | train 2.992806 | val 3.332702 | tok/s 2624723.66 | norm 0.3685 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16277 | train 3.144567 | val 3.332702 | tok/s 2628476.91 | norm 0.4006 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16278 | train 2.925107 | val 3.332702 | tok/s 2625649.72 | norm 0.3752 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16279 | train 3.009368 | val 3.332702 | tok/s 2625911.44 | norm 0.3892 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16280 | train 2.984772 | val 3.332702 | tok/s 2619110.01 | norm 0.4649 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16281 | train 2.902526 | val 3.332702 | tok/s 2622479.68 | norm 0.4212 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16282 | train 2.952331 | val 3.332702 | tok/s 2619311.02 | norm 0.3869 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16283 | train 2.935790 | val 3.332702 | tok/s 2625804.49 | norm 0.3737 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16284 | train 3.001392 | val 3.332702 | tok/s 2624011.91 | norm 0.3333 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16285 | train 2.859223 | val 3.332702 | tok/s 2628540.23 | norm 0.3525 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16286 | train 2.908041 | val 3.332702 | tok/s 2627349.57 | norm 0.3503 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16287 | train 3.070086 | val 3.332702 | tok/s 2626617.65 | norm 0.3828 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16288 | train 2.968539 | val 3.332702 | tok/s 2627817.19 | norm 0.4207 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16289 | train 2.950283 | val 3.332702 | tok/s 2629171.49 | norm 0.3710 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16290 | train 2.973378 | val 3.332702 | tok/s 2608043.41 | norm 0.3721 | dt 201.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16291 | train 2.986620 | val 3.332702 | tok/s 2625551.57 | norm 0.3648 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16292 | train 2.916094 | val 3.332702 | tok/s 2629131.84 | norm 0.3652 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16293 | train 3.044385 | val 3.332702 | tok/s 2633435.19 | norm 0.3866 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16294 | train 2.994176 | val 3.332702 | tok/s 2616783.37 | norm 0.3964 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16295 | train 2.924292 | val 3.332702 | tok/s 2631125.12 | norm 0.3586 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16296 | train 2.950059 | val 3.332702 | tok/s 2628067.84 | norm 0.3576 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16297 | train 3.022015 | val 3.332702 | tok/s 2626931.76 | norm 0.3675 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16298 | train 2.881752 | val 3.332702 | tok/s 2630385.78 | norm 0.4082 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16299 | train 2.863823 | val 3.332702 | tok/s 2623295.86 | norm 0.3894 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16300 | train 3.107919 | val 3.332702 | tok/s 2628197.75 | norm 0.3469 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16301 | train 2.940256 | val 3.332702 | tok/s 2626980.28 | norm 0.3412 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16302 | train 2.928228 | val 3.332702 | tok/s 2628323.84 | norm 0.3403 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16303 | train 2.896566 | val 3.332702 | tok/s 2626276.00 | norm 0.3552 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16304 | train 2.946405 | val 3.332702 | tok/s 2628539.41 | norm 0.3397 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16305 | train 3.023403 | val 3.332702 | tok/s 2628659.85 | norm 0.3452 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16306 | train 2.960733 | val 3.332702 | tok/s 2628302.26 | norm 0.3412 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16307 | train 2.963376 | val 3.332702 | tok/s 2630540.91 | norm 0.4040 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16308 | train 3.007230 | val 3.332702 | tok/s 2626034.83 | norm 0.4989 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16309 | train 2.858328 | val 3.332702 | tok/s 2628832.88 | norm 0.3829 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16310 | train 2.906495 | val 3.332702 | tok/s 2632645.72 | norm 0.3632 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16311 | train 3.001317 | val 3.332702 | tok/s 2629359.14 | norm 0.3618 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16312 | train 3.036568 | val 3.332702 | tok/s 2628222.42 | norm 0.3451 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16313 | train 3.053910 | val 3.332702 | tok/s 2624479.04 | norm 0.3449 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16314 | train 2.936292 | val 3.332702 | tok/s 2627346.13 | norm 0.3787 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16315 | train 2.918098 | val 3.332702 | tok/s 2498605.42 | norm 0.4109 | dt 209.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16316 | train 2.997556 | val 3.332702 | tok/s 2479109.36 | norm 0.4062 | dt 211.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16317 | train 3.065304 | val 3.332702 | tok/s 2631143.20 | norm 0.3669 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16318 | train 2.911286 | val 3.332702 | tok/s 2627902.79 | norm 0.3828 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16319 | train 2.930047 | val 3.332702 | tok/s 2630205.67 | norm 0.3827 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16320 | train 2.867535 | val 3.332702 | tok/s 2634387.17 | norm 0.3914 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16321 | train 2.778636 | val 3.332702 | tok/s 2525039.97 | norm 0.4199 | dt 207.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16322 | train 2.978869 | val 3.332702 | tok/s 2631986.90 | norm 0.4030 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16323 | train 3.022873 | val 3.332702 | tok/s 2625491.21 | norm 0.3927 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16324 | train 2.943183 | val 3.332702 | tok/s 2627970.53 | norm 0.3591 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16325 | train 2.976862 | val 3.332702 | tok/s 2629320.76 | norm 0.3462 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16326 | train 3.001178 | val 3.332702 | tok/s 2627461.66 | norm 0.3750 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16327 | train 3.075976 | val 3.332702 | tok/s 2630787.09 | norm 0.3931 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16328 | train 2.916576 | val 3.332702 | tok/s 2626106.96 | norm 0.3677 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16329 | train 3.077141 | val 3.332702 | tok/s 2630723.32 | norm 0.3910 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16330 | train 3.023118 | val 3.332702 | tok/s 2625935.29 | norm 0.3974 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16331 | train 2.991012 | val 3.332702 | tok/s 2627549.04 | norm 0.3669 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16332 | train 2.917913 | val 3.332702 | tok/s 2628306.12 | norm 0.3263 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16333 | train 2.965193 | val 3.332702 | tok/s 2627059.49 | norm 0.3454 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16334 | train 2.903022 | val 3.332702 | tok/s 2628324.67 | norm 0.3740 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16335 | train 2.926700 | val 3.332702 | tok/s 2630938.32 | norm 0.3936 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16336 | train 3.021700 | val 3.332702 | tok/s 2628510.71 | norm 0.3776 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16337 | train 2.923236 | val 3.332702 | tok/s 2630589.73 | norm 0.3675 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16338 | train 2.970526 | val 3.332702 | tok/s 2623506.88 | norm 0.6675 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16339 | train 2.998269 | val 3.332702 | tok/s 2626586.15 | norm 0.3833 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16340 | train 2.958666 | val 3.332702 | tok/s 2631146.60 | norm 0.3736 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16341 | train 3.002928 | val 3.332702 | tok/s 2626776.55 | norm 0.3949 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16342 | train 2.963339 | val 3.332702 | tok/s 2631004.81 | norm 0.3570 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16343 | train 2.847982 | val 3.332702 | tok/s 2625469.78 | norm 0.3978 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16344 | train 3.015250 | val 3.332702 | tok/s 2628073.86 | norm 0.4418 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16345 | train 2.937000 | val 3.332702 | tok/s 2621398.39 | norm 0.3783 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16346 | train 2.986590 | val 3.332702 | tok/s 2627451.98 | norm 0.3781 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16347 | train 2.976205 | val 3.332702 | tok/s 2633899.13 | norm 0.4183 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16348 | train 2.980764 | val 3.332702 | tok/s 2631209.18 | norm 0.3968 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16349 | train 3.085498 | val 3.332702 | tok/s 2632658.59 | norm 0.3817 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16350 | train 3.009711 | val 3.332702 | tok/s 2601524.89 | norm 0.3935 | dt 201.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16351 | train 2.925840 | val 3.332702 | tok/s 2628563.71 | norm 0.3712 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16352 | train 2.931016 | val 3.332702 | tok/s 2633368.06 | norm 0.3873 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16353 | train 2.939820 | val 3.332702 | tok/s 2631472.57 | norm 0.4245 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16354 | train 2.891269 | val 3.332702 | tok/s 2634015.13 | norm 0.3923 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16355 | train 3.008122 | val 3.332702 | tok/s 2626442.59 | norm 0.3584 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16356 | train 3.043521 | val 3.332702 | tok/s 2530845.87 | norm 0.3642 | dt 207.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16357 | train 2.975839 | val 3.332702 | tok/s 2467326.83 | norm 0.3701 | dt 212.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16358 | train 2.983479 | val 3.332702 | tok/s 2630275.78 | norm 0.3767 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16359 | train 2.998648 | val 3.332702 | tok/s 2630944.97 | norm 0.3636 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16360 | train 3.190394 | val 3.332702 | tok/s 2631350.34 | norm 0.3894 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16361 | train 3.013143 | val 3.332702 | tok/s 2628398.22 | norm 0.3456 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16362 | train 2.951788 | val 3.332702 | tok/s 2630559.27 | norm 0.3714 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16363 | train 3.065583 | val 3.332702 | tok/s 2630154.66 | norm 0.4147 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16364 | train 2.941877 | val 3.332702 | tok/s 2580946.68 | norm 0.3821 | dt 203.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16365 | train 2.972663 | val 3.332702 | tok/s 2629959.23 | norm 0.3922 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16366 | train 2.962480 | val 3.332702 | tok/s 2608380.19 | norm 0.3906 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16367 | train 3.041408 | val 3.332702 | tok/s 2626514.37 | norm 0.3983 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16368 | train 2.937994 | val 3.332702 | tok/s 2626888.61 | norm 0.3944 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16369 | train 2.992370 | val 3.332702 | tok/s 2631537.93 | norm 0.4126 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16370 | train 2.889771 | val 3.332702 | tok/s 2628468.91 | norm 0.3641 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16371 | train 2.818915 | val 3.332702 | tok/s 2627979.11 | norm 0.3696 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16372 | train 3.045124 | val 3.332702 | tok/s 2629454.39 | norm 0.3751 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16373 | train 3.150007 | val 3.332702 | tok/s 2629580.99 | norm 0.4176 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16374 | train 2.959845 | val 3.332702 | tok/s 2631453.47 | norm 0.3766 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16375 | train 3.030056 | val 3.332702 | tok/s 2630658.49 | norm 0.3723 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16376 | train 2.798250 | val 3.332702 | tok/s 2633612.01 | norm 0.3380 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16377 | train 2.844104 | val 3.332702 | tok/s 2628168.57 | norm 0.3215 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16378 | train 2.943988 | val 3.332702 | tok/s 2630341.80 | norm 0.4153 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16379 | train 3.058424 | val 3.332702 | tok/s 2631409.16 | norm 0.3936 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16380 | train 2.859684 | val 3.332702 | tok/s 2632503.16 | norm 0.3170 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16381 | train 2.967839 | val 3.332702 | tok/s 2633606.26 | norm 0.3370 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16382 | train 2.989783 | val 3.332702 | tok/s 2635651.84 | norm 0.3726 | dt 198.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16383 | train 2.863239 | val 3.332702 | tok/s 2630211.29 | norm 0.3745 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16384 | train 2.919754 | val 3.332702 | tok/s 2632220.67 | norm 0.3579 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16385 | train 2.993057 | val 3.332702 | tok/s 2632193.29 | norm 0.3688 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16386 | train 2.987440 | val 3.332702 | tok/s 2631695.19 | norm 0.3884 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16387 | train 2.956976 | val 3.332702 | tok/s 2631409.34 | norm 0.3805 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16388 | train 2.907162 | val 3.332702 | tok/s 2629546.92 | norm 0.3719 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16389 | train 2.953167 | val 3.332702 | tok/s 2629130.07 | norm 0.3823 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16390 | train 3.011681 | val 3.332702 | tok/s 2629474.60 | norm 0.3746 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16391 | train 2.861093 | val 3.332702 | tok/s 2627735.24 | norm 0.3872 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16392 | train 2.916909 | val 3.332702 | tok/s 2630924.14 | norm 0.3657 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16393 | train 2.900250 | val 3.332702 | tok/s 2630354.18 | norm 0.3762 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16394 | train 2.993097 | val 3.332702 | tok/s 2625271.70 | norm 0.3810 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16395 | train 2.847027 | val 3.332702 | tok/s 2629131.56 | norm 0.3473 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16396 | train 3.015840 | val 3.332702 | tok/s 2631378.98 | norm 0.3589 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16397 | train 2.930312 | val 3.332702 | tok/s 2625633.14 | norm 0.4014 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16398 | train 2.887496 | val 3.332702 | tok/s 2628459.92 | norm 0.3992 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16399 | train 2.917870 | val 3.332702 | tok/s 2625674.52 | norm 0.4229 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16400 | train 2.831931 | val 3.332702 | tok/s 2629511.54 | norm 0.4222 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16401 | train 2.907687 | val 3.332702 | tok/s 2631204.57 | norm 0.4218 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16402 | train 2.951453 | val 3.332702 | tok/s 2630149.07 | norm 0.3946 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16403 | train 2.867232 | val 3.332702 | tok/s 2634115.95 | norm 0.3903 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16404 | train 2.989507 | val 3.332702 | tok/s 2633029.57 | norm 0.3725 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16405 | train 3.207915 | val 3.332702 | tok/s 2609815.24 | norm 0.3907 | dt 200.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16406 | train 2.899759 | val 3.332702 | tok/s 2573318.94 | norm 0.3942 | dt 203.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16407 | train 2.832713 | val 3.332702 | tok/s 2628566.53 | norm 0.3318 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16408 | train 3.157033 | val 3.332702 | tok/s 2629896.51 | norm 0.3982 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16409 | train 3.078116 | val 3.332702 | tok/s 2629891.18 | norm 0.4114 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16410 | train 2.932920 | val 3.332702 | tok/s 2625928.71 | norm 0.3326 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16411 | train 2.968308 | val 3.332702 | tok/s 2630846.12 | norm 0.3193 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16412 | train 3.143070 | val 3.332702 | tok/s 2632271.76 | norm 0.3728 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16413 | train 2.972451 | val 3.332702 | tok/s 2631077.37 | norm 0.3731 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16414 | train 2.880427 | val 3.332702 | tok/s 2631629.96 | norm 0.3908 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16415 | train 2.950701 | val 3.332702 | tok/s 2479939.90 | norm 0.3961 | dt 211.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16416 | train 2.941535 | val 3.332702 | tok/s 2632012.62 | norm 0.3944 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16417 | train 2.927945 | val 3.332702 | tok/s 2576657.92 | norm 0.3935 | dt 203.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16418 | train 2.847665 | val 3.332702 | tok/s 2629739.15 | norm 0.4232 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16419 | train 3.075070 | val 3.332702 | tok/s 2630190.70 | norm 0.4203 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16420 | train 3.023808 | val 3.332702 | tok/s 2629228.41 | norm 0.3826 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16421 | train 2.945716 | val 3.332702 | tok/s 2631179.40 | norm 0.3531 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16422 | train 3.044507 | val 3.332702 | tok/s 2632452.11 | norm 0.3849 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16423 | train 3.016097 | val 3.332702 | tok/s 2633439.32 | norm 0.3456 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16424 | train 2.982776 | val 3.332702 | tok/s 2629820.95 | norm 0.3481 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16425 | train 2.854018 | val 3.332702 | tok/s 2634141.38 | norm 0.3976 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16426 | train 2.976787 | val 3.332702 | tok/s 2631630.30 | norm 0.3574 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16427 | train 2.910728 | val 3.332702 | tok/s 2635432.21 | norm 0.3649 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16428 | train 2.990846 | val 3.332702 | tok/s 2634315.41 | norm 0.4099 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16429 | train 2.932226 | val 3.332702 | tok/s 2630708.35 | norm 0.3636 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16430 | train 2.940788 | val 3.332702 | tok/s 2630969.99 | norm 0.3242 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16431 | train 2.908902 | val 3.332702 | tok/s 2632191.62 | norm 0.3428 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16432 | train 2.946650 | val 3.332702 | tok/s 2632070.46 | norm 0.3602 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16433 | train 2.998382 | val 3.332702 | tok/s 2630277.96 | norm 0.3610 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16434 | train 3.175566 | val 3.332702 | tok/s 2632914.12 | norm 0.5196 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16435 | train 2.977314 | val 3.332702 | tok/s 2632603.70 | norm 0.3692 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16436 | train 3.015733 | val 3.332702 | tok/s 2632178.75 | norm 0.3465 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16437 | train 2.995379 | val 3.332702 | tok/s 2630873.85 | norm 0.3401 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16438 | train 2.939661 | val 3.332702 | tok/s 2629284.06 | norm 0.3713 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16439 | train 2.980247 | val 3.332702 | tok/s 2628897.89 | norm 0.3736 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16440 | train 2.896221 | val 3.332702 | tok/s 2630279.53 | norm 0.3615 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16441 | train 3.208950 | val 3.332702 | tok/s 2616088.31 | norm 0.4869 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16442 | train 2.902402 | val 3.332702 | tok/s 2632404.50 | norm 0.3487 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16443 | train 2.931823 | val 3.332702 | tok/s 2633234.96 | norm 0.3071 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16444 | train 2.920764 | val 3.332702 | tok/s 2628850.81 | norm 0.3285 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16445 | train 2.933465 | val 3.332702 | tok/s 2630900.34 | norm 0.3709 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16446 | train 2.981358 | val 3.332702 | tok/s 2632683.55 | norm 0.3833 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16447 | train 2.943597 | val 3.332702 | tok/s 2632906.40 | norm 0.3553 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16448 | train 2.920505 | val 3.332702 | tok/s 2629344.59 | norm 0.3832 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16449 | train 2.982576 | val 3.332702 | tok/s 2627340.18 | norm 0.3608 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16450 | train 2.908996 | val 3.332702 | tok/s 2632576.06 | norm 0.3556 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16451 | train 3.052891 | val 3.332702 | tok/s 2629938.69 | norm 0.3753 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16452 | train 2.908515 | val 3.332702 | tok/s 2631918.93 | norm 0.3807 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16453 | train 2.896249 | val 3.332702 | tok/s 2633137.32 | norm 0.3884 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16454 | train 2.938776 | val 3.332702 | tok/s 2629249.61 | norm 0.4253 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16455 | train 2.970582 | val 3.332702 | tok/s 2632742.90 | norm 0.4003 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16456 | train 2.913946 | val 3.332702 | tok/s 2623996.53 | norm 0.3696 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16457 | train 2.927612 | val 3.332702 | tok/s 2633851.03 | norm 0.4052 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16458 | train 2.999191 | val 3.332702 | tok/s 2633287.54 | norm 0.3749 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16459 | train 2.905918 | val 3.332702 | tok/s 2596346.94 | norm 0.3768 | dt 201.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16460 | train 3.042324 | val 3.332702 | tok/s 2629779.99 | norm 0.3760 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16461 | train 3.015959 | val 3.332702 | tok/s 2631206.00 | norm 0.3795 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16462 | train 2.934304 | val 3.332702 | tok/s 2634831.74 | norm 0.3865 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16463 | train 2.895342 | val 3.332702 | tok/s 2632765.87 | norm 0.4042 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16464 | train 3.049659 | val 3.332702 | tok/s 2634701.65 | norm 0.4450 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16465 | train 2.753824 | val 3.332702 | tok/s 2635785.18 | norm 0.4161 | dt 198.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16466 | train 2.970894 | val 3.332702 | tok/s 2614674.99 | norm 0.3623 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16467 | train 2.941083 | val 3.332702 | tok/s 2632761.92 | norm 0.3553 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16468 | train 2.890881 | val 3.332702 | tok/s 2630734.98 | norm 0.3839 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16469 | train 3.522302 | val 3.332702 | tok/s 2631570.23 | norm 0.8009 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16470 | train 3.117361 | val 3.332702 | tok/s 2628870.58 | norm 0.4664 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16471 | train 2.981743 | val 3.332702 | tok/s 2631689.51 | norm 0.3766 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16472 | train 2.886163 | val 3.332702 | tok/s 2628309.26 | norm 0.3277 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16473 | train 3.071522 | val 3.332702 | tok/s 2631654.81 | norm 0.3385 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16474 | train 2.851882 | val 3.332702 | tok/s 2634155.07 | norm 0.3848 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16475 | train 2.987083 | val 3.332702 | tok/s 2633047.14 | norm 0.4040 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16476 | train 2.943000 | val 3.332702 | tok/s 2495245.43 | norm 0.3722 | dt 210.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16477 | train 2.877444 | val 3.332702 | tok/s 2541999.15 | norm 0.3816 | dt 206.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16478 | train 3.013692 | val 3.332702 | tok/s 2630651.19 | norm 0.4022 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16479 | train 2.949813 | val 3.332702 | tok/s 2631755.38 | norm 0.3510 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16480 | train 2.917817 | val 3.332702 | tok/s 2628997.20 | norm 0.3762 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16481 | train 2.880326 | val 3.332702 | tok/s 2633080.65 | norm 0.3597 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16482 | train 2.882757 | val 3.332702 | tok/s 2629019.08 | norm 0.3758 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16483 | train 2.925686 | val 3.332702 | tok/s 2629876.75 | norm 0.3983 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16484 | train 2.971836 | val 3.332702 | tok/s 2631040.19 | norm 0.3918 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16485 | train 2.930826 | val 3.332702 | tok/s 2630535.78 | norm 0.3735 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16486 | train 2.952866 | val 3.332702 | tok/s 2628523.96 | norm 0.3586 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16487 | train 2.984321 | val 3.332702 | tok/s 2633915.55 | norm 0.3746 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16488 | train 2.970352 | val 3.332702 | tok/s 2628804.65 | norm 0.3716 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16489 | train 2.934988 | val 3.332702 | tok/s 2629314.78 | norm 0.3661 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16490 | train 3.190426 | val 3.332702 | tok/s 2627794.01 | norm 0.6130 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16491 | train 2.909994 | val 3.332702 | tok/s 2630468.86 | norm 0.3422 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16492 | train 2.888984 | val 3.332702 | tok/s 2631035.41 | norm 0.3366 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16493 | train 2.942686 | val 3.332702 | tok/s 2631134.55 | norm 0.4456 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16494 | train 2.917225 | val 3.332702 | tok/s 2627945.35 | norm 0.3651 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16495 | train 3.028128 | val 3.332702 | tok/s 2628568.45 | norm 0.3375 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16496 | train 2.990000 | val 3.332702 | tok/s 2630314.77 | norm 0.3576 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16497 | train 2.930989 | val 3.332702 | tok/s 2630100.26 | norm 0.3728 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16498 | train 2.827954 | val 3.332702 | tok/s 2632358.27 | norm 0.3677 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16499 | train 2.952988 | val 3.332702 | tok/s 2628891.78 | norm 0.3713 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 16500 | val_loss 3.329592 step 16500 | train 2.929598 | val 3.329592 | tok/s 2611904.16 | norm 0.3573 | dt 200.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16501 | train 2.943617 | val 3.329592 | tok/s 2628428.27 | norm 0.3499 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16502 | train 3.069329 | val 3.329592 | tok/s 2629115.04 | norm 0.3695 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16503 | train 3.086515 | val 3.329592 | tok/s 2633307.12 | norm 0.3602 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16504 | train 3.188398 | val 3.329592 | tok/s 2633713.66 | norm 0.3944 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16505 | train 2.897164 | val 3.329592 | tok/s 2628451.84 | norm 0.4019 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16506 | train 2.903976 | val 3.329592 | tok/s 2630000.80 | norm 0.3417 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16507 | train 2.874182 | val 3.329592 | tok/s 2628744.91 | norm 0.4028 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16508 | train 2.898184 | val 3.329592 | tok/s 2631383.80 | norm 0.3973 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16509 | train 2.963160 | val 3.329592 | tok/s 2631834.63 | norm 0.3520 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16510 | train 3.064275 | val 3.329592 | tok/s 2628989.77 | norm 0.3486 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16511 | train 3.054198 | val 3.329592 | tok/s 2630596.66 | norm 0.3628 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16512 | train 2.916672 | val 3.329592 | tok/s 2631797.45 | norm 0.3292 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16513 | train 3.027226 | val 3.329592 | tok/s 2631487.66 | norm 0.3305 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16514 | train 3.031419 | val 3.329592 | tok/s 2634489.84 | norm 0.3558 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16515 | train 3.051615 | val 3.329592 | tok/s 2634745.15 | norm 0.3851 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16516 | train 2.956716 | val 3.329592 | tok/s 2631094.43 | norm 0.3699 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16517 | train 2.944366 | val 3.329592 | tok/s 2631636.42 | norm 0.3874 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16518 | train 2.888712 | val 3.329592 | tok/s 2632932.80 | norm 0.3663 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16519 | train 2.889365 | val 3.329592 | tok/s 2632840.42 | norm 0.3834 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16520 | train 2.833162 | val 3.329592 | tok/s 2628165.44 | norm 0.4096 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16521 | train 2.916463 | val 3.329592 | tok/s 2632864.91 | norm 0.3999 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16522 | train 3.103443 | val 3.329592 | tok/s 2632763.74 | norm 0.3958 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16523 | train 3.059929 | val 3.329592 | tok/s 2633697.33 | norm 0.3835 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16524 | train 2.870730 | val 3.329592 | tok/s 2631842.87 | norm 0.3633 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16525 | train 2.985983 | val 3.329592 | tok/s 2630855.91 | norm 0.3649 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16526 | train 3.007097 | val 3.329592 | tok/s 2628851.38 | norm 0.3612 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16527 | train 3.081253 | val 3.329592 | tok/s 2627599.63 | norm 0.3875 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16528 | train 2.962111 | val 3.329592 | tok/s 2631706.01 | norm 0.3765 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16529 | train 2.883610 | val 3.329592 | tok/s 2628071.56 | norm 0.3319 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16530 | train 3.228946 | val 3.329592 | tok/s 2629341.50 | norm 0.4078 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16531 | train 2.866653 | val 3.329592 | tok/s 2630121.33 | norm 0.4227 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16532 | train 2.910249 | val 3.329592 | tok/s 2631899.87 | norm 0.3899 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16533 | train 2.880156 | val 3.329592 | tok/s 2629553.18 | norm 0.4169 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16534 | train 2.925878 | val 3.329592 | tok/s 2628732.00 | norm 0.4086 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16535 | train 2.875251 | val 3.329592 | tok/s 2635969.04 | norm 0.4318 | dt 198.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16536 | train 2.920555 | val 3.329592 | tok/s 2633210.25 | norm 0.4279 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16537 | train 3.046536 | val 3.329592 | tok/s 2631911.36 | norm 0.3967 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16538 | train 2.923110 | val 3.329592 | tok/s 2631375.24 | norm 0.3864 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16539 | train 2.955023 | val 3.329592 | tok/s 2634850.59 | norm 0.3677 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16540 | train 2.989912 | val 3.329592 | tok/s 2628433.55 | norm 0.3886 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16541 | train 2.903807 | val 3.329592 | tok/s 2623408.44 | norm 0.4069 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16542 | train 3.032562 | val 3.329592 | tok/s 2628064.22 | norm 0.4071 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16543 | train 2.995304 | val 3.329592 | tok/s 2625315.42 | norm 0.4276 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16544 | train 3.045511 | val 3.329592 | tok/s 2631626.33 | norm 0.4373 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16545 | train 2.899216 | val 3.329592 | tok/s 2629489.84 | norm 0.4009 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16546 | train 2.860018 | val 3.329592 | tok/s 2628696.69 | norm 0.3755 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16547 | train 2.996685 | val 3.329592 | tok/s 2634051.55 | norm 0.3929 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16548 | train 2.996361 | val 3.329592 | tok/s 2629923.67 | norm 0.3434 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16549 | train 3.012788 | val 3.329592 | tok/s 2626820.06 | norm 0.3673 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16550 | train 2.984809 | val 3.329592 | tok/s 2631028.51 | norm 0.4069 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16551 | train 3.044319 | val 3.329592 | tok/s 2631168.51 | norm 0.3674 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16552 | train 2.954381 | val 3.329592 | tok/s 2532936.64 | norm 0.3638 | dt 206.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16553 | train 2.946995 | val 3.329592 | tok/s 2613209.66 | norm 0.3616 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16554 | train 2.960286 | val 3.329592 | tok/s 2480614.51 | norm 0.3910 | dt 211.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16555 | train 2.944335 | val 3.329592 | tok/s 2630781.26 | norm 0.3939 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16556 | train 2.980156 | val 3.329592 | tok/s 2631742.40 | norm 0.3501 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16557 | train 2.961254 | val 3.329592 | tok/s 2630660.78 | norm 0.3510 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16558 | train 2.854939 | val 3.329592 | tok/s 2628761.55 | norm 0.3713 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16559 | train 2.858581 | val 3.329592 | tok/s 2633178.18 | norm 0.3771 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16560 | train 2.790957 | val 3.329592 | tok/s 2630623.55 | norm 0.3659 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16561 | train 2.769734 | val 3.329592 | tok/s 2563884.02 | norm 0.3794 | dt 204.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16562 | train 3.068805 | val 3.329592 | tok/s 2630891.66 | norm 0.3783 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16563 | train 2.917261 | val 3.329592 | tok/s 2634595.05 | norm 0.3597 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16564 | train 2.928566 | val 3.329592 | tok/s 2631911.65 | norm 0.3671 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16565 | train 2.979600 | val 3.329592 | tok/s 2634086.29 | norm 0.4130 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16566 | train 2.994583 | val 3.329592 | tok/s 2634587.42 | norm 0.4250 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16567 | train 2.948945 | val 3.329592 | tok/s 2631177.47 | norm 0.3935 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16568 | train 2.932124 | val 3.329592 | tok/s 2630372.01 | norm 0.3529 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16569 | train 2.960734 | val 3.329592 | tok/s 2614487.49 | norm 0.3577 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16570 | train 2.848920 | val 3.329592 | tok/s 2632458.87 | norm 0.3946 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16571 | train 2.882573 | val 3.329592 | tok/s 2628726.34 | norm 0.3942 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16572 | train 2.972351 | val 3.329592 | tok/s 2630354.49 | norm 0.3788 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16573 | train 3.041585 | val 3.329592 | tok/s 2631190.38 | norm 0.4231 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16574 | train 2.781330 | val 3.329592 | tok/s 2630565.91 | norm 0.4228 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16575 | train 2.983958 | val 3.329592 | tok/s 2629969.66 | norm 0.3887 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16576 | train 2.848946 | val 3.329592 | tok/s 2627636.49 | norm 0.3622 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16577 | train 2.894599 | val 3.329592 | tok/s 2630231.34 | norm 0.3592 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16578 | train 3.012915 | val 3.329592 | tok/s 2627197.28 | norm 0.3650 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16579 | train 2.914128 | val 3.329592 | tok/s 2629826.79 | norm 0.3481 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16580 | train 2.968530 | val 3.329592 | tok/s 2627005.90 | norm 0.3789 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16581 | train 2.876767 | val 3.329592 | tok/s 2624024.26 | norm 0.3693 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16582 | train 3.014963 | val 3.329592 | tok/s 2625092.91 | norm 0.3417 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16583 | train 2.993948 | val 3.329592 | tok/s 2626815.78 | norm 0.3677 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16584 | train 2.980530 | val 3.329592 | tok/s 2629387.97 | norm 0.3705 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16585 | train 3.085792 | val 3.329592 | tok/s 2628292.60 | norm 0.3682 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16586 | train 2.983528 | val 3.329592 | tok/s 2391804.40 | norm 0.3912 | dt 219.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16587 | train 2.842403 | val 3.329592 | tok/s 2344433.52 | norm 0.3455 | dt 223.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16588 | train 2.919540 | val 3.329592 | tok/s 2573835.75 | norm 0.3507 | dt 203.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16589 | train 2.966459 | val 3.329592 | tok/s 2622929.00 | norm 0.3853 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16590 | train 2.961941 | val 3.329592 | tok/s 2626866.24 | norm 0.3462 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16591 | train 2.980659 | val 3.329592 | tok/s 2628226.80 | norm 0.3640 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16592 | train 3.042044 | val 3.329592 | tok/s 2628636.70 | norm 0.3956 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16593 | train 2.895771 | val 3.329592 | tok/s 2629728.19 | norm 0.3686 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16594 | train 3.065687 | val 3.329592 | tok/s 2629634.86 | norm 0.3458 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16595 | train 2.987496 | val 3.329592 | tok/s 2621926.48 | norm 0.3844 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16596 | train 2.896758 | val 3.329592 | tok/s 2618546.77 | norm 0.3963 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16597 | train 2.920093 | val 3.329592 | tok/s 2628754.02 | norm 0.3600 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16598 | train 2.990879 | val 3.329592 | tok/s 2633029.27 | norm 0.3686 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16599 | train 2.873119 | val 3.329592 | tok/s 2624541.25 | norm 0.3990 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16600 | train 3.010071 | val 3.329592 | tok/s 2627260.12 | norm 0.3488 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16601 | train 2.973828 | val 3.329592 | tok/s 2629730.37 | norm 0.3508 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16602 | train 2.983310 | val 3.329592 | tok/s 2632135.10 | norm 0.3460 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16603 | train 3.022236 | val 3.329592 | tok/s 2634898.56 | norm 0.3462 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16604 | train 3.000227 | val 3.329592 | tok/s 2631681.32 | norm 0.3517 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16605 | train 2.940522 | val 3.329592 | tok/s 2630734.90 | norm 0.3598 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16606 | train 2.939958 | val 3.329592 | tok/s 2629525.40 | norm 0.3721 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16607 | train 3.027197 | val 3.329592 | tok/s 2629608.14 | norm 0.3732 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16608 | train 3.023171 | val 3.329592 | tok/s 2628962.17 | norm 0.3718 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16609 | train 3.136110 | val 3.329592 | tok/s 2628112.53 | norm 0.3720 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16610 | train 2.853611 | val 3.329592 | tok/s 2627373.49 | norm 0.4127 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16611 | train 3.008121 | val 3.329592 | tok/s 2633152.01 | norm 0.4086 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16612 | train 2.947819 | val 3.329592 | tok/s 2629699.15 | norm 0.4111 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16613 | train 3.070446 | val 3.329592 | tok/s 2628308.30 | norm 0.4187 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16614 | train 3.049746 | val 3.329592 | tok/s 2629439.44 | norm 0.3987 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16615 | train 2.896493 | val 3.329592 | tok/s 2632683.56 | norm 0.3965 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16616 | train 3.084589 | val 3.329592 | tok/s 2629320.52 | norm 0.3836 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16617 | train 2.917970 | val 3.329592 | tok/s 2626407.34 | norm 0.3578 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16618 | train 2.991084 | val 3.329592 | tok/s 2631228.45 | norm 0.3914 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16619 | train 2.999208 | val 3.329592 | tok/s 2627949.85 | norm 0.3485 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16620 | train 2.957832 | val 3.329592 | tok/s 2629685.51 | norm 0.3641 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16621 | train 2.955251 | val 3.329592 | tok/s 2628050.51 | norm 0.3901 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16622 | train 3.301557 | val 3.329592 | tok/s 2628790.80 | norm 0.4201 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16623 | train 2.942588 | val 3.329592 | tok/s 2631290.64 | norm 0.3870 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16624 | train 2.970418 | val 3.329592 | tok/s 2631206.27 | norm 0.3891 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16625 | train 2.893228 | val 3.329592 | tok/s 2629412.06 | norm 0.4249 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16626 | train 2.941410 | val 3.329592 | tok/s 2629616.94 | norm 0.3965 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16627 | train 2.889028 | val 3.329592 | tok/s 2628563.68 | norm 0.3482 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16628 | train 2.837582 | val 3.329592 | tok/s 2629155.60 | norm 0.3502 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16629 | train 2.886804 | val 3.329592 | tok/s 2629605.38 | norm 0.3688 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16630 | train 2.990342 | val 3.329592 | tok/s 2627430.03 | norm 0.3636 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16631 | train 2.911886 | val 3.329592 | tok/s 2630651.28 | norm 0.3492 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16632 | train 2.910615 | val 3.329592 | tok/s 2628932.85 | norm 0.3658 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16633 | train 3.147935 | val 3.329592 | tok/s 2631824.35 | norm 0.4060 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16634 | train 2.973779 | val 3.329592 | tok/s 2631509.56 | norm 0.4074 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16635 | train 2.917756 | val 3.329592 | tok/s 2631418.85 | norm 0.3555 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16636 | train 2.982098 | val 3.329592 | tok/s 2633926.68 | norm 0.3949 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16637 | train 2.894931 | val 3.329592 | tok/s 2628438.73 | norm 0.3910 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16638 | train 2.867909 | val 3.329592 | tok/s 2629962.25 | norm 0.3711 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16639 | train 2.844464 | val 3.329592 | tok/s 2628351.90 | norm 0.4149 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16640 | train 2.993616 | val 3.329592 | tok/s 2633350.47 | norm 0.3721 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16641 | train 2.948968 | val 3.329592 | tok/s 2631186.78 | norm 0.3561 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16642 | train 3.145914 | val 3.329592 | tok/s 2630505.66 | norm 0.4303 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16643 | train 2.964103 | val 3.329592 | tok/s 2629055.31 | norm 0.4132 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16644 | train 3.109062 | val 3.329592 | tok/s 2624964.72 | norm 0.3786 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16645 | train 2.963068 | val 3.329592 | tok/s 2631222.17 | norm 0.3666 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16646 | train 2.908094 | val 3.329592 | tok/s 2630116.99 | norm 0.3694 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16647 | train 3.117543 | val 3.329592 | tok/s 2626580.97 | norm 0.4983 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16648 | train 2.973742 | val 3.329592 | tok/s 2629391.61 | norm 0.4103 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16649 | train 2.870696 | val 3.329592 | tok/s 2627965.11 | norm 0.4206 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16650 | train 3.187442 | val 3.329592 | tok/s 2629836.17 | norm 0.4226 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16651 | train 2.951991 | val 3.329592 | tok/s 2630372.05 | norm 0.4145 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16652 | train 2.911240 | val 3.329592 | tok/s 2630453.47 | norm 0.3980 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16653 | train 2.978934 | val 3.329592 | tok/s 2632747.70 | norm 0.4002 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16654 | train 3.053170 | val 3.329592 | tok/s 2632208.88 | norm 0.4143 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16655 | train 2.960042 | val 3.329592 | tok/s 2627826.95 | norm 0.3562 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16656 | train 3.006490 | val 3.329592 | tok/s 2632346.31 | norm 0.3598 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16657 | train 2.951345 | val 3.329592 | tok/s 2631164.01 | norm 0.4184 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16658 | train 3.030498 | val 3.329592 | tok/s 2631891.64 | norm 0.4370 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16659 | train 2.975140 | val 3.329592 | tok/s 2629522.23 | norm 0.3572 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16660 | train 2.955983 | val 3.329592 | tok/s 2629726.87 | norm 0.3762 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16661 | train 2.958495 | val 3.329592 | tok/s 2631207.89 | norm 0.4005 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16662 | train 3.037580 | val 3.329592 | tok/s 2631108.73 | norm 0.3703 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16663 | train 2.969589 | val 3.329592 | tok/s 2630490.84 | norm 0.3693 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16664 | train 2.931848 | val 3.329592 | tok/s 2625891.91 | norm 0.3744 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16665 | train 3.010837 | val 3.329592 | tok/s 2627995.15 | norm 0.3692 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16666 | train 3.143682 | val 3.329592 | tok/s 2631267.05 | norm 0.3786 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16667 | train 3.044814 | val 3.329592 | tok/s 2633146.72 | norm 0.3870 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16668 | train 2.909286 | val 3.329592 | tok/s 2626824.69 | norm 0.4258 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16669 | train 2.962097 | val 3.329592 | tok/s 2624572.09 | norm 0.3759 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16670 | train 2.980266 | val 3.329592 | tok/s 2633193.93 | norm 0.3644 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16671 | train 2.900010 | val 3.329592 | tok/s 2547947.65 | norm 0.3501 | dt 205.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16672 | train 3.080856 | val 3.329592 | tok/s 2631260.58 | norm 0.3780 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16673 | train 2.963522 | val 3.329592 | tok/s 2244341.13 | norm 0.3723 | dt 233.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16674 | train 3.007449 | val 3.329592 | tok/s 2628089.28 | norm 0.3731 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16675 | train 2.943048 | val 3.329592 | tok/s 2627609.55 | norm 0.3860 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16676 | train 2.896890 | val 3.329592 | tok/s 2628602.84 | norm 0.3648 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16677 | train 3.039429 | val 3.329592 | tok/s 2622596.75 | norm 0.4535 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16678 | train 3.068183 | val 3.329592 | tok/s 2629312.50 | norm 0.4454 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16679 | train 2.917051 | val 3.329592 | tok/s 2626543.06 | norm 0.3732 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16680 | train 2.879776 | val 3.329592 | tok/s 2626431.22 | norm 0.3352 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16681 | train 2.907839 | val 3.329592 | tok/s 2373177.84 | norm 0.3642 | dt 220.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16682 | train 2.971375 | val 3.329592 | tok/s 2628399.40 | norm 0.3714 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16683 | train 2.925133 | val 3.329592 | tok/s 2629038.41 | norm 0.3617 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16684 | train 3.062025 | val 3.329592 | tok/s 2630305.15 | norm 0.3584 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16685 | train 2.968720 | val 3.329592 | tok/s 2503341.18 | norm 0.3616 | dt 209.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16686 | train 2.906444 | val 3.329592 | tok/s 2624754.97 | norm 0.3992 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16687 | train 2.843055 | val 3.329592 | tok/s 2630260.05 | norm 0.3870 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16688 | train 3.084729 | val 3.329592 | tok/s 2629163.06 | norm 0.3522 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16689 | train 3.096741 | val 3.329592 | tok/s 2625900.89 | norm 0.7194 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16690 | train 2.870008 | val 3.329592 | tok/s 2628182.10 | norm 0.3322 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16691 | train 3.005389 | val 3.329592 | tok/s 2630124.88 | norm 0.3200 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16692 | train 2.913015 | val 3.329592 | tok/s 2632324.79 | norm 0.3516 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16693 | train 2.962507 | val 3.329592 | tok/s 2633572.22 | norm 0.3847 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16694 | train 2.956771 | val 3.329592 | tok/s 2630415.69 | norm 0.3471 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16695 | train 2.890482 | val 3.329592 | tok/s 2627471.08 | norm 0.3612 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16696 | train 3.129066 | val 3.329592 | tok/s 2631570.35 | norm 0.4023 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16697 | train 2.884951 | val 3.329592 | tok/s 2633559.35 | norm 0.3730 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16698 | train 2.941521 | val 3.329592 | tok/s 2628537.74 | norm 0.3455 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16699 | train 2.909312 | val 3.329592 | tok/s 2629287.79 | norm 0.3362 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16700 | train 3.049868 | val 3.329592 | tok/s 2627109.28 | norm 0.3654 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16701 | train 3.050744 | val 3.329592 | tok/s 2625783.86 | norm 0.3806 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16702 | train 3.049726 | val 3.329592 | tok/s 2628843.90 | norm 0.3893 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16703 | train 2.878180 | val 3.329592 | tok/s 2626114.80 | norm 0.3969 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16704 | train 2.927050 | val 3.329592 | tok/s 2629334.85 | norm 0.3827 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16705 | train 2.981010 | val 3.329592 | tok/s 2630445.25 | norm 0.3930 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16706 | train 3.066862 | val 3.329592 | tok/s 2625578.49 | norm 0.4030 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16707 | train 2.951830 | val 3.329592 | tok/s 2629658.04 | norm 0.3757 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16708 | train 2.908369 | val 3.329592 | tok/s 2628986.30 | norm 0.4056 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16709 | train 2.943802 | val 3.329592 | tok/s 2630375.21 | norm 0.4064 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16710 | train 2.942114 | val 3.329592 | tok/s 2628266.73 | norm 0.3752 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16711 | train 2.938130 | val 3.329592 | tok/s 2628036.96 | norm 0.3669 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16712 | train 2.882933 | val 3.329592 | tok/s 2630683.17 | norm 0.3494 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16713 | train 3.168835 | val 3.329592 | tok/s 2629060.63 | norm 0.4044 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16714 | train 2.949477 | val 3.329592 | tok/s 2627764.66 | norm 0.3709 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16715 | train 2.884011 | val 3.329592 | tok/s 2631709.84 | norm 0.3749 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16716 | train 2.845532 | val 3.329592 | tok/s 2631749.75 | norm 0.3804 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16717 | train 2.948956 | val 3.329592 | tok/s 2631622.97 | norm 0.3828 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16718 | train 2.911294 | val 3.329592 | tok/s 2628470.24 | norm 0.4415 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16719 | train 3.036387 | val 3.329592 | tok/s 2631907.49 | norm 0.4073 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16720 | train 2.945513 | val 3.329592 | tok/s 2630625.17 | norm 0.3548 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16721 | train 2.939273 | val 3.329592 | tok/s 2628649.48 | norm 0.3501 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16722 | train 2.931163 | val 3.329592 | tok/s 2631200.27 | norm 0.3691 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16723 | train 2.948293 | val 3.329592 | tok/s 2629480.78 | norm 0.3721 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16724 | train 3.159268 | val 3.329592 | tok/s 2629433.74 | norm 0.4257 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16725 | train 2.934171 | val 3.329592 | tok/s 2631948.03 | norm 0.3903 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16726 | train 3.063105 | val 3.329592 | tok/s 2629581.25 | norm 0.3726 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16727 | train 2.984594 | val 3.329592 | tok/s 2630511.62 | norm 0.3832 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16728 | train 3.002160 | val 3.329592 | tok/s 2628651.07 | norm 0.3608 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16729 | train 2.909666 | val 3.329592 | tok/s 2629332.52 | norm 0.3877 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16730 | train 2.923123 | val 3.329592 | tok/s 2630837.26 | norm 0.3636 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16731 | train 2.955617 | val 3.329592 | tok/s 2630446.49 | norm 0.3481 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16732 | train 2.920553 | val 3.329592 | tok/s 2629611.85 | norm 0.3758 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16733 | train 3.075345 | val 3.329592 | tok/s 2632476.05 | norm 0.3694 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16734 | train 2.959701 | val 3.329592 | tok/s 2630328.70 | norm 0.3643 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16735 | train 3.033640 | val 3.329592 | tok/s 2627816.36 | norm 0.3934 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16736 | train 2.917441 | val 3.329592 | tok/s 2633560.24 | norm 0.4237 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16737 | train 2.997811 | val 3.329592 | tok/s 2630099.58 | norm 0.3859 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16738 | train 2.990616 | val 3.329592 | tok/s 2633150.56 | norm 0.3581 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16739 | train 2.963226 | val 3.329592 | tok/s 2631156.97 | norm 0.3455 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16740 | train 2.994232 | val 3.329592 | tok/s 2630473.34 | norm 0.3789 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16741 | train 2.892745 | val 3.329592 | tok/s 2627987.89 | norm 0.3752 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16742 | train 3.013996 | val 3.329592 | tok/s 2627221.16 | norm 0.3566 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16743 | train 2.912695 | val 3.329592 | tok/s 2631235.55 | norm 0.3529 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16744 | train 3.000646 | val 3.329592 | tok/s 2631438.73 | norm 0.3603 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16745 | train 2.935330 | val 3.329592 | tok/s 2628835.47 | norm 0.3476 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16746 | train 3.058056 | val 3.329592 | tok/s 2630585.86 | norm 0.3622 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16747 | train 2.946620 | val 3.329592 | tok/s 2628948.62 | norm 0.3641 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16748 | train 3.065200 | val 3.329592 | tok/s 2631930.47 | norm 0.3941 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16749 | train 2.866822 | val 3.329592 | tok/s 2632643.64 | norm 0.3532 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 16750 | val_loss 3.327171 step 16750 | train 2.890616 | val 3.327171 | tok/s 2608531.58 | norm 0.3703 | dt 200.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16751 | train 2.862179 | val 3.327171 | tok/s 2627761.94 | norm 0.3931 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16752 | train 2.897200 | val 3.327171 | tok/s 2628721.12 | norm 0.3678 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16753 | train 2.949580 | val 3.327171 | tok/s 2628354.04 | norm 0.3883 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16754 | train 3.104130 | val 3.327171 | tok/s 2629062.40 | norm 0.7056 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16755 | train 3.060508 | val 3.327171 | tok/s 2627879.97 | norm 0.4424 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16756 | train 2.933547 | val 3.327171 | tok/s 2627462.38 | norm 0.3694 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16757 | train 2.973746 | val 3.327171 | tok/s 2630923.73 | norm 0.4468 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16758 | train 2.924274 | val 3.327171 | tok/s 2628015.72 | norm 0.3590 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16759 | train 2.919013 | val 3.327171 | tok/s 2628310.09 | norm 0.3558 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16760 | train 2.858535 | val 3.327171 | tok/s 2628127.55 | norm 0.3664 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16761 | train 2.895845 | val 3.327171 | tok/s 2632222.25 | norm 0.3606 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16762 | train 3.048162 | val 3.327171 | tok/s 2629194.72 | norm 0.3766 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16763 | train 2.900561 | val 3.327171 | tok/s 2630938.82 | norm 0.3685 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16764 | train 3.039323 | val 3.327171 | tok/s 2630541.44 | norm 0.3748 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16765 | train 2.919265 | val 3.327171 | tok/s 2631450.87 | norm 0.3723 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16766 | train 2.934793 | val 3.327171 | tok/s 2627964.13 | norm 0.3602 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16767 | train 3.013440 | val 3.327171 | tok/s 2629291.29 | norm 0.3627 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16768 | train 3.039184 | val 3.327171 | tok/s 2628916.51 | norm 0.3894 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16769 | train 3.083839 | val 3.327171 | tok/s 2629310.93 | norm 0.4084 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16770 | train 2.943883 | val 3.327171 | tok/s 2629422.40 | norm 0.3707 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16771 | train 2.846781 | val 3.327171 | tok/s 2631471.78 | norm 0.3407 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16772 | train 3.051466 | val 3.327171 | tok/s 2630074.44 | norm 0.3222 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16773 | train 3.040030 | val 3.327171 | tok/s 2629212.73 | norm 0.3465 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16774 | train 2.922307 | val 3.327171 | tok/s 2629736.12 | norm 0.3765 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16775 | train 2.889187 | val 3.327171 | tok/s 2630804.85 | norm 0.4047 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16776 | train 2.947264 | val 3.327171 | tok/s 2631755.71 | norm 0.3882 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16777 | train 2.927793 | val 3.327171 | tok/s 2630827.19 | norm 0.3565 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16778 | train 3.083421 | val 3.327171 | tok/s 2626715.28 | norm 0.3719 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16779 | train 3.012176 | val 3.327171 | tok/s 2632636.88 | norm 0.3839 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16780 | train 2.963928 | val 3.327171 | tok/s 2631426.00 | norm 0.3632 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16781 | train 2.971074 | val 3.327171 | tok/s 2632353.43 | norm 0.3850 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16782 | train 2.948356 | val 3.327171 | tok/s 2628429.62 | norm 0.3963 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16783 | train 2.918390 | val 3.327171 | tok/s 2626848.44 | norm 0.4134 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16784 | train 3.102609 | val 3.327171 | tok/s 2630015.24 | norm 0.4001 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16785 | train 2.848818 | val 3.327171 | tok/s 2632205.42 | norm 0.3683 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16786 | train 2.839049 | val 3.327171 | tok/s 2632227.42 | norm 0.3839 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16787 | train 2.967250 | val 3.327171 | tok/s 2624037.21 | norm 0.3410 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16788 | train 2.918802 | val 3.327171 | tok/s 2627085.92 | norm 0.3533 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16789 | train 2.986622 | val 3.327171 | tok/s 2628507.45 | norm 0.3664 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16790 | train 3.111157 | val 3.327171 | tok/s 2625873.49 | norm 0.3800 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16791 | train 2.874554 | val 3.327171 | tok/s 2627311.94 | norm 0.4100 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16792 | train 2.948357 | val 3.327171 | tok/s 2629943.44 | norm 0.4137 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16793 | train 2.972464 | val 3.327171 | tok/s 2632426.50 | norm 0.3740 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16794 | train 3.082430 | val 3.327171 | tok/s 2630292.01 | norm 0.4960 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16795 | train 2.949846 | val 3.327171 | tok/s 2629693.73 | norm 0.4132 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16796 | train 2.858375 | val 3.327171 | tok/s 2629951.21 | norm 0.3615 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16797 | train 2.861338 | val 3.327171 | tok/s 2625505.74 | norm 0.3683 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16798 | train 2.961822 | val 3.327171 | tok/s 2626700.05 | norm 0.3751 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16799 | train 3.065846 | val 3.327171 | tok/s 2628929.78 | norm 0.4013 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16800 | train 2.951458 | val 3.327171 | tok/s 2627840.78 | norm 0.3828 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16801 | train 2.903851 | val 3.327171 | tok/s 2630024.57 | norm 0.3955 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16802 | train 3.024632 | val 3.327171 | tok/s 2630999.42 | norm 0.4204 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16803 | train 2.927527 | val 3.327171 | tok/s 2631065.08 | norm 0.3810 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16804 | train 2.949566 | val 3.327171 | tok/s 2631928.38 | norm 0.3938 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16805 | train 2.936110 | val 3.327171 | tok/s 2631597.27 | norm 0.4220 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16806 | train 3.075166 | val 3.327171 | tok/s 2633292.08 | norm 0.5411 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16807 | train 2.962173 | val 3.327171 | tok/s 2627348.30 | norm 0.3771 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16808 | train 2.931057 | val 3.327171 | tok/s 2629601.92 | norm 0.3398 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16809 | train 2.996111 | val 3.327171 | tok/s 2627415.21 | norm 0.3613 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16810 | train 2.834659 | val 3.327171 | tok/s 2628759.99 | norm 0.3483 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16811 | train 2.865431 | val 3.327171 | tok/s 2631162.80 | norm 0.3569 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16812 | train 2.993176 | val 3.327171 | tok/s 2629633.62 | norm 0.3893 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16813 | train 3.007312 | val 3.327171 | tok/s 2627484.87 | norm 0.3591 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16814 | train 2.942694 | val 3.327171 | tok/s 2631584.53 | norm 0.3807 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16815 | train 3.008061 | val 3.327171 | tok/s 2629509.03 | norm 0.4206 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16816 | train 3.040226 | val 3.327171 | tok/s 2629617.72 | norm 0.3894 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16817 | train 2.887691 | val 3.327171 | tok/s 2627282.09 | norm 0.4037 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16818 | train 2.945827 | val 3.327171 | tok/s 2630315.55 | norm 0.4139 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16819 | train 2.874001 | val 3.327171 | tok/s 2628304.54 | norm 0.4498 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16820 | train 2.944889 | val 3.327171 | tok/s 2629221.43 | norm 0.4251 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16821 | train 2.918520 | val 3.327171 | tok/s 2628634.48 | norm 0.3939 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16822 | train 3.024872 | val 3.327171 | tok/s 2625541.45 | norm 0.3864 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16823 | train 2.950869 | val 3.327171 | tok/s 2629760.77 | norm 0.3794 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16824 | train 2.880044 | val 3.327171 | tok/s 2631462.50 | norm 0.3491 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16825 | train 2.954588 | val 3.327171 | tok/s 2627780.91 | norm 0.3823 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16826 | train 2.916511 | val 3.327171 | tok/s 2628803.25 | norm 0.3467 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16827 | train 2.962415 | val 3.327171 | tok/s 2628751.82 | norm 0.3646 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16828 | train 2.894587 | val 3.327171 | tok/s 2627969.76 | norm 0.3720 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16829 | train 3.017219 | val 3.327171 | tok/s 2628290.94 | norm 0.3861 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16830 | train 2.886865 | val 3.327171 | tok/s 2630443.96 | norm 0.3700 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16831 | train 2.974598 | val 3.327171 | tok/s 2632157.13 | norm 0.3773 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16832 | train 2.913091 | val 3.327171 | tok/s 2628790.61 | norm 0.4320 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16833 | train 3.024748 | val 3.327171 | tok/s 2630509.01 | norm 0.4022 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16834 | train 2.932854 | val 3.327171 | tok/s 2631508.98 | norm 0.3744 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16835 | train 2.870324 | val 3.327171 | tok/s 2632409.22 | norm 0.3669 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16836 | train 2.911123 | val 3.327171 | tok/s 2632400.84 | norm 0.3857 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16837 | train 2.889914 | val 3.327171 | tok/s 2626062.36 | norm 0.3930 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16838 | train 2.935707 | val 3.327171 | tok/s 2628145.79 | norm 0.3819 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16839 | train 2.986074 | val 3.327171 | tok/s 2630856.23 | norm 0.3834 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16840 | train 2.829070 | val 3.327171 | tok/s 2630915.98 | norm 0.3781 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16841 | train 2.978240 | val 3.327171 | tok/s 2631623.33 | norm 0.3889 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16842 | train 2.953657 | val 3.327171 | tok/s 2630334.12 | norm 0.3853 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16843 | train 3.032626 | val 3.327171 | tok/s 2634706.86 | norm 0.3638 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16844 | train 2.912760 | val 3.327171 | tok/s 2604931.02 | norm 0.3492 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16845 | train 2.923326 | val 3.327171 | tok/s 2628425.80 | norm 0.3303 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16846 | train 3.035801 | val 3.327171 | tok/s 2629138.63 | norm 0.3415 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16847 | train 3.043417 | val 3.327171 | tok/s 2631040.09 | norm 0.3410 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16848 | train 2.897897 | val 3.327171 | tok/s 2621202.83 | norm 0.3422 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16849 | train 2.790702 | val 3.327171 | tok/s 2631780.49 | norm 0.3720 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16850 | train 3.016051 | val 3.327171 | tok/s 2629607.51 | norm 0.3758 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16851 | train 2.990895 | val 3.327171 | tok/s 2631815.06 | norm 0.3913 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16852 | train 2.935758 | val 3.327171 | tok/s 2629259.68 | norm 0.3706 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16853 | train 2.873832 | val 3.327171 | tok/s 2628205.04 | norm 0.3664 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16854 | train 3.066963 | val 3.327171 | tok/s 2627036.09 | norm 0.4368 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16855 | train 3.119392 | val 3.327171 | tok/s 2631362.90 | norm 0.3911 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16856 | train 2.945190 | val 3.327171 | tok/s 2628602.18 | norm 0.3441 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16857 | train 3.035418 | val 3.327171 | tok/s 2630819.56 | norm 0.3601 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16858 | train 3.001078 | val 3.327171 | tok/s 2628528.39 | norm 0.3589 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16859 | train 2.863970 | val 3.327171 | tok/s 2630674.55 | norm 0.3526 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16860 | train 2.925533 | val 3.327171 | tok/s 2624311.08 | norm 0.3756 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16861 | train 2.779673 | val 3.327171 | tok/s 2631752.42 | norm 0.3661 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16862 | train 2.870124 | val 3.327171 | tok/s 2630639.49 | norm 0.3665 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16863 | train 3.034072 | val 3.327171 | tok/s 2624144.69 | norm 0.4464 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16864 | train 2.928672 | val 3.327171 | tok/s 2626930.07 | norm 0.4400 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16865 | train 3.084307 | val 3.327171 | tok/s 2633743.73 | norm 0.5709 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16866 | train 2.978812 | val 3.327171 | tok/s 2627707.95 | norm 0.3455 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16867 | train 2.942932 | val 3.327171 | tok/s 2630268.39 | norm 0.3551 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16868 | train 2.908811 | val 3.327171 | tok/s 2627507.51 | norm 0.3798 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16869 | train 2.935679 | val 3.327171 | tok/s 2633419.98 | norm 0.4054 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16870 | train 2.976340 | val 3.327171 | tok/s 2628092.58 | norm 0.4016 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16871 | train 2.977695 | val 3.327171 | tok/s 2633338.45 | norm 0.3946 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16872 | train 2.911651 | val 3.327171 | tok/s 2628461.61 | norm 0.3698 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16873 | train 2.988331 | val 3.327171 | tok/s 2630358.47 | norm 0.3819 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16874 | train 2.881111 | val 3.327171 | tok/s 2629481.22 | norm 0.3850 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16875 | train 2.989382 | val 3.327171 | tok/s 2630393.82 | norm 0.3928 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16876 | train 2.937328 | val 3.327171 | tok/s 2628727.44 | norm 0.3835 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16877 | train 3.005121 | val 3.327171 | tok/s 2628794.83 | norm 0.3786 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16878 | train 2.966022 | val 3.327171 | tok/s 2629712.21 | norm 0.3561 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16879 | train 2.969632 | val 3.327171 | tok/s 2629126.63 | norm 0.3591 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16880 | train 2.982210 | val 3.327171 | tok/s 2625695.43 | norm 0.3885 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16881 | train 2.944415 | val 3.327171 | tok/s 2630604.21 | norm 0.3652 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16882 | train 2.946107 | val 3.327171 | tok/s 2624698.48 | norm 0.3363 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16883 | train 2.906319 | val 3.327171 | tok/s 2629438.16 | norm 0.3673 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16884 | train 2.978296 | val 3.327171 | tok/s 2630400.60 | norm 0.4040 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16885 | train 3.004955 | val 3.327171 | tok/s 2630414.04 | norm 0.3994 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16886 | train 2.843548 | val 3.327171 | tok/s 2631558.61 | norm 0.3812 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16887 | train 3.033449 | val 3.327171 | tok/s 2630638.20 | norm 0.4063 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16888 | train 2.829592 | val 3.327171 | tok/s 2628521.87 | norm 0.3595 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16889 | train 3.002900 | val 3.327171 | tok/s 2627869.61 | norm 0.3682 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16890 | train 2.894076 | val 3.327171 | tok/s 2624825.17 | norm 0.3956 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16891 | train 2.933084 | val 3.327171 | tok/s 2626379.96 | norm 0.3823 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16892 | train 2.814506 | val 3.327171 | tok/s 2625389.78 | norm 0.4287 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16893 | train 2.948714 | val 3.327171 | tok/s 2628662.83 | norm 0.4304 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16894 | train 2.932089 | val 3.327171 | tok/s 2630171.89 | norm 0.3652 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16895 | train 2.971047 | val 3.327171 | tok/s 2630764.82 | norm 0.3901 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16896 | train 2.954596 | val 3.327171 | tok/s 2629948.79 | norm 0.3775 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16897 | train 3.063866 | val 3.327171 | tok/s 2628813.31 | norm 0.3834 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16898 | train 2.931275 | val 3.327171 | tok/s 2625579.72 | norm 0.4242 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16899 | train 2.942908 | val 3.327171 | tok/s 2625193.80 | norm 0.3808 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16900 | train 3.011052 | val 3.327171 | tok/s 2632409.32 | norm 0.3583 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16901 | train 3.081479 | val 3.327171 | tok/s 2627957.52 | norm 0.3691 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16902 | train 2.924816 | val 3.327171 | tok/s 2625002.34 | norm 0.3828 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16903 | train 2.966982 | val 3.327171 | tok/s 2625983.10 | norm 0.3595 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16904 | train 2.896596 | val 3.327171 | tok/s 2626233.82 | norm 0.3689 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16905 | train 2.970983 | val 3.327171 | tok/s 2629161.95 | norm 0.3658 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16906 | train 2.791336 | val 3.327171 | tok/s 2625806.52 | norm 0.3701 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16907 | train 2.844357 | val 3.327171 | tok/s 2630321.57 | norm 0.3850 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16908 | train 2.960846 | val 3.327171 | tok/s 2627749.69 | norm 0.3602 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16909 | train 3.011359 | val 3.327171 | tok/s 2626252.16 | norm 0.3546 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16910 | train 2.898346 | val 3.327171 | tok/s 2628952.09 | norm 0.3807 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16911 | train 2.910381 | val 3.327171 | tok/s 2627066.22 | norm 0.3020 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16912 | train 2.878760 | val 3.327171 | tok/s 2631004.07 | norm 0.3605 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16913 | train 2.966666 | val 3.327171 | tok/s 2625814.15 | norm 0.3704 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16914 | train 3.086539 | val 3.327171 | tok/s 2445482.75 | norm 0.3818 | dt 214.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16915 | train 2.966016 | val 3.327171 | tok/s 2620891.60 | norm 0.3729 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16916 | train 3.199462 | val 3.327171 | tok/s 2631436.30 | norm 0.3651 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16917 | train 3.003823 | val 3.327171 | tok/s 2632035.34 | norm 0.3567 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16918 | train 2.781161 | val 3.327171 | tok/s 2624332.92 | norm 0.5536 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16919 | train 2.944329 | val 3.327171 | tok/s 2629355.09 | norm 0.4024 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16920 | train 3.011402 | val 3.327171 | tok/s 2631394.72 | norm 0.3478 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16921 | train 3.212864 | val 3.327171 | tok/s 2627530.37 | norm 0.4136 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16922 | train 3.140579 | val 3.327171 | tok/s 2628447.35 | norm 0.4180 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16923 | train 2.954747 | val 3.327171 | tok/s 2628682.26 | norm 0.4193 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16924 | train 3.032147 | val 3.327171 | tok/s 2630004.99 | norm 0.3808 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16925 | train 2.956361 | val 3.327171 | tok/s 2632206.66 | norm 0.3192 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16926 | train 2.937929 | val 3.327171 | tok/s 2629729.50 | norm 0.3643 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16927 | train 2.967939 | val 3.327171 | tok/s 2632965.07 | norm 0.3752 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16928 | train 3.065583 | val 3.327171 | tok/s 2633704.17 | norm 0.3937 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16929 | train 3.094337 | val 3.327171 | tok/s 2628295.94 | norm 0.3601 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16930 | train 3.009844 | val 3.327171 | tok/s 2588246.48 | norm 0.3580 | dt 202.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16931 | train 3.003410 | val 3.327171 | tok/s 2631406.61 | norm 0.4091 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16932 | train 2.985120 | val 3.327171 | tok/s 2633265.32 | norm 0.3585 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16933 | train 3.023083 | val 3.327171 | tok/s 2630058.82 | norm 0.3460 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16934 | train 3.157884 | val 3.327171 | tok/s 2627748.94 | norm 0.3846 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16935 | train 2.990239 | val 3.327171 | tok/s 2627428.76 | norm 0.3552 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16936 | train 3.083383 | val 3.327171 | tok/s 2628748.04 | norm 0.3748 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16937 | train 2.983955 | val 3.327171 | tok/s 2626074.81 | norm 0.3756 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16938 | train 3.011472 | val 3.327171 | tok/s 2625567.69 | norm 0.3686 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16939 | train 2.933924 | val 3.327171 | tok/s 2625796.45 | norm 0.3683 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16940 | train 2.987345 | val 3.327171 | tok/s 2627827.72 | norm 0.3804 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16941 | train 2.873422 | val 3.327171 | tok/s 2627359.14 | norm 0.3797 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16942 | train 2.915030 | val 3.327171 | tok/s 2628926.13 | norm 0.3863 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16943 | train 2.952413 | val 3.327171 | tok/s 2628019.26 | norm 0.3789 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16944 | train 2.870230 | val 3.327171 | tok/s 2629273.08 | norm 0.3760 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16945 | train 2.953276 | val 3.327171 | tok/s 2627027.69 | norm 0.3813 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16946 | train 2.965156 | val 3.327171 | tok/s 2629682.85 | norm 0.3889 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16947 | train 2.886362 | val 3.327171 | tok/s 2627989.87 | norm 0.3687 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16948 | train 2.927851 | val 3.327171 | tok/s 2630552.99 | norm 0.3645 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16949 | train 2.891602 | val 3.327171 | tok/s 2629642.24 | norm 0.4151 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16950 | train 2.871442 | val 3.327171 | tok/s 2631445.12 | norm 0.3723 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16951 | train 2.977794 | val 3.327171 | tok/s 2630114.41 | norm 0.3470 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16952 | train 2.888618 | val 3.327171 | tok/s 2626591.78 | norm 0.3436 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16953 | train 3.048376 | val 3.327171 | tok/s 2631081.66 | norm 0.3647 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16954 | train 2.961723 | val 3.327171 | tok/s 2630162.51 | norm 0.4012 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16955 | train 2.889626 | val 3.327171 | tok/s 2628841.09 | norm 0.3852 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16956 | train 2.830182 | val 3.327171 | tok/s 2632399.02 | norm 0.3767 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16957 | train 2.923322 | val 3.327171 | tok/s 2628854.30 | norm 0.3861 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16958 | train 2.829908 | val 3.327171 | tok/s 2631404.05 | norm 0.3844 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16959 | train 2.914480 | val 3.327171 | tok/s 2626957.52 | norm 0.3856 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16960 | train 3.003332 | val 3.327171 | tok/s 2629347.37 | norm 0.3761 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16961 | train 2.883393 | val 3.327171 | tok/s 2629593.87 | norm 0.3680 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16962 | train 2.995405 | val 3.327171 | tok/s 2632409.63 | norm 0.3648 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16963 | train 2.949244 | val 3.327171 | tok/s 2628544.06 | norm 0.3791 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16964 | train 2.884835 | val 3.327171 | tok/s 2631589.46 | norm 0.3551 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16965 | train 2.918760 | val 3.327171 | tok/s 2628527.54 | norm 0.3593 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16966 | train 2.917955 | val 3.327171 | tok/s 2633329.84 | norm 0.3933 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16967 | train 3.038241 | val 3.327171 | tok/s 2627773.72 | norm 0.3760 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16968 | train 2.980327 | val 3.327171 | tok/s 2631026.02 | norm 0.3568 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16969 | train 3.125780 | val 3.327171 | tok/s 2629538.52 | norm 0.3890 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16970 | train 2.871285 | val 3.327171 | tok/s 2627227.60 | norm 0.3432 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16971 | train 2.869488 | val 3.327171 | tok/s 2626631.46 | norm 0.3667 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16972 | train 2.935949 | val 3.327171 | tok/s 2630258.81 | norm 0.3872 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16973 | train 3.004435 | val 3.327171 | tok/s 2630167.32 | norm 0.4033 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16974 | train 3.015793 | val 3.327171 | tok/s 2631498.12 | norm 0.4231 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16975 | train 2.997052 | val 3.327171 | tok/s 2632027.62 | norm 0.3863 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16976 | train 3.044968 | val 3.327171 | tok/s 2627291.25 | norm 0.3639 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16977 | train 2.999880 | val 3.327171 | tok/s 2630591.58 | norm 0.3659 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16978 | train 2.940322 | val 3.327171 | tok/s 2632142.44 | norm 0.3638 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16979 | train 2.902441 | val 3.327171 | tok/s 2554716.09 | norm 0.3475 | dt 205.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16980 | train 2.912079 | val 3.327171 | tok/s 2625653.45 | norm 0.3180 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16981 | train 2.858276 | val 3.327171 | tok/s 2631286.45 | norm 0.3303 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16982 | train 3.027005 | val 3.327171 | tok/s 2630476.03 | norm 0.3446 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16983 | train 2.887418 | val 3.327171 | tok/s 2628248.40 | norm 0.3583 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16984 | train 2.919057 | val 3.327171 | tok/s 2628687.92 | norm 0.3665 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16985 | train 3.120631 | val 3.327171 | tok/s 2629944.53 | norm 0.4250 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16986 | train 2.934637 | val 3.327171 | tok/s 2629540.31 | norm 0.3787 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16987 | train 3.018332 | val 3.327171 | tok/s 2629721.69 | norm 0.3446 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16988 | train 2.995001 | val 3.327171 | tok/s 2631150.34 | norm 0.4273 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16989 | train 2.923954 | val 3.327171 | tok/s 2630217.89 | norm 0.3891 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16990 | train 2.923130 | val 3.327171 | tok/s 2628046.78 | norm 0.3762 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16991 | train 2.958458 | val 3.327171 | tok/s 2635284.65 | norm 0.3759 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16992 | train 2.954651 | val 3.327171 | tok/s 2634026.78 | norm 0.3850 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16993 | train 2.992196 | val 3.327171 | tok/s 2629260.37 | norm 0.3910 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16994 | train 2.950177 | val 3.327171 | tok/s 2628933.67 | norm 0.4092 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16995 | train 2.905854 | val 3.327171 | tok/s 2630562.86 | norm 0.3912 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16996 | train 2.965218 | val 3.327171 | tok/s 2630055.65 | norm 0.3847 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16997 | train 2.974689 | val 3.327171 | tok/s 2630855.75 | norm 0.3648 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16998 | train 3.010570 | val 3.327171 | tok/s 2632706.69 | norm 0.3677 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 16999 | train 2.979135 | val 3.327171 | tok/s 2628964.02 | norm 0.3548 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 17000 | val_loss 3.326383 step 17000 | train 2.911856 | val 3.326383 | tok/s 2613105.67 | norm 0.3615 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17001 | train 2.794895 | val 3.326383 | tok/s 2627130.34 | norm 0.3589 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17002 | train 2.935936 | val 3.326383 | tok/s 2630292.63 | norm 0.3760 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17003 | train 2.824749 | val 3.326383 | tok/s 2626664.86 | norm 0.3780 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17004 | train 3.043532 | val 3.326383 | tok/s 2632401.39 | norm 0.3894 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17005 | train 2.929983 | val 3.326383 | tok/s 2631622.87 | norm 0.3490 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17006 | train 3.071404 | val 3.326383 | tok/s 2631375.57 | norm 0.3399 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17007 | train 2.863545 | val 3.326383 | tok/s 2630498.54 | norm 0.3503 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17008 | train 3.038408 | val 3.326383 | tok/s 2630940.87 | norm 0.3755 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17009 | train 3.136772 | val 3.326383 | tok/s 2632677.72 | norm 0.4025 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17010 | train 3.120668 | val 3.326383 | tok/s 2612730.67 | norm 0.3852 | dt 200.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17011 | train 2.934585 | val 3.326383 | tok/s 2627698.42 | norm 0.3564 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17012 | train 2.943508 | val 3.326383 | tok/s 2634647.92 | norm 0.3502 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17013 | train 2.875884 | val 3.326383 | tok/s 2627777.58 | norm 0.3665 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17014 | train 2.872412 | val 3.326383 | tok/s 2629537.85 | norm 0.4015 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17015 | train 2.903357 | val 3.326383 | tok/s 2629758.14 | norm 0.3888 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17016 | train 2.950169 | val 3.326383 | tok/s 2631913.24 | norm 0.3628 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17017 | train 2.897824 | val 3.326383 | tok/s 2630348.93 | norm 0.3739 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17018 | train 2.923999 | val 3.326383 | tok/s 2633434.90 | norm 0.3780 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17019 | train 2.852564 | val 3.326383 | tok/s 2632850.46 | norm 0.3548 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17020 | train 3.045758 | val 3.326383 | tok/s 2629158.31 | norm 0.4000 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17021 | train 3.067791 | val 3.326383 | tok/s 2628449.63 | norm 0.3726 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17022 | train 2.904727 | val 3.326383 | tok/s 2634161.19 | norm 0.3711 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17023 | train 2.884132 | val 3.326383 | tok/s 2632641.36 | norm 0.3533 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17024 | train 3.152978 | val 3.326383 | tok/s 2629883.46 | norm 0.3748 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17025 | train 3.003006 | val 3.326383 | tok/s 2629518.83 | norm 0.4084 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17026 | train 2.990170 | val 3.326383 | tok/s 2634626.19 | norm 0.3738 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17027 | train 2.973010 | val 3.326383 | tok/s 2634494.22 | norm 0.3474 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17028 | train 2.883883 | val 3.326383 | tok/s 2628597.78 | norm 0.3396 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17029 | train 2.958229 | val 3.326383 | tok/s 2632537.80 | norm 0.3586 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17030 | train 2.923910 | val 3.326383 | tok/s 2631854.23 | norm 0.3665 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17031 | train 2.967314 | val 3.326383 | tok/s 2627706.25 | norm 0.3999 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17032 | train 2.923741 | val 3.326383 | tok/s 2632858.31 | norm 0.4034 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17033 | train 2.967393 | val 3.326383 | tok/s 2628250.92 | norm 0.3655 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17034 | train 2.909706 | val 3.326383 | tok/s 2630921.54 | norm 0.3678 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17035 | train 2.947123 | val 3.326383 | tok/s 2615712.82 | norm 0.3785 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17036 | train 2.955199 | val 3.326383 | tok/s 2630743.35 | norm 0.3648 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17037 | train 2.949877 | val 3.326383 | tok/s 2631843.82 | norm 0.3957 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17038 | train 3.076876 | val 3.326383 | tok/s 2630551.08 | norm 0.3739 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17039 | train 2.928607 | val 3.326383 | tok/s 2628631.98 | norm 0.3423 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17040 | train 3.011681 | val 3.326383 | tok/s 2630752.24 | norm 0.3728 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17041 | train 2.845869 | val 3.326383 | tok/s 2630775.81 | norm 0.3945 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17042 | train 2.911865 | val 3.326383 | tok/s 2629286.54 | norm 0.3604 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17043 | train 3.034222 | val 3.326383 | tok/s 2629391.54 | norm 0.3327 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17044 | train 2.902847 | val 3.326383 | tok/s 2634396.04 | norm 0.3766 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17045 | train 3.007694 | val 3.326383 | tok/s 2628267.41 | norm 0.3459 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17046 | train 2.807745 | val 3.326383 | tok/s 2632225.56 | norm 0.3370 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17047 | train 3.117582 | val 3.326383 | tok/s 2633541.03 | norm 0.3779 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17048 | train 3.063769 | val 3.326383 | tok/s 2632188.07 | norm 0.3254 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17049 | train 2.965303 | val 3.326383 | tok/s 2632847.07 | norm 0.3553 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17050 | train 3.052953 | val 3.326383 | tok/s 2623776.81 | norm 0.3878 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17051 | train 3.083256 | val 3.326383 | tok/s 2633112.59 | norm 0.4344 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17052 | train 2.869624 | val 3.326383 | tok/s 2629253.18 | norm 0.4130 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17053 | train 2.927026 | val 3.326383 | tok/s 2633178.49 | norm 0.3767 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17054 | train 3.058281 | val 3.326383 | tok/s 2633171.08 | norm 0.3971 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17055 | train 2.924174 | val 3.326383 | tok/s 2629527.65 | norm 0.4046 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17056 | train 2.963650 | val 3.326383 | tok/s 2633048.68 | norm 0.3697 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17057 | train 2.903162 | val 3.326383 | tok/s 2630420.53 | norm 0.3739 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17058 | train 2.983927 | val 3.326383 | tok/s 2634577.29 | norm 0.3768 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17059 | train 3.010649 | val 3.326383 | tok/s 2634440.65 | norm 0.3789 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17060 | train 3.011373 | val 3.326383 | tok/s 2618826.41 | norm 0.4007 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17061 | train 2.888108 | val 3.326383 | tok/s 2632404.42 | norm 0.3740 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17062 | train 2.903132 | val 3.326383 | tok/s 2628833.69 | norm 0.3602 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17063 | train 3.010993 | val 3.326383 | tok/s 2630333.17 | norm 0.3819 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17064 | train 3.050196 | val 3.326383 | tok/s 2632514.57 | norm 0.3697 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17065 | train 2.931901 | val 3.326383 | tok/s 2630377.15 | norm 0.3689 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17066 | train 3.102606 | val 3.326383 | tok/s 2634824.65 | norm 0.3490 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17067 | train 2.985965 | val 3.326383 | tok/s 2635692.65 | norm 0.3342 | dt 198.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17068 | train 2.970849 | val 3.326383 | tok/s 2631669.55 | norm 0.3477 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17069 | train 3.015094 | val 3.326383 | tok/s 2624656.24 | norm 0.3721 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17070 | train 2.965829 | val 3.326383 | tok/s 2631901.95 | norm 0.3591 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17071 | train 3.017514 | val 3.326383 | tok/s 2634058.03 | norm 0.3793 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17072 | train 2.922037 | val 3.326383 | tok/s 2635144.40 | norm 0.3552 | dt 198.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17073 | train 2.985879 | val 3.326383 | tok/s 2632872.38 | norm 0.3500 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17074 | train 2.811838 | val 3.326383 | tok/s 2627802.04 | norm 0.3623 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17075 | train 2.955728 | val 3.326383 | tok/s 2632878.54 | norm 0.3407 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17076 | train 2.915632 | val 3.326383 | tok/s 2629163.64 | norm 0.3480 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17077 | train 2.916566 | val 3.326383 | tok/s 2630466.85 | norm 0.3954 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17078 | train 2.859851 | val 3.326383 | tok/s 2631620.90 | norm 0.4371 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17079 | train 2.948290 | val 3.326383 | tok/s 2629936.85 | norm 0.3720 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17080 | train 2.960445 | val 3.326383 | tok/s 2627237.72 | norm 0.3479 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17081 | train 2.952109 | val 3.326383 | tok/s 2630332.83 | norm 0.3726 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17082 | train 2.961955 | val 3.326383 | tok/s 2628929.72 | norm 0.3837 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17083 | train 3.046447 | val 3.326383 | tok/s 2626648.42 | norm 0.3856 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17084 | train 2.919698 | val 3.326383 | tok/s 2627090.94 | norm 0.3306 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17085 | train 2.851645 | val 3.326383 | tok/s 2624069.46 | norm 0.3624 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17086 | train 2.994054 | val 3.326383 | tok/s 2485852.77 | norm 0.3981 | dt 210.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17087 | train 3.217920 | val 3.326383 | tok/s 2629884.95 | norm 0.4493 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17088 | train 3.079138 | val 3.326383 | tok/s 2629827.07 | norm 0.3875 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17089 | train 2.979985 | val 3.326383 | tok/s 2628211.07 | norm 0.3557 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17090 | train 2.987539 | val 3.326383 | tok/s 2630036.05 | norm 0.3854 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17091 | train 2.901437 | val 3.326383 | tok/s 2630254.81 | norm 0.4096 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17092 | train 2.873328 | val 3.326383 | tok/s 2632773.39 | norm 0.4050 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17093 | train 2.988117 | val 3.326383 | tok/s 2565860.31 | norm 0.3876 | dt 204.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17094 | train 2.904993 | val 3.326383 | tok/s 2633351.95 | norm 0.4116 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17095 | train 2.908751 | val 3.326383 | tok/s 2626201.53 | norm 0.4035 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17096 | train 2.865373 | val 3.326383 | tok/s 2628452.41 | norm 0.3545 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17097 | train 2.981462 | val 3.326383 | tok/s 2628352.22 | norm 0.3458 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17098 | train 3.200163 | val 3.326383 | tok/s 2631832.07 | norm 0.4340 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17099 | train 2.951018 | val 3.326383 | tok/s 2628959.77 | norm 0.4007 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17100 | train 2.991582 | val 3.326383 | tok/s 2630302.65 | norm 0.3619 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17101 | train 3.015012 | val 3.326383 | tok/s 2629788.79 | norm 0.3484 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17102 | train 2.859743 | val 3.326383 | tok/s 2631710.45 | norm 0.3883 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17103 | train 3.047860 | val 3.326383 | tok/s 2626537.68 | norm 0.3785 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17104 | train 2.996034 | val 3.326383 | tok/s 2631609.09 | norm 0.3450 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17105 | train 2.915681 | val 3.326383 | tok/s 2628930.89 | norm 0.3473 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17106 | train 2.923790 | val 3.326383 | tok/s 2629310.90 | norm 0.3764 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17107 | train 2.874994 | val 3.326383 | tok/s 2627456.85 | norm 0.4043 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17108 | train 2.863663 | val 3.326383 | tok/s 2627956.93 | norm 0.3826 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17109 | train 3.028082 | val 3.326383 | tok/s 2633309.49 | norm 0.3533 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17110 | train 2.840425 | val 3.326383 | tok/s 2625479.60 | norm 0.3499 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17111 | train 2.971998 | val 3.326383 | tok/s 2628788.90 | norm 0.3667 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17112 | train 2.845879 | val 3.326383 | tok/s 2629701.27 | norm 0.3758 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17113 | train 2.975082 | val 3.326383 | tok/s 2630349.04 | norm 0.3698 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17114 | train 2.902572 | val 3.326383 | tok/s 2630449.79 | norm 0.3854 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17115 | train 2.984006 | val 3.326383 | tok/s 2632147.18 | norm 0.4426 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17116 | train 2.996670 | val 3.326383 | tok/s 2626735.54 | norm 0.3611 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17117 | train 3.097127 | val 3.326383 | tok/s 2624853.98 | norm 0.4185 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17118 | train 2.991367 | val 3.326383 | tok/s 2626488.85 | norm 0.3812 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17119 | train 2.907776 | val 3.326383 | tok/s 2630909.34 | norm 0.3576 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17120 | train 2.979179 | val 3.326383 | tok/s 2630177.00 | norm 0.4213 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17121 | train 2.954865 | val 3.326383 | tok/s 2626979.55 | norm 0.3858 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17122 | train 3.026863 | val 3.326383 | tok/s 2627071.47 | norm 0.3604 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17123 | train 2.818194 | val 3.326383 | tok/s 2630671.14 | norm 0.3715 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17124 | train 2.963313 | val 3.326383 | tok/s 2633500.05 | norm 0.3927 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17125 | train 2.890324 | val 3.326383 | tok/s 2631781.88 | norm 0.4045 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17126 | train 2.995097 | val 3.326383 | tok/s 2628439.98 | norm 0.3904 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17127 | train 3.004492 | val 3.326383 | tok/s 2630482.20 | norm 0.3929 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17128 | train 3.012856 | val 3.326383 | tok/s 2628706.34 | norm 0.3634 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17129 | train 2.958702 | val 3.326383 | tok/s 2630168.62 | norm 0.3492 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17130 | train 3.133540 | val 3.326383 | tok/s 2629969.95 | norm 0.5213 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17131 | train 2.893694 | val 3.326383 | tok/s 2630431.42 | norm 0.3718 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17132 | train 2.898466 | val 3.326383 | tok/s 2633739.30 | norm 0.3941 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17133 | train 2.915278 | val 3.326383 | tok/s 2625215.12 | norm 0.3971 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17134 | train 2.966908 | val 3.326383 | tok/s 2630034.24 | norm 0.4640 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17135 | train 3.036180 | val 3.326383 | tok/s 2629815.13 | norm 0.4057 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17136 | train 3.015114 | val 3.326383 | tok/s 2631509.29 | norm 0.3591 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17137 | train 2.908131 | val 3.326383 | tok/s 2628551.18 | norm 0.3762 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17138 | train 2.941504 | val 3.326383 | tok/s 2630312.35 | norm 0.3825 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17139 | train 3.057620 | val 3.326383 | tok/s 2633635.05 | norm 0.3854 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17140 | train 2.980714 | val 3.326383 | tok/s 2630452.92 | norm 0.3648 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17141 | train 2.915072 | val 3.326383 | tok/s 2627079.76 | norm 0.3956 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17142 | train 2.964921 | val 3.326383 | tok/s 2629939.53 | norm 0.3866 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17143 | train 3.032489 | val 3.326383 | tok/s 2632676.61 | norm 0.3665 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17144 | train 2.956948 | val 3.326383 | tok/s 2629006.93 | norm 0.3780 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17145 | train 2.918685 | val 3.326383 | tok/s 2629886.08 | norm 0.3780 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17146 | train 2.907330 | val 3.326383 | tok/s 2626719.36 | norm 0.4008 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17147 | train 2.875517 | val 3.326383 | tok/s 2629485.03 | norm 0.3774 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17148 | train 3.056014 | val 3.326383 | tok/s 2630631.44 | norm 0.3588 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17149 | train 2.959510 | val 3.326383 | tok/s 2630123.09 | norm 0.3555 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17150 | train 2.902469 | val 3.326383 | tok/s 2630350.34 | norm 0.4093 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17151 | train 3.137367 | val 3.326383 | tok/s 2630726.03 | norm 0.3849 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17152 | train 2.935156 | val 3.326383 | tok/s 2631748.23 | norm 0.3498 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17153 | train 2.931596 | val 3.326383 | tok/s 2628065.71 | norm 0.3168 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17154 | train 2.878848 | val 3.326383 | tok/s 2630718.52 | norm 0.3678 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17155 | train 3.009746 | val 3.326383 | tok/s 2628332.46 | norm 0.4051 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17156 | train 2.937948 | val 3.326383 | tok/s 2626118.27 | norm 0.3826 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17157 | train 2.950717 | val 3.326383 | tok/s 2628305.12 | norm 0.3603 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17158 | train 2.890867 | val 3.326383 | tok/s 2630907.85 | norm 0.3925 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17159 | train 2.993519 | val 3.326383 | tok/s 2625434.30 | norm 0.4330 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17160 | train 2.914180 | val 3.326383 | tok/s 2629304.62 | norm 0.4646 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17161 | train 2.960021 | val 3.326383 | tok/s 2626840.43 | norm 0.4119 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17162 | train 2.874238 | val 3.326383 | tok/s 2620300.00 | norm 0.3616 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17163 | train 2.872293 | val 3.326383 | tok/s 2626609.90 | norm 0.3654 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17164 | train 2.990902 | val 3.326383 | tok/s 2484259.88 | norm 0.3739 | dt 211.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17165 | train 3.072466 | val 3.326383 | tok/s 2562666.54 | norm 0.3611 | dt 204.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17166 | train 2.993819 | val 3.326383 | tok/s 2623565.10 | norm 0.3733 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17167 | train 2.900230 | val 3.326383 | tok/s 2625221.37 | norm 0.3674 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17168 | train 3.001517 | val 3.326383 | tok/s 2627160.52 | norm 0.3592 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17169 | train 2.912546 | val 3.326383 | tok/s 2625681.00 | norm 0.3373 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17170 | train 3.090400 | val 3.326383 | tok/s 2631568.79 | norm 0.3359 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17171 | train 2.938754 | val 3.326383 | tok/s 2629062.15 | norm 0.3668 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17172 | train 2.968022 | val 3.326383 | tok/s 2630333.08 | norm 0.3542 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17173 | train 3.007512 | val 3.326383 | tok/s 2627804.73 | norm 0.3844 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17174 | train 3.029432 | val 3.326383 | tok/s 2629788.04 | norm 0.3872 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17175 | train 3.007732 | val 3.326383 | tok/s 2628873.11 | norm 0.3684 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17176 | train 2.955302 | val 3.326383 | tok/s 2631973.50 | norm 0.4013 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17177 | train 2.949550 | val 3.326383 | tok/s 2623090.16 | norm 0.4282 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17178 | train 3.057595 | val 3.326383 | tok/s 2629081.93 | norm 0.4102 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17179 | train 2.941626 | val 3.326383 | tok/s 2628569.86 | norm 0.3958 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17180 | train 2.868817 | val 3.326383 | tok/s 2629930.16 | norm 0.4979 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17181 | train 3.083370 | val 3.326383 | tok/s 2628922.00 | norm 0.3507 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17182 | train 2.977126 | val 3.326383 | tok/s 2631188.65 | norm 0.3300 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17183 | train 2.967851 | val 3.326383 | tok/s 2629722.96 | norm 0.3842 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17184 | train 2.891262 | val 3.326383 | tok/s 2626951.08 | norm 0.3640 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17185 | train 2.968487 | val 3.326383 | tok/s 2616175.03 | norm 0.3535 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17186 | train 2.852462 | val 3.326383 | tok/s 2630547.28 | norm 0.3454 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17187 | train 2.909827 | val 3.326383 | tok/s 2630129.26 | norm 0.3785 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17188 | train 3.041779 | val 3.326383 | tok/s 2624474.21 | norm 0.3820 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17189 | train 2.948359 | val 3.326383 | tok/s 2627629.30 | norm 0.3429 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17190 | train 2.981510 | val 3.326383 | tok/s 2628614.36 | norm 0.3565 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17191 | train 3.061107 | val 3.326383 | tok/s 2627530.12 | norm 0.3685 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17192 | train 2.866123 | val 3.326383 | tok/s 2627976.59 | norm 0.3737 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17193 | train 3.068760 | val 3.326383 | tok/s 2630317.25 | norm 0.3641 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17194 | train 2.774531 | val 3.326383 | tok/s 2629397.74 | norm 0.4086 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17195 | train 3.089750 | val 3.326383 | tok/s 2629636.76 | norm 0.4430 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17196 | train 3.051867 | val 3.326383 | tok/s 2628560.07 | norm 0.3758 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17197 | train 2.956622 | val 3.326383 | tok/s 2630745.12 | norm 0.3268 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17198 | train 2.983172 | val 3.326383 | tok/s 2632341.31 | norm 0.3643 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17199 | train 2.935944 | val 3.326383 | tok/s 2627799.39 | norm 0.3912 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17200 | train 2.945670 | val 3.326383 | tok/s 2630532.58 | norm 0.3676 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17201 | train 2.882850 | val 3.326383 | tok/s 2628781.36 | norm 0.3833 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17202 | train 3.030691 | val 3.326383 | tok/s 2630594.11 | norm 0.3437 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17203 | train 2.930086 | val 3.326383 | tok/s 2634696.30 | norm 0.3536 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17204 | train 2.940107 | val 3.326383 | tok/s 2629822.64 | norm 0.3465 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17205 | train 3.022506 | val 3.326383 | tok/s 2629805.66 | norm 0.3557 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17206 | train 2.901686 | val 3.326383 | tok/s 2630956.80 | norm 0.3570 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17207 | train 2.944605 | val 3.326383 | tok/s 2630099.59 | norm 0.3976 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17208 | train 3.008990 | val 3.326383 | tok/s 2631136.40 | norm 0.3884 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17209 | train 2.982994 | val 3.326383 | tok/s 2631078.32 | norm 0.3826 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17210 | train 3.021127 | val 3.326383 | tok/s 2629521.20 | norm 0.3899 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17211 | train 3.003786 | val 3.326383 | tok/s 2625481.61 | norm 0.3590 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17212 | train 3.132311 | val 3.326383 | tok/s 2631335.27 | norm 0.4612 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17213 | train 3.016268 | val 3.326383 | tok/s 2631226.86 | norm 0.4179 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17214 | train 2.931571 | val 3.326383 | tok/s 2630051.33 | norm 0.3806 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17215 | train 2.884433 | val 3.326383 | tok/s 2629422.80 | norm 0.3823 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17216 | train 2.905683 | val 3.326383 | tok/s 2632464.01 | norm 0.3613 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17217 | train 3.044872 | val 3.326383 | tok/s 2627419.60 | norm 0.3568 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17218 | train 2.980785 | val 3.326383 | tok/s 2628331.06 | norm 0.4149 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17219 | train 3.033385 | val 3.326383 | tok/s 2630538.82 | norm 0.4182 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17220 | train 3.002101 | val 3.326383 | tok/s 2630192.28 | norm 0.3756 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17221 | train 2.890898 | val 3.326383 | tok/s 2630916.13 | norm 0.3910 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17222 | train 3.004828 | val 3.326383 | tok/s 2630469.44 | norm 0.4202 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17223 | train 2.930918 | val 3.326383 | tok/s 2628495.76 | norm 0.3536 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17224 | train 2.847360 | val 3.326383 | tok/s 2629583.93 | norm 0.3355 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17225 | train 2.894658 | val 3.326383 | tok/s 2625867.50 | norm 0.3669 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17226 | train 2.953000 | val 3.326383 | tok/s 2631770.50 | norm 0.3604 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17227 | train 2.962336 | val 3.326383 | tok/s 2630622.46 | norm 0.3855 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17228 | train 2.894449 | val 3.326383 | tok/s 2628429.30 | norm 0.3818 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17229 | train 2.898577 | val 3.326383 | tok/s 2629977.08 | norm 0.3910 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17230 | train 2.941564 | val 3.326383 | tok/s 2630597.25 | norm 0.3776 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17231 | train 3.178957 | val 3.326383 | tok/s 2629927.79 | norm 0.5784 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17232 | train 2.895235 | val 3.326383 | tok/s 2630208.00 | norm 0.4183 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17233 | train 2.983379 | val 3.326383 | tok/s 2627839.12 | norm 0.3162 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17234 | train 2.873116 | val 3.326383 | tok/s 2628518.92 | norm 0.3354 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17235 | train 2.855149 | val 3.326383 | tok/s 2629245.55 | norm 0.3898 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17236 | train 3.060928 | val 3.326383 | tok/s 2631531.41 | norm 0.3848 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17237 | train 3.020345 | val 3.326383 | tok/s 2630376.35 | norm 0.3693 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17238 | train 3.033592 | val 3.326383 | tok/s 2629717.48 | norm 0.3457 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17239 | train 2.918335 | val 3.326383 | tok/s 2631663.50 | norm 0.3415 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17240 | train 2.922184 | val 3.326383 | tok/s 2629457.04 | norm 0.3404 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17241 | train 2.900149 | val 3.326383 | tok/s 2633208.96 | norm 0.3526 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17242 | train 3.085260 | val 3.326383 | tok/s 2622657.54 | norm 0.3606 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17243 | train 3.076647 | val 3.326383 | tok/s 2631992.46 | norm 0.3798 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17244 | train 2.898049 | val 3.326383 | tok/s 2629271.63 | norm 0.3558 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17245 | train 2.936140 | val 3.326383 | tok/s 2630103.47 | norm 0.3575 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17246 | train 2.957601 | val 3.326383 | tok/s 2622196.16 | norm 0.3837 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17247 | train 2.963068 | val 3.326383 | tok/s 2625865.11 | norm 0.4004 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17248 | train 2.919035 | val 3.326383 | tok/s 2629673.78 | norm 0.4211 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17249 | train 2.973649 | val 3.326383 | tok/s 2629408.26 | norm 0.3833 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 17250 | val_loss 3.331177 step 17250 | train 2.784407 | val 3.331177 | tok/s 2613099.72 | norm 0.3666 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17251 | train 2.901735 | val 3.331177 | tok/s 2629288.49 | norm 0.3831 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17252 | train 3.020449 | val 3.331177 | tok/s 2626868.79 | norm 0.3914 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17253 | train 2.909624 | val 3.331177 | tok/s 2626463.22 | norm 0.3467 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17254 | train 2.957018 | val 3.331177 | tok/s 2629629.97 | norm 0.3767 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17255 | train 2.872926 | val 3.331177 | tok/s 2631714.86 | norm 0.3483 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17256 | train 2.924825 | val 3.331177 | tok/s 2631021.96 | norm 0.3413 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17257 | train 2.975688 | val 3.331177 | tok/s 2617309.99 | norm 0.3420 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17258 | train 2.902345 | val 3.331177 | tok/s 2623866.12 | norm 0.3578 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17259 | train 2.827655 | val 3.331177 | tok/s 2630258.52 | norm 0.3709 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17260 | train 2.957361 | val 3.331177 | tok/s 2631361.36 | norm 0.3600 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17261 | train 2.978994 | val 3.331177 | tok/s 2630808.76 | norm 0.3597 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17262 | train 2.901107 | val 3.331177 | tok/s 2627040.44 | norm 0.3946 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17263 | train 3.001373 | val 3.331177 | tok/s 2630019.64 | norm 0.3771 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17264 | train 2.944112 | val 3.331177 | tok/s 2626105.95 | norm 0.3750 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17265 | train 2.835543 | val 3.331177 | tok/s 2604570.94 | norm 0.3778 | dt 201.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17266 | train 2.946458 | val 3.331177 | tok/s 2629923.98 | norm 0.3548 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17267 | train 2.924129 | val 3.331177 | tok/s 2625231.70 | norm 0.3457 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17268 | train 2.900948 | val 3.331177 | tok/s 2630046.23 | norm 0.3432 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17269 | train 2.864051 | val 3.331177 | tok/s 2629212.54 | norm 0.3769 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17270 | train 3.020952 | val 3.331177 | tok/s 2629239.51 | norm 0.3575 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17271 | train 2.987213 | val 3.331177 | tok/s 2628563.12 | norm 0.3540 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17272 | train 3.282847 | val 3.331177 | tok/s 2632019.97 | norm 0.4288 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17273 | train 3.013428 | val 3.331177 | tok/s 2628482.43 | norm 0.3866 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17274 | train 2.958701 | val 3.331177 | tok/s 2631171.29 | norm 0.3744 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17275 | train 2.880917 | val 3.331177 | tok/s 2629076.44 | norm 0.4021 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17276 | train 2.883104 | val 3.331177 | tok/s 2627281.47 | norm 0.3536 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17277 | train 3.005870 | val 3.331177 | tok/s 2628676.10 | norm 0.3593 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17278 | train 2.882163 | val 3.331177 | tok/s 2632116.43 | norm 0.3529 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17279 | train 2.895520 | val 3.331177 | tok/s 2629881.99 | norm 0.3742 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17280 | train 2.876132 | val 3.331177 | tok/s 2629462.70 | norm 0.3714 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17281 | train 2.999309 | val 3.331177 | tok/s 2627900.11 | norm 0.3568 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17282 | train 2.953650 | val 3.331177 | tok/s 2626051.06 | norm 0.3763 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17283 | train 2.874780 | val 3.331177 | tok/s 2628666.65 | norm 0.4142 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17284 | train 2.890386 | val 3.331177 | tok/s 2628866.75 | norm 0.4248 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17285 | train 3.066285 | val 3.331177 | tok/s 2627454.12 | norm 0.4176 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17286 | train 2.931920 | val 3.331177 | tok/s 2632258.17 | norm 0.3910 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17287 | train 2.949346 | val 3.331177 | tok/s 2630529.90 | norm 0.3664 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17288 | train 2.933396 | val 3.331177 | tok/s 2629515.64 | norm 0.3801 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17289 | train 3.028218 | val 3.331177 | tok/s 2184631.24 | norm 0.4079 | dt 239.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17290 | train 3.242536 | val 3.331177 | tok/s 2628342.96 | norm 0.4315 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17291 | train 2.869612 | val 3.331177 | tok/s 2631380.62 | norm 0.3648 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17292 | train 3.024879 | val 3.331177 | tok/s 2627773.89 | norm 0.3255 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17293 | train 2.945602 | val 3.331177 | tok/s 2630307.77 | norm 0.3912 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17294 | train 2.991233 | val 3.331177 | tok/s 2631971.20 | norm 0.3707 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17295 | train 2.865023 | val 3.331177 | tok/s 2631173.54 | norm 0.3569 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17296 | train 2.895515 | val 3.331177 | tok/s 2632876.98 | norm 0.3755 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17297 | train 3.024325 | val 3.331177 | tok/s 2615841.31 | norm 0.4071 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17298 | train 2.958375 | val 3.331177 | tok/s 2629920.96 | norm 0.4011 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17299 | train 2.908126 | val 3.331177 | tok/s 2625653.20 | norm 0.3941 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17300 | train 3.066968 | val 3.331177 | tok/s 2627583.32 | norm 0.3593 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17301 | train 2.916288 | val 3.331177 | tok/s 2625352.80 | norm 0.4075 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17302 | train 2.970384 | val 3.331177 | tok/s 2626626.58 | norm 0.4065 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17303 | train 3.027624 | val 3.331177 | tok/s 2628662.34 | norm 0.3892 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17304 | train 2.957155 | val 3.331177 | tok/s 2628919.47 | norm 0.3896 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17305 | train 3.011422 | val 3.331177 | tok/s 2631567.38 | norm 0.4029 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17306 | train 3.052319 | val 3.331177 | tok/s 2626068.81 | norm 0.3922 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17307 | train 2.931711 | val 3.331177 | tok/s 2629543.86 | norm 0.3159 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17308 | train 2.908139 | val 3.331177 | tok/s 2625912.16 | norm 0.3552 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17309 | train 2.887990 | val 3.331177 | tok/s 2629029.01 | norm 0.3789 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17310 | train 2.908471 | val 3.331177 | tok/s 2626663.80 | norm 0.3698 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17311 | train 2.953729 | val 3.331177 | tok/s 2630698.11 | norm 0.3815 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17312 | train 2.909184 | val 3.331177 | tok/s 2628607.52 | norm 0.4007 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17313 | train 3.009714 | val 3.331177 | tok/s 2629470.15 | norm 0.3626 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17314 | train 2.970280 | val 3.331177 | tok/s 2631436.19 | norm 0.3628 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17315 | train 2.949234 | val 3.331177 | tok/s 2629011.96 | norm 0.3960 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17316 | train 2.970926 | val 3.331177 | tok/s 2627334.20 | norm 0.3714 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17317 | train 3.275909 | val 3.331177 | tok/s 2630931.40 | norm 0.4919 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17318 | train 2.927999 | val 3.331177 | tok/s 2626213.01 | norm 0.4222 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17319 | train 2.954309 | val 3.331177 | tok/s 2630953.40 | norm 0.3871 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17320 | train 2.805126 | val 3.331177 | tok/s 2628943.48 | norm 0.3891 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17321 | train 2.901913 | val 3.331177 | tok/s 2632484.79 | norm 0.3841 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17322 | train 3.044888 | val 3.331177 | tok/s 2631383.25 | norm 0.4839 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17323 | train 2.847189 | val 3.331177 | tok/s 2626954.46 | norm 0.4225 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17324 | train 2.948044 | val 3.331177 | tok/s 2628905.55 | norm 0.3823 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17325 | train 3.146625 | val 3.331177 | tok/s 2634422.05 | norm 0.5007 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17326 | train 2.947673 | val 3.331177 | tok/s 2632455.40 | norm 0.3420 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17327 | train 2.929482 | val 3.331177 | tok/s 2628945.60 | norm 0.3397 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17328 | train 2.865439 | val 3.331177 | tok/s 2634093.16 | norm 0.3691 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17329 | train 2.819017 | val 3.331177 | tok/s 2630674.30 | norm 0.3605 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17330 | train 2.971406 | val 3.331177 | tok/s 2632517.22 | norm 0.3674 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17331 | train 2.990285 | val 3.331177 | tok/s 2633194.98 | norm 0.3515 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17332 | train 3.013836 | val 3.331177 | tok/s 2629431.74 | norm 0.3653 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17333 | train 2.923092 | val 3.331177 | tok/s 2625845.72 | norm 0.3728 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17334 | train 2.958934 | val 3.331177 | tok/s 2621927.34 | norm 0.3849 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17335 | train 3.163978 | val 3.331177 | tok/s 2627086.03 | norm 0.4358 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17336 | train 2.989110 | val 3.331177 | tok/s 2628587.79 | norm 0.4018 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17337 | train 2.968035 | val 3.331177 | tok/s 2622953.17 | norm 0.3785 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17338 | train 2.965728 | val 3.331177 | tok/s 2625040.03 | norm 0.3798 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17339 | train 2.972031 | val 3.331177 | tok/s 2627493.26 | norm 0.4203 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17340 | train 3.251760 | val 3.331177 | tok/s 2625333.30 | norm 0.4093 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17341 | train 2.979444 | val 3.331177 | tok/s 2571434.41 | norm 0.3576 | dt 203.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17342 | train 2.937808 | val 3.331177 | tok/s 2627646.05 | norm 0.3502 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17343 | train 3.110193 | val 3.331177 | tok/s 2629747.04 | norm 0.3970 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17344 | train 2.984459 | val 3.331177 | tok/s 2631713.38 | norm 0.3802 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17345 | train 3.001736 | val 3.331177 | tok/s 2628656.49 | norm 0.3843 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17346 | train 2.974235 | val 3.331177 | tok/s 2631121.07 | norm 0.3757 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17347 | train 2.894137 | val 3.331177 | tok/s 2526132.08 | norm 0.3783 | dt 207.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17348 | train 2.977278 | val 3.331177 | tok/s 2626101.61 | norm 0.4608 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17349 | train 2.907178 | val 3.331177 | tok/s 2627392.77 | norm 0.3213 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17350 | train 2.953017 | val 3.331177 | tok/s 2629206.31 | norm 0.3323 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17351 | train 2.828011 | val 3.331177 | tok/s 2623865.75 | norm 0.3613 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17352 | train 2.944191 | val 3.331177 | tok/s 2619885.92 | norm 0.3669 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17353 | train 2.835000 | val 3.331177 | tok/s 2629260.45 | norm 0.3693 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17354 | train 2.970296 | val 3.331177 | tok/s 2625025.67 | norm 0.3750 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17355 | train 3.035825 | val 3.331177 | tok/s 2628503.71 | norm 0.3715 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17356 | train 2.925124 | val 3.331177 | tok/s 2627822.99 | norm 0.3948 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17357 | train 2.966088 | val 3.331177 | tok/s 2624598.81 | norm 0.3792 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17358 | train 3.007929 | val 3.331177 | tok/s 2632310.88 | norm 0.3462 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17359 | train 2.894307 | val 3.331177 | tok/s 2625396.46 | norm 0.3543 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17360 | train 2.982715 | val 3.331177 | tok/s 2628360.39 | norm 0.3651 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17361 | train 2.890335 | val 3.331177 | tok/s 2628089.21 | norm 0.3284 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17362 | train 3.041638 | val 3.331177 | tok/s 2628450.29 | norm 0.3619 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17363 | train 2.983207 | val 3.331177 | tok/s 2626413.13 | norm 0.3621 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17364 | train 2.891128 | val 3.331177 | tok/s 2575472.52 | norm 0.3309 | dt 203.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17365 | train 3.023100 | val 3.331177 | tok/s 2630236.50 | norm 0.3741 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17366 | train 2.953544 | val 3.331177 | tok/s 2618648.17 | norm 0.3960 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17367 | train 2.868489 | val 3.331177 | tok/s 2625088.91 | norm 0.3661 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17368 | train 3.142865 | val 3.331177 | tok/s 2610886.35 | norm 0.3415 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17369 | train 3.011400 | val 3.331177 | tok/s 2627687.42 | norm 0.3169 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17370 | train 2.916881 | val 3.331177 | tok/s 2631616.79 | norm 0.3466 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17371 | train 2.983735 | val 3.331177 | tok/s 2628289.77 | norm 0.3841 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17372 | train 2.891948 | val 3.331177 | tok/s 2629921.13 | norm 0.3614 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17373 | train 2.955515 | val 3.331177 | tok/s 2624501.57 | norm 0.3502 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17374 | train 2.851771 | val 3.331177 | tok/s 2625888.75 | norm 0.3710 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17375 | train 2.910334 | val 3.331177 | tok/s 2627483.54 | norm 0.3670 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17376 | train 2.901056 | val 3.331177 | tok/s 2620690.26 | norm 0.3696 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17377 | train 2.987365 | val 3.331177 | tok/s 2628725.47 | norm 0.3804 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17378 | train 3.018390 | val 3.331177 | tok/s 2627764.00 | norm 0.3655 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17379 | train 2.969972 | val 3.331177 | tok/s 2626752.80 | norm 0.3909 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17380 | train 3.076623 | val 3.331177 | tok/s 2623916.72 | norm 0.3904 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17381 | train 2.863955 | val 3.331177 | tok/s 2625587.85 | norm 0.3854 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17382 | train 2.888194 | val 3.331177 | tok/s 2627710.48 | norm 0.3653 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17383 | train 2.940913 | val 3.331177 | tok/s 2628460.59 | norm 0.3921 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17384 | train 2.854328 | val 3.331177 | tok/s 2629900.71 | norm 0.3976 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17385 | train 3.060781 | val 3.331177 | tok/s 2624412.57 | norm 0.3779 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17386 | train 2.962052 | val 3.331177 | tok/s 2623485.62 | norm 0.3800 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17387 | train 2.947824 | val 3.331177 | tok/s 2626267.96 | norm 0.3914 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17388 | train 3.007829 | val 3.331177 | tok/s 2626672.19 | norm 0.3986 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17389 | train 2.955595 | val 3.331177 | tok/s 2630826.13 | norm 0.3993 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17390 | train 2.929504 | val 3.331177 | tok/s 2614088.44 | norm 0.4152 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17391 | train 3.061791 | val 3.331177 | tok/s 2628446.14 | norm 0.4380 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17392 | train 2.882100 | val 3.331177 | tok/s 2628586.30 | norm 0.3776 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17393 | train 2.740320 | val 3.331177 | tok/s 2625229.86 | norm 0.4644 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17394 | train 2.948756 | val 3.331177 | tok/s 2628978.77 | norm 0.4310 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17395 | train 2.969785 | val 3.331177 | tok/s 2628291.73 | norm 0.3558 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17396 | train 2.870057 | val 3.331177 | tok/s 2628743.35 | norm 0.3718 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17397 | train 3.024514 | val 3.331177 | tok/s 2628747.25 | norm 0.3580 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17398 | train 2.960724 | val 3.331177 | tok/s 2623959.55 | norm 0.3496 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17399 | train 3.083647 | val 3.331177 | tok/s 2624074.91 | norm 0.3553 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17400 | train 3.042766 | val 3.331177 | tok/s 2625766.53 | norm 0.3755 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17401 | train 3.000599 | val 3.331177 | tok/s 2623734.01 | norm 0.3586 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17402 | train 3.003622 | val 3.331177 | tok/s 2629055.66 | norm 0.3872 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17403 | train 2.867790 | val 3.331177 | tok/s 2630165.30 | norm 0.3584 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17404 | train 2.810872 | val 3.331177 | tok/s 2628945.01 | norm 0.3622 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17405 | train 2.958082 | val 3.331177 | tok/s 2625320.56 | norm 0.3588 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17406 | train 2.967669 | val 3.331177 | tok/s 2628545.60 | norm 0.3643 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17407 | train 3.112123 | val 3.331177 | tok/s 2628681.14 | norm 0.3561 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17408 | train 3.070538 | val 3.331177 | tok/s 2626358.70 | norm 0.3438 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17409 | train 2.851310 | val 3.331177 | tok/s 2632669.45 | norm 0.3783 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17410 | train 2.898618 | val 3.331177 | tok/s 2623104.32 | norm 0.3837 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17411 | train 2.915470 | val 3.331177 | tok/s 2624675.03 | norm 0.3567 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17412 | train 2.979106 | val 3.331177 | tok/s 2628107.25 | norm 0.3660 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17413 | train 2.990807 | val 3.331177 | tok/s 2629325.48 | norm 0.3879 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17414 | train 2.882735 | val 3.331177 | tok/s 2631538.86 | norm 0.4024 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17415 | train 3.146098 | val 3.331177 | tok/s 2627885.30 | norm 0.4383 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17416 | train 2.917701 | val 3.331177 | tok/s 2628629.35 | norm 0.3876 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17417 | train 3.041009 | val 3.331177 | tok/s 2631944.29 | norm 0.3444 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17418 | train 2.912868 | val 3.331177 | tok/s 2627297.10 | norm 0.3963 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17419 | train 2.910350 | val 3.331177 | tok/s 2628521.88 | norm 0.3820 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17420 | train 2.984255 | val 3.331177 | tok/s 2625380.10 | norm 0.3565 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17421 | train 3.030540 | val 3.331177 | tok/s 2626928.23 | norm 0.3899 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17422 | train 2.969118 | val 3.331177 | tok/s 2629818.87 | norm 0.4031 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17423 | train 2.968264 | val 3.331177 | tok/s 2629841.29 | norm 0.3757 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17424 | train 3.005689 | val 3.331177 | tok/s 2628297.37 | norm 0.3521 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17425 | train 3.001607 | val 3.331177 | tok/s 2631625.05 | norm 0.3923 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17426 | train 2.985393 | val 3.331177 | tok/s 2625273.82 | norm 0.3715 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17427 | train 3.012779 | val 3.331177 | tok/s 2626961.95 | norm 0.3642 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17428 | train 2.936957 | val 3.331177 | tok/s 2624144.51 | norm 0.3875 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17429 | train 3.002161 | val 3.331177 | tok/s 2620425.17 | norm 0.4149 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17430 | train 2.867846 | val 3.331177 | tok/s 2631799.46 | norm 0.3916 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17431 | train 3.010749 | val 3.331177 | tok/s 2627963.67 | norm 0.3450 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17432 | train 2.931918 | val 3.331177 | tok/s 2628344.31 | norm 0.3526 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17433 | train 2.997580 | val 3.331177 | tok/s 2625634.94 | norm 0.3667 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17434 | train 3.084765 | val 3.331177 | tok/s 2628918.54 | norm 0.3915 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17435 | train 2.949306 | val 3.331177 | tok/s 2524790.93 | norm 0.3421 | dt 207.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17436 | train 3.021894 | val 3.331177 | tok/s 2464750.52 | norm 0.3744 | dt 212.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17437 | train 2.885994 | val 3.331177 | tok/s 2614436.92 | norm 0.4110 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17438 | train 2.932172 | val 3.331177 | tok/s 2618190.54 | norm 0.3535 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17439 | train 2.848217 | val 3.331177 | tok/s 2625055.59 | norm 0.3343 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17440 | train 2.779802 | val 3.331177 | tok/s 2623992.03 | norm 0.3805 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17441 | train 2.915795 | val 3.331177 | tok/s 2627188.17 | norm 0.3920 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17442 | train 2.845837 | val 3.331177 | tok/s 2624982.82 | norm 0.4005 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17443 | train 2.979722 | val 3.331177 | tok/s 2627172.79 | norm 0.4074 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17444 | train 3.023536 | val 3.331177 | tok/s 2626625.67 | norm 0.3790 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17445 | train 3.050740 | val 3.331177 | tok/s 2624996.98 | norm 0.3456 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17446 | train 2.854347 | val 3.331177 | tok/s 2628524.37 | norm 0.3681 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17447 | train 2.774512 | val 3.331177 | tok/s 2628086.97 | norm 0.3834 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17448 | train 3.012747 | val 3.331177 | tok/s 2621169.22 | norm 0.3553 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17449 | train 3.020837 | val 3.331177 | tok/s 2628914.41 | norm 0.3730 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17450 | train 2.817693 | val 3.331177 | tok/s 2619636.42 | norm 0.5222 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17451 | train 3.001119 | val 3.331177 | tok/s 2626773.30 | norm 0.3727 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17452 | train 2.962469 | val 3.331177 | tok/s 2622804.58 | norm 0.3630 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17453 | train 2.906251 | val 3.331177 | tok/s 2624593.17 | norm 0.3980 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17454 | train 2.986828 | val 3.331177 | tok/s 2614382.98 | norm 0.3763 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17455 | train 2.981955 | val 3.331177 | tok/s 2629789.78 | norm 0.3944 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17456 | train 3.041275 | val 3.331177 | tok/s 2631837.88 | norm 0.3839 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17457 | train 2.978727 | val 3.331177 | tok/s 2629352.52 | norm 0.3519 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17458 | train 3.032228 | val 3.331177 | tok/s 2630317.72 | norm 0.3798 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17459 | train 2.990033 | val 3.331177 | tok/s 2626678.83 | norm 0.3583 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17460 | train 2.937424 | val 3.331177 | tok/s 2625629.30 | norm 0.3248 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17461 | train 2.940742 | val 3.331177 | tok/s 2625843.61 | norm 0.3557 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17462 | train 2.982350 | val 3.331177 | tok/s 2621705.54 | norm 0.3702 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17463 | train 3.019446 | val 3.331177 | tok/s 2626306.53 | norm 0.3827 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17464 | train 2.891241 | val 3.331177 | tok/s 2628021.33 | norm 0.3568 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17465 | train 2.983640 | val 3.331177 | tok/s 2629235.79 | norm 0.3759 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17466 | train 3.154386 | val 3.331177 | tok/s 2631808.46 | norm 0.4546 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17467 | train 2.922301 | val 3.331177 | tok/s 2624945.75 | norm 0.3438 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17468 | train 2.882820 | val 3.331177 | tok/s 2628491.24 | norm 0.4013 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17469 | train 3.042567 | val 3.331177 | tok/s 2628134.88 | norm 0.4153 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17470 | train 2.962298 | val 3.331177 | tok/s 2626883.63 | norm 0.3433 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17471 | train 2.944352 | val 3.331177 | tok/s 2629648.05 | norm 0.3524 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17472 | train 2.956261 | val 3.331177 | tok/s 2628853.70 | norm 0.4256 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17473 | train 2.952076 | val 3.331177 | tok/s 2538679.42 | norm 0.3258 | dt 206.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17474 | train 2.962134 | val 3.331177 | tok/s 2625936.18 | norm 0.3574 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17475 | train 2.934019 | val 3.331177 | tok/s 2625141.36 | norm 0.3877 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17476 | train 2.909581 | val 3.331177 | tok/s 2632437.43 | norm 0.3754 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17477 | train 2.939445 | val 3.331177 | tok/s 2505975.81 | norm 0.3790 | dt 209.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17478 | train 2.874837 | val 3.331177 | tok/s 2612021.60 | norm 0.4055 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17479 | train 2.875349 | val 3.331177 | tok/s 2621019.35 | norm 0.4006 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17480 | train 2.960933 | val 3.331177 | tok/s 2622344.07 | norm 0.3858 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17481 | train 2.997760 | val 3.331177 | tok/s 2628733.21 | norm 0.3932 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17482 | train 3.007989 | val 3.331177 | tok/s 2622628.73 | norm 0.3581 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17483 | train 2.988891 | val 3.331177 | tok/s 2626083.62 | norm 0.3348 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17484 | train 3.006244 | val 3.331177 | tok/s 2619208.02 | norm 0.4999 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17485 | train 2.825940 | val 3.331177 | tok/s 2624074.83 | norm 0.3337 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17486 | train 2.906607 | val 3.331177 | tok/s 2631076.26 | norm 0.3348 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17487 | train 2.932462 | val 3.331177 | tok/s 2627599.41 | norm 0.3817 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17488 | train 2.949110 | val 3.331177 | tok/s 2625903.02 | norm 0.3741 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17489 | train 3.023059 | val 3.331177 | tok/s 2623575.49 | norm 0.3815 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17490 | train 2.986752 | val 3.331177 | tok/s 2622665.91 | norm 0.3899 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17491 | train 2.891732 | val 3.331177 | tok/s 2628264.10 | norm 0.3591 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17492 | train 2.921669 | val 3.331177 | tok/s 2624802.91 | norm 0.3182 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17493 | train 3.105206 | val 3.331177 | tok/s 2627716.88 | norm 0.3543 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17494 | train 2.873219 | val 3.331177 | tok/s 2626032.04 | norm 0.3631 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17495 | train 2.830031 | val 3.331177 | tok/s 2628786.37 | norm 0.3764 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17496 | train 2.902446 | val 3.331177 | tok/s 2628091.03 | norm 0.3904 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17497 | train 2.856643 | val 3.331177 | tok/s 2627907.61 | norm 0.3792 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17498 | train 2.955010 | val 3.331177 | tok/s 2628681.36 | norm 0.3702 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17499 | train 2.983043 | val 3.331177 | tok/s 2623782.67 | norm 0.3567 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 17500 | val_loss 3.325012 step 17500 | train 2.926024 | val 3.325012 | tok/s 2608318.81 | norm 0.3666 | dt 201.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17501 | train 2.982258 | val 3.325012 | tok/s 2600466.26 | norm 0.3771 | dt 201.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17502 | train 2.906838 | val 3.325012 | tok/s 2627327.34 | norm 0.3310 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17503 | train 2.990910 | val 3.325012 | tok/s 2625191.95 | norm 0.3620 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17504 | train 3.021262 | val 3.325012 | tok/s 2627353.82 | norm 0.3868 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17505 | train 2.945887 | val 3.325012 | tok/s 2625075.53 | norm 0.3808 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17506 | train 2.911635 | val 3.325012 | tok/s 2599116.09 | norm 0.4339 | dt 201.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17507 | train 3.008769 | val 3.325012 | tok/s 2627250.85 | norm 0.3985 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17508 | train 2.995888 | val 3.325012 | tok/s 2624923.48 | norm 0.3828 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17509 | train 2.857755 | val 3.325012 | tok/s 2628579.04 | norm 0.3813 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17510 | train 3.058223 | val 3.325012 | tok/s 2620491.36 | norm 0.4037 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17511 | train 3.027802 | val 3.325012 | tok/s 2625445.50 | norm 0.3964 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17512 | train 2.984229 | val 3.325012 | tok/s 2627271.81 | norm 0.3812 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17513 | train 2.934000 | val 3.325012 | tok/s 2629095.87 | norm 0.3930 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17514 | train 2.947565 | val 3.325012 | tok/s 2627839.97 | norm 0.3854 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17515 | train 2.975156 | val 3.325012 | tok/s 2623964.71 | norm 0.3334 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17516 | train 2.933563 | val 3.325012 | tok/s 2622837.57 | norm 0.3721 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17517 | train 2.993063 | val 3.325012 | tok/s 2631756.75 | norm 0.3728 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17518 | train 3.038085 | val 3.325012 | tok/s 2629700.15 | norm 0.3987 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17519 | train 3.019662 | val 3.325012 | tok/s 2628445.31 | norm 0.4102 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17520 | train 2.918210 | val 3.325012 | tok/s 2626449.23 | norm 0.3653 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17521 | train 2.873861 | val 3.325012 | tok/s 2626991.81 | norm 0.3962 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17522 | train 3.078363 | val 3.325012 | tok/s 2630691.35 | norm 0.3787 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17523 | train 2.961234 | val 3.325012 | tok/s 2628831.70 | norm 0.3569 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17524 | train 3.024489 | val 3.325012 | tok/s 2628357.72 | norm 0.3822 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17525 | train 2.842921 | val 3.325012 | tok/s 2628351.29 | norm 0.4254 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17526 | train 2.986943 | val 3.325012 | tok/s 2629193.49 | norm 0.3676 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17527 | train 3.003100 | val 3.325012 | tok/s 2629477.46 | norm 0.4117 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17528 | train 2.866777 | val 3.325012 | tok/s 2622149.78 | norm 0.3593 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17529 | train 2.937253 | val 3.325012 | tok/s 2623345.68 | norm 0.3872 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17530 | train 2.945536 | val 3.325012 | tok/s 2627487.41 | norm 0.4208 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17531 | train 2.953300 | val 3.325012 | tok/s 2626192.21 | norm 0.3854 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17532 | train 2.999721 | val 3.325012 | tok/s 2628936.47 | norm 0.3613 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17533 | train 2.946743 | val 3.325012 | tok/s 2628173.26 | norm 0.3536 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17534 | train 2.997262 | val 3.325012 | tok/s 2626553.44 | norm 0.4170 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17535 | train 2.840950 | val 3.325012 | tok/s 2626172.57 | norm 0.3935 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17536 | train 2.891253 | val 3.325012 | tok/s 2624111.69 | norm 0.3773 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17537 | train 2.930374 | val 3.325012 | tok/s 2621274.52 | norm 0.3777 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17538 | train 3.184854 | val 3.325012 | tok/s 2619586.13 | norm 0.3746 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17539 | train 3.048003 | val 3.325012 | tok/s 2630038.98 | norm 0.3588 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17540 | train 2.904475 | val 3.325012 | tok/s 2629373.91 | norm 0.3595 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17541 | train 3.046808 | val 3.325012 | tok/s 2627141.13 | norm 0.3485 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17542 | train 2.941896 | val 3.325012 | tok/s 2634183.71 | norm 0.3735 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17543 | train 2.952584 | val 3.325012 | tok/s 2627712.82 | norm 0.3962 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17544 | train 2.921660 | val 3.325012 | tok/s 2625884.13 | norm 0.3629 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17545 | train 2.938365 | val 3.325012 | tok/s 2627530.31 | norm 0.3667 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17546 | train 3.055141 | val 3.325012 | tok/s 2623851.29 | norm 0.4743 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17547 | train 2.985477 | val 3.325012 | tok/s 2628889.64 | norm 0.3324 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17548 | train 3.080464 | val 3.325012 | tok/s 2627717.43 | norm 0.3350 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17549 | train 2.875199 | val 3.325012 | tok/s 2629711.40 | norm 0.4144 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17550 | train 2.891254 | val 3.325012 | tok/s 2626940.12 | norm 0.3537 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17551 | train 2.982629 | val 3.325012 | tok/s 2627669.60 | norm 0.3751 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17552 | train 2.817045 | val 3.325012 | tok/s 2622809.62 | norm 0.3932 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17553 | train 3.007999 | val 3.325012 | tok/s 2626097.36 | norm 0.3888 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17554 | train 2.938615 | val 3.325012 | tok/s 2630448.67 | norm 0.3989 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17555 | train 2.849074 | val 3.325012 | tok/s 2628762.21 | norm 0.3830 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17556 | train 2.928562 | val 3.325012 | tok/s 2625628.48 | norm 0.4163 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17557 | train 2.824054 | val 3.325012 | tok/s 2618766.68 | norm 0.4271 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17558 | train 2.939605 | val 3.325012 | tok/s 2624842.47 | norm 0.4229 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17559 | train 2.781200 | val 3.325012 | tok/s 2627338.68 | norm 0.3928 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17560 | train 2.815674 | val 3.325012 | tok/s 2625820.85 | norm 0.3759 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17561 | train 2.938693 | val 3.325012 | tok/s 2628100.75 | norm 0.5558 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17562 | train 2.961490 | val 3.325012 | tok/s 2628635.94 | norm 0.3236 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17563 | train 2.929822 | val 3.325012 | tok/s 2627745.30 | norm 0.3219 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17564 | train 2.906993 | val 3.325012 | tok/s 2629267.80 | norm 0.3486 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17565 | train 2.887906 | val 3.325012 | tok/s 2626769.97 | norm 0.3976 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17566 | train 3.015884 | val 3.325012 | tok/s 2629295.19 | norm 0.3796 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17567 | train 2.888323 | val 3.325012 | tok/s 2628524.99 | norm 0.3785 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17568 | train 2.922825 | val 3.325012 | tok/s 2628626.51 | norm 0.3832 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17569 | train 2.961593 | val 3.325012 | tok/s 2627605.83 | norm 0.3553 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17570 | train 3.067539 | val 3.325012 | tok/s 2627386.74 | norm 0.3837 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17571 | train 3.014548 | val 3.325012 | tok/s 2630605.57 | norm 0.3358 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17572 | train 2.990302 | val 3.325012 | tok/s 2628780.82 | norm 0.3346 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17573 | train 2.908571 | val 3.325012 | tok/s 2628793.76 | norm 0.3816 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17574 | train 3.007245 | val 3.325012 | tok/s 2628466.80 | norm 0.3613 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17575 | train 2.889852 | val 3.325012 | tok/s 2626907.45 | norm 0.3856 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17576 | train 2.872893 | val 3.325012 | tok/s 2631279.04 | norm 0.3952 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17577 | train 2.881864 | val 3.325012 | tok/s 2609725.95 | norm 0.3702 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17578 | train 2.917212 | val 3.325012 | tok/s 2631068.13 | norm 0.3874 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17579 | train 2.886053 | val 3.325012 | tok/s 2632241.52 | norm 0.4019 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17580 | train 2.968164 | val 3.325012 | tok/s 2628890.68 | norm 0.3603 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17581 | train 2.919790 | val 3.325012 | tok/s 2631373.71 | norm 0.3551 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17582 | train 2.951417 | val 3.325012 | tok/s 2627696.69 | norm 0.3613 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17583 | train 2.885855 | val 3.325012 | tok/s 2626594.47 | norm 0.3567 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17584 | train 3.076548 | val 3.325012 | tok/s 2634732.49 | norm 0.3518 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17585 | train 2.962935 | val 3.325012 | tok/s 2624619.96 | norm 0.3343 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17586 | train 2.937274 | val 3.325012 | tok/s 2630532.04 | norm 0.3661 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17587 | train 2.935761 | val 3.325012 | tok/s 2628964.79 | norm 0.3877 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17588 | train 2.969574 | val 3.325012 | tok/s 2628850.20 | norm 0.3802 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17589 | train 2.824088 | val 3.325012 | tok/s 2630054.26 | norm 0.3693 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17590 | train 3.013666 | val 3.325012 | tok/s 2633084.83 | norm 0.3922 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17591 | train 3.145454 | val 3.325012 | tok/s 2631859.90 | norm 0.3943 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17592 | train 3.085042 | val 3.325012 | tok/s 2439446.37 | norm 0.3816 | dt 214.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17593 | train 3.005497 | val 3.325012 | tok/s 2613885.13 | norm 0.3951 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17594 | train 2.800213 | val 3.325012 | tok/s 2630662.91 | norm 0.3775 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17595 | train 3.065188 | val 3.325012 | tok/s 2626986.08 | norm 0.3707 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17596 | train 2.906736 | val 3.325012 | tok/s 2631778.41 | norm 0.3450 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17597 | train 3.011533 | val 3.325012 | tok/s 2628791.18 | norm 0.3388 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17598 | train 3.114984 | val 3.325012 | tok/s 2632342.87 | norm 0.3697 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17599 | train 2.963343 | val 3.325012 | tok/s 2626767.14 | norm 0.4170 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17600 | train 3.010376 | val 3.325012 | tok/s 2625884.07 | norm 0.4022 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17601 | train 2.824319 | val 3.325012 | tok/s 2628956.45 | norm 0.3832 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17602 | train 2.955704 | val 3.325012 | tok/s 2628923.20 | norm 0.4092 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17603 | train 2.954722 | val 3.325012 | tok/s 2629993.12 | norm 0.4040 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17604 | train 2.889471 | val 3.325012 | tok/s 2624567.12 | norm 0.3901 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17605 | train 2.987779 | val 3.325012 | tok/s 2624091.17 | norm 0.3318 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17606 | train 2.899939 | val 3.325012 | tok/s 2345968.76 | norm 0.3487 | dt 223.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17607 | train 2.956358 | val 3.325012 | tok/s 2556759.97 | norm 0.3853 | dt 205.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17608 | train 2.940529 | val 3.325012 | tok/s 2624993.52 | norm 0.4247 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17609 | train 2.906607 | val 3.325012 | tok/s 2630448.36 | norm 0.4094 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17610 | train 2.871857 | val 3.325012 | tok/s 2627847.78 | norm 0.3718 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17611 | train 3.063894 | val 3.325012 | tok/s 2626518.58 | norm 0.3562 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17612 | train 2.997461 | val 3.325012 | tok/s 2627528.90 | norm 0.3463 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17613 | train 2.960762 | val 3.325012 | tok/s 2629440.18 | norm 0.3829 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17614 | train 2.876265 | val 3.325012 | tok/s 2624350.83 | norm 0.3974 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17615 | train 2.952857 | val 3.325012 | tok/s 2629013.37 | norm 0.4131 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17616 | train 2.844628 | val 3.325012 | tok/s 2632195.88 | norm 0.4118 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17617 | train 3.024088 | val 3.325012 | tok/s 2626228.71 | norm 0.3743 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17618 | train 2.955319 | val 3.325012 | tok/s 2629271.76 | norm 0.3782 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17619 | train 3.046227 | val 3.325012 | tok/s 2628333.99 | norm 0.3766 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17620 | train 2.954519 | val 3.325012 | tok/s 2630775.72 | norm 0.3579 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17621 | train 2.893020 | val 3.325012 | tok/s 2627524.58 | norm 0.3623 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17622 | train 2.913735 | val 3.325012 | tok/s 2496711.35 | norm 0.3593 | dt 209.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17623 | train 3.018853 | val 3.325012 | tok/s 2629904.35 | norm 0.3839 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17624 | train 2.955369 | val 3.325012 | tok/s 2628437.06 | norm 0.3610 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17625 | train 2.912848 | val 3.325012 | tok/s 2596909.98 | norm 0.3599 | dt 201.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17626 | train 3.016923 | val 3.325012 | tok/s 2625684.02 | norm 0.5805 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17627 | train 2.839540 | val 3.325012 | tok/s 2625363.45 | norm 0.3688 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17628 | train 2.986969 | val 3.325012 | tok/s 2633741.67 | norm 0.3904 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17629 | train 3.027927 | val 3.325012 | tok/s 2625570.76 | norm 0.4522 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17630 | train 3.096713 | val 3.325012 | tok/s 2627973.75 | norm 0.4104 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17631 | train 3.137319 | val 3.325012 | tok/s 2625529.47 | norm 0.3871 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17632 | train 2.927138 | val 3.325012 | tok/s 2626958.17 | norm 0.3505 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17633 | train 3.040113 | val 3.325012 | tok/s 2630373.63 | norm 0.3617 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17634 | train 3.013980 | val 3.325012 | tok/s 2629128.00 | norm 0.3775 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17635 | train 2.950023 | val 3.325012 | tok/s 2629013.27 | norm 0.3960 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17636 | train 2.963527 | val 3.325012 | tok/s 2627366.31 | norm 0.3686 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17637 | train 2.845234 | val 3.325012 | tok/s 2627464.78 | norm 0.3617 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17638 | train 3.004786 | val 3.325012 | tok/s 2628757.66 | norm 0.3838 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17639 | train 2.871384 | val 3.325012 | tok/s 2624750.91 | norm 0.3932 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17640 | train 2.859688 | val 3.325012 | tok/s 2486156.27 | norm 0.3954 | dt 210.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17641 | train 2.909416 | val 3.325012 | tok/s 2627047.37 | norm 0.3804 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17642 | train 2.997718 | val 3.325012 | tok/s 2623861.78 | norm 0.3872 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17643 | train 2.989156 | val 3.325012 | tok/s 2633372.13 | norm 0.4402 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17644 | train 2.978752 | val 3.325012 | tok/s 2630599.99 | norm 0.3881 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17645 | train 2.802148 | val 3.325012 | tok/s 2631993.20 | norm 0.3599 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17646 | train 2.965690 | val 3.325012 | tok/s 2626206.64 | norm 0.3849 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17647 | train 2.908668 | val 3.325012 | tok/s 2607123.76 | norm 0.3606 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17648 | train 2.875914 | val 3.325012 | tok/s 2630550.39 | norm 0.3367 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17649 | train 2.913927 | val 3.325012 | tok/s 2630122.47 | norm 0.3394 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17650 | train 3.023311 | val 3.325012 | tok/s 2628962.99 | norm 0.3943 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17651 | train 2.894572 | val 3.325012 | tok/s 2629924.54 | norm 0.3716 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17652 | train 2.970254 | val 3.325012 | tok/s 2631151.18 | norm 0.3282 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17653 | train 3.012879 | val 3.325012 | tok/s 2635534.64 | norm 0.3521 | dt 198.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17654 | train 2.943910 | val 3.325012 | tok/s 2625967.51 | norm 0.3765 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17655 | train 2.899929 | val 3.325012 | tok/s 2628626.40 | norm 0.3891 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17656 | train 2.878759 | val 3.325012 | tok/s 2621352.25 | norm 0.3630 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17657 | train 2.917362 | val 3.325012 | tok/s 2629700.22 | norm 0.3810 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17658 | train 2.890642 | val 3.325012 | tok/s 2633308.28 | norm 0.3893 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17659 | train 3.019088 | val 3.325012 | tok/s 2625931.05 | norm 0.3992 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17660 | train 3.076538 | val 3.325012 | tok/s 2631373.49 | norm 0.3741 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17661 | train 2.899547 | val 3.325012 | tok/s 2627741.85 | norm 0.3779 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17662 | train 2.962572 | val 3.325012 | tok/s 2628295.41 | norm 0.4100 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17663 | train 2.934927 | val 3.325012 | tok/s 2632264.86 | norm 0.3939 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17664 | train 2.859246 | val 3.325012 | tok/s 2631987.03 | norm 0.3714 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17665 | train 2.836659 | val 3.325012 | tok/s 2633430.92 | norm 0.3692 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17666 | train 2.941559 | val 3.325012 | tok/s 2628487.17 | norm 0.4041 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17667 | train 2.956088 | val 3.325012 | tok/s 2628829.19 | norm 0.4108 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17668 | train 2.973911 | val 3.325012 | tok/s 2634495.78 | norm 0.4024 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17669 | train 2.873312 | val 3.325012 | tok/s 2627156.11 | norm 0.3867 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17670 | train 2.991357 | val 3.325012 | tok/s 2630756.27 | norm 0.3843 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17671 | train 2.906958 | val 3.325012 | tok/s 2624607.88 | norm 0.3744 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17672 | train 2.854204 | val 3.325012 | tok/s 2628930.89 | norm 0.3436 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17673 | train 3.008335 | val 3.325012 | tok/s 2633169.92 | norm 0.3799 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17674 | train 2.942552 | val 3.325012 | tok/s 2626003.84 | norm 0.4427 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17675 | train 3.019219 | val 3.325012 | tok/s 2630135.36 | norm 0.3577 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17676 | train 2.950281 | val 3.325012 | tok/s 2626183.94 | norm 0.3472 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17677 | train 2.998947 | val 3.325012 | tok/s 2628174.64 | norm 0.3895 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17678 | train 2.971728 | val 3.325012 | tok/s 2630478.19 | norm 0.3892 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17679 | train 3.090204 | val 3.325012 | tok/s 2625674.12 | norm 0.3933 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17680 | train 3.025501 | val 3.325012 | tok/s 2624818.75 | norm 0.3889 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17681 | train 3.207456 | val 3.325012 | tok/s 2622249.69 | norm 0.4167 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17682 | train 3.037583 | val 3.325012 | tok/s 2624057.86 | norm 0.3728 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17683 | train 3.011799 | val 3.325012 | tok/s 2628948.53 | norm 0.3569 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17684 | train 2.953559 | val 3.325012 | tok/s 2624714.86 | norm 0.3760 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17685 | train 3.021589 | val 3.325012 | tok/s 2631839.91 | norm 0.5847 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17686 | train 2.990955 | val 3.325012 | tok/s 2628861.24 | norm 0.3192 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17687 | train 3.042178 | val 3.325012 | tok/s 2627117.62 | norm 0.3376 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17688 | train 3.047185 | val 3.325012 | tok/s 2629983.71 | norm 0.3467 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17689 | train 2.938102 | val 3.325012 | tok/s 2629398.58 | norm 0.3601 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17690 | train 2.941745 | val 3.325012 | tok/s 2626002.02 | norm 0.4002 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17691 | train 2.910042 | val 3.325012 | tok/s 2622450.64 | norm 0.4226 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17692 | train 2.934077 | val 3.325012 | tok/s 2630826.64 | norm 0.3630 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17693 | train 2.970705 | val 3.325012 | tok/s 2624490.26 | norm 0.3521 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17694 | train 2.919927 | val 3.325012 | tok/s 2628417.16 | norm 0.3615 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17695 | train 3.132385 | val 3.325012 | tok/s 2628364.31 | norm 0.4298 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17696 | train 2.990596 | val 3.325012 | tok/s 2627025.85 | norm 0.3768 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17697 | train 3.004079 | val 3.325012 | tok/s 2629626.46 | norm 0.3204 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17698 | train 2.913712 | val 3.325012 | tok/s 2469602.02 | norm 0.3470 | dt 212.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17699 | train 3.003629 | val 3.325012 | tok/s 2605529.46 | norm 0.3845 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17700 | train 2.985541 | val 3.325012 | tok/s 2630117.89 | norm 0.3450 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17701 | train 2.898868 | val 3.325012 | tok/s 2631484.69 | norm 0.3830 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17702 | train 2.864903 | val 3.325012 | tok/s 2628459.96 | norm 0.3702 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17703 | train 2.848628 | val 3.325012 | tok/s 2630215.13 | norm 0.3487 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17704 | train 2.967947 | val 3.325012 | tok/s 2630360.83 | norm 0.3628 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17705 | train 2.907553 | val 3.325012 | tok/s 2634080.74 | norm 0.4003 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17706 | train 2.974857 | val 3.325012 | tok/s 2629057.50 | norm 0.3872 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17707 | train 2.862131 | val 3.325012 | tok/s 2632952.31 | norm 0.3691 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17708 | train 3.115428 | val 3.325012 | tok/s 2627118.69 | norm 0.4144 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17709 | train 3.154765 | val 3.325012 | tok/s 2626950.49 | norm 0.5643 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17710 | train 2.994414 | val 3.325012 | tok/s 2631543.02 | norm 0.3625 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17711 | train 2.917511 | val 3.325012 | tok/s 2628764.62 | norm 0.3577 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17712 | train 2.942042 | val 3.325012 | tok/s 2632358.55 | norm 0.3613 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17713 | train 2.962916 | val 3.325012 | tok/s 2628604.74 | norm 0.3624 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17714 | train 3.050468 | val 3.325012 | tok/s 2629848.96 | norm 0.3629 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17715 | train 2.796550 | val 3.325012 | tok/s 2631739.68 | norm 0.4045 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17716 | train 2.830300 | val 3.325012 | tok/s 2508427.86 | norm 0.3873 | dt 209.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17717 | train 2.984638 | val 3.325012 | tok/s 2627209.67 | norm 0.3661 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17718 | train 2.771108 | val 3.325012 | tok/s 2630207.18 | norm 0.4010 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17719 | train 2.952434 | val 3.325012 | tok/s 2628613.13 | norm 0.4229 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17720 | train 3.016672 | val 3.325012 | tok/s 2633036.25 | norm 0.3790 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17721 | train 3.064718 | val 3.325012 | tok/s 2630100.43 | norm 0.3369 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17722 | train 2.933299 | val 3.325012 | tok/s 2631762.51 | norm 0.3371 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17723 | train 2.950395 | val 3.325012 | tok/s 2627571.24 | norm 0.3304 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17724 | train 3.016327 | val 3.325012 | tok/s 2629692.57 | norm 0.3599 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17725 | train 2.951243 | val 3.325012 | tok/s 2629401.06 | norm 0.3632 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17726 | train 3.017099 | val 3.325012 | tok/s 2621492.36 | norm 0.4095 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17727 | train 2.982482 | val 3.325012 | tok/s 2627538.11 | norm 0.3938 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17728 | train 2.898368 | val 3.325012 | tok/s 2629790.10 | norm 0.3706 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17729 | train 2.929475 | val 3.325012 | tok/s 2629517.92 | norm 0.3839 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17730 | train 3.043483 | val 3.325012 | tok/s 2627434.11 | norm 0.3972 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17731 | train 2.932111 | val 3.325012 | tok/s 2625650.31 | norm 0.3948 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17732 | train 3.076145 | val 3.325012 | tok/s 2631344.47 | norm 0.4077 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17733 | train 2.960743 | val 3.325012 | tok/s 2625198.01 | norm 0.3623 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17734 | train 3.128662 | val 3.325012 | tok/s 2629853.39 | norm 0.4341 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17735 | train 2.885600 | val 3.325012 | tok/s 2627379.77 | norm 0.3902 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17736 | train 2.909441 | val 3.325012 | tok/s 2627014.84 | norm 0.3415 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17737 | train 2.988445 | val 3.325012 | tok/s 2631456.93 | norm 0.3941 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17738 | train 2.933398 | val 3.325012 | tok/s 2629552.62 | norm 0.3665 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17739 | train 2.932259 | val 3.325012 | tok/s 2631538.49 | norm 0.3685 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17740 | train 3.001847 | val 3.325012 | tok/s 2627699.56 | norm 0.3949 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17741 | train 2.911809 | val 3.325012 | tok/s 2624227.01 | norm 0.3697 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17742 | train 2.946615 | val 3.325012 | tok/s 2629496.80 | norm 0.3798 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17743 | train 2.993838 | val 3.325012 | tok/s 2627193.26 | norm 0.3997 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17744 | train 2.945106 | val 3.325012 | tok/s 2477668.34 | norm 0.3812 | dt 211.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17745 | train 2.943496 | val 3.325012 | tok/s 2628069.56 | norm 0.3781 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17746 | train 2.865271 | val 3.325012 | tok/s 2628773.29 | norm 0.3689 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17747 | train 2.967697 | val 3.325012 | tok/s 2627904.01 | norm 0.3328 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17748 | train 2.934785 | val 3.325012 | tok/s 2630462.01 | norm 0.3500 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17749 | train 2.954011 | val 3.325012 | tok/s 2626748.16 | norm 0.3785 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 17750 | val_loss 3.325837 step 17750 | train 2.889166 | val 3.325837 | tok/s 2479633.08 | norm 0.3836 | dt 211.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17751 | train 2.985761 | val 3.325837 | tok/s 2622406.58 | norm 0.3814 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17752 | train 2.936295 | val 3.325837 | tok/s 2627092.27 | norm 0.3796 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17753 | train 3.012227 | val 3.325837 | tok/s 2630268.77 | norm 0.3778 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17754 | train 2.922247 | val 3.325837 | tok/s 2631767.46 | norm 0.3585 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17755 | train 2.928617 | val 3.325837 | tok/s 2630095.84 | norm 0.3613 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17756 | train 2.880876 | val 3.325837 | tok/s 2627715.19 | norm 0.3656 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17757 | train 2.896973 | val 3.325837 | tok/s 2629402.22 | norm 0.3446 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17758 | train 2.917625 | val 3.325837 | tok/s 2550962.51 | norm 0.3393 | dt 205.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17759 | train 2.961788 | val 3.325837 | tok/s 2579992.45 | norm 0.3565 | dt 203.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17760 | train 2.944618 | val 3.325837 | tok/s 2629826.33 | norm 0.3890 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17761 | train 2.980545 | val 3.325837 | tok/s 2629698.99 | norm 0.3622 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17762 | train 2.939751 | val 3.325837 | tok/s 2629017.00 | norm 0.3485 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17763 | train 3.044976 | val 3.325837 | tok/s 2627294.74 | norm 0.3500 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17764 | train 2.998158 | val 3.325837 | tok/s 2620640.08 | norm 0.3361 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17765 | train 2.896331 | val 3.325837 | tok/s 2622110.08 | norm 0.3985 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17766 | train 2.897118 | val 3.325837 | tok/s 2625884.89 | norm 0.3918 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17767 | train 3.087553 | val 3.325837 | tok/s 2614604.18 | norm 0.3890 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17768 | train 2.742521 | val 3.325837 | tok/s 2623501.71 | norm 0.3741 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17769 | train 2.920276 | val 3.325837 | tok/s 2625131.72 | norm 0.3846 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17770 | train 2.842034 | val 3.325837 | tok/s 2625782.27 | norm 0.4196 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17771 | train 2.989585 | val 3.325837 | tok/s 2625772.43 | norm 0.3961 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17772 | train 2.901300 | val 3.325837 | tok/s 2630950.54 | norm 0.3886 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17773 | train 3.164639 | val 3.325837 | tok/s 2632820.64 | norm 0.3909 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17774 | train 2.928719 | val 3.325837 | tok/s 2629292.24 | norm 0.3886 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17775 | train 2.850825 | val 3.325837 | tok/s 2626142.13 | norm 0.4058 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17776 | train 3.016961 | val 3.325837 | tok/s 2632458.05 | norm 0.3705 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17777 | train 2.836276 | val 3.325837 | tok/s 2628986.52 | norm 0.3617 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17778 | train 2.876807 | val 3.325837 | tok/s 2626951.03 | norm 0.3690 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17779 | train 2.984777 | val 3.325837 | tok/s 2623217.10 | norm 0.3797 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17780 | train 3.248960 | val 3.325837 | tok/s 2627153.64 | norm 0.6219 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17781 | train 3.073901 | val 3.325837 | tok/s 2624633.63 | norm 0.3673 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17782 | train 2.914179 | val 3.325837 | tok/s 2629220.13 | norm 0.3527 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17783 | train 3.036268 | val 3.325837 | tok/s 2632144.89 | norm 0.4878 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17784 | train 2.814947 | val 3.325837 | tok/s 2630741.12 | norm 0.3891 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17785 | train 2.992705 | val 3.325837 | tok/s 2630256.15 | norm 0.3657 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17786 | train 3.041061 | val 3.325837 | tok/s 2627630.33 | norm 0.3885 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17787 | train 3.000926 | val 3.325837 | tok/s 2629045.72 | norm 0.3735 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17788 | train 2.908582 | val 3.325837 | tok/s 2625643.98 | norm 0.3428 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17789 | train 2.949084 | val 3.325837 | tok/s 2625794.49 | norm 0.3641 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17790 | train 2.995117 | val 3.325837 | tok/s 2629373.71 | norm 0.4012 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17791 | train 2.844252 | val 3.325837 | tok/s 2627826.92 | norm 0.3700 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17792 | train 2.877077 | val 3.325837 | tok/s 2606835.10 | norm 0.3375 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17793 | train 2.901501 | val 3.325837 | tok/s 2566736.81 | norm 0.3443 | dt 204.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17794 | train 2.834443 | val 3.325837 | tok/s 2625048.76 | norm 0.3541 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17795 | train 2.928572 | val 3.325837 | tok/s 2629427.52 | norm 0.4053 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17796 | train 2.990635 | val 3.325837 | tok/s 2629036.94 | norm 0.3465 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17797 | train 3.065967 | val 3.325837 | tok/s 2628763.87 | norm 0.4080 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17798 | train 3.040513 | val 3.325837 | tok/s 2629262.36 | norm 0.3797 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17799 | train 3.065889 | val 3.325837 | tok/s 2489658.07 | norm 0.3528 | dt 210.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17800 | train 3.092878 | val 3.325837 | tok/s 2555888.76 | norm 0.3782 | dt 205.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17801 | train 2.871137 | val 3.325837 | tok/s 2625324.86 | norm 0.3901 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17802 | train 2.917150 | val 3.325837 | tok/s 2628410.29 | norm 0.4085 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17803 | train 2.935411 | val 3.325837 | tok/s 2626464.10 | norm 0.3782 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17804 | train 2.870010 | val 3.325837 | tok/s 2628741.55 | norm 0.3854 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17805 | train 2.894317 | val 3.325837 | tok/s 2627734.37 | norm 0.3830 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17806 | train 2.986307 | val 3.325837 | tok/s 2630129.58 | norm 0.4080 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17807 | train 2.990187 | val 3.325837 | tok/s 2606512.19 | norm 0.3607 | dt 201.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17808 | train 2.940032 | val 3.325837 | tok/s 2626529.94 | norm 0.3393 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17809 | train 2.936898 | val 3.325837 | tok/s 2627367.43 | norm 0.4073 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17810 | train 2.999151 | val 3.325837 | tok/s 2632011.57 | norm 0.3954 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17811 | train 3.022146 | val 3.325837 | tok/s 2627684.14 | norm 0.3683 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17812 | train 3.037848 | val 3.325837 | tok/s 2628248.52 | norm 0.3814 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17813 | train 3.180963 | val 3.325837 | tok/s 2629434.68 | norm 0.3909 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17814 | train 2.910745 | val 3.325837 | tok/s 2630853.03 | norm 0.3696 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17815 | train 2.880662 | val 3.325837 | tok/s 2631053.75 | norm 0.3716 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17816 | train 3.026380 | val 3.325837 | tok/s 2624225.84 | norm 0.3565 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17817 | train 3.079119 | val 3.325837 | tok/s 2628264.87 | norm 0.4285 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17818 | train 2.857946 | val 3.325837 | tok/s 2632475.96 | norm 0.3757 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17819 | train 2.836833 | val 3.325837 | tok/s 2627131.51 | norm 0.3530 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17820 | train 2.942156 | val 3.325837 | tok/s 2631715.78 | norm 0.4097 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17821 | train 2.965835 | val 3.325837 | tok/s 2629403.73 | norm 0.3281 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17822 | train 3.045529 | val 3.325837 | tok/s 2630439.09 | norm 0.3673 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17823 | train 2.979050 | val 3.325837 | tok/s 2625084.58 | norm 0.3504 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17824 | train 2.995889 | val 3.325837 | tok/s 2630417.07 | norm 0.3671 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17825 | train 2.882592 | val 3.325837 | tok/s 2634218.75 | norm 0.3704 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17826 | train 2.959124 | val 3.325837 | tok/s 2622487.36 | norm 0.3656 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17827 | train 3.005720 | val 3.325837 | tok/s 2633745.66 | norm 0.3856 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17828 | train 2.938478 | val 3.325837 | tok/s 2628900.73 | norm 0.3719 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17829 | train 3.008404 | val 3.325837 | tok/s 2625208.41 | norm 0.3779 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17830 | train 2.957416 | val 3.325837 | tok/s 2627611.72 | norm 0.3638 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17831 | train 3.038340 | val 3.325837 | tok/s 2623452.70 | norm 0.4086 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17832 | train 2.946752 | val 3.325837 | tok/s 2630073.03 | norm 0.3971 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17833 | train 2.810243 | val 3.325837 | tok/s 2628761.55 | norm 0.3482 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17834 | train 3.038498 | val 3.325837 | tok/s 2631945.94 | norm 0.3432 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17835 | train 2.847336 | val 3.325837 | tok/s 2629281.79 | norm 0.3471 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17836 | train 2.946270 | val 3.325837 | tok/s 2628606.69 | norm 0.3611 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17837 | train 2.921635 | val 3.325837 | tok/s 2632106.45 | norm 0.3717 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17838 | train 2.909237 | val 3.325837 | tok/s 2628761.86 | norm 0.4188 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17839 | train 2.960370 | val 3.325837 | tok/s 2574457.17 | norm 0.3842 | dt 203.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17840 | train 2.837421 | val 3.325837 | tok/s 2622208.50 | norm 0.3458 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17841 | train 2.849065 | val 3.325837 | tok/s 2626382.56 | norm 0.3776 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17842 | train 2.986156 | val 3.325837 | tok/s 2629531.09 | norm 0.3947 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17843 | train 3.011721 | val 3.325837 | tok/s 2626200.90 | norm 0.3575 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17844 | train 2.945841 | val 3.325837 | tok/s 2630194.99 | norm 0.4015 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17845 | train 3.047376 | val 3.325837 | tok/s 2617319.46 | norm 0.3913 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17846 | train 2.868802 | val 3.325837 | tok/s 2608352.38 | norm 0.3444 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17847 | train 2.924509 | val 3.325837 | tok/s 2631580.86 | norm 0.3874 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17848 | train 3.038187 | val 3.325837 | tok/s 2625310.15 | norm 0.3997 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17849 | train 2.878128 | val 3.325837 | tok/s 2614327.12 | norm 0.3682 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17850 | train 2.929421 | val 3.325837 | tok/s 2628768.51 | norm 0.3929 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17851 | train 2.924305 | val 3.325837 | tok/s 2629032.65 | norm 0.3935 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17852 | train 2.904722 | val 3.325837 | tok/s 2634703.89 | norm 0.3580 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17853 | train 2.902447 | val 3.325837 | tok/s 2631991.07 | norm 0.3775 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17854 | train 2.975386 | val 3.325837 | tok/s 2628888.50 | norm 0.3710 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17855 | train 2.856833 | val 3.325837 | tok/s 2486323.51 | norm 0.3760 | dt 210.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17856 | train 3.032332 | val 3.325837 | tok/s 2626132.69 | norm 0.4167 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17857 | train 2.953762 | val 3.325837 | tok/s 2630312.87 | norm 0.4046 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17858 | train 2.947933 | val 3.325837 | tok/s 2631570.82 | norm 0.3778 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17859 | train 3.010256 | val 3.325837 | tok/s 2629489.24 | norm 0.3836 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17860 | train 2.926532 | val 3.325837 | tok/s 2630475.17 | norm 0.3737 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17861 | train 3.080241 | val 3.325837 | tok/s 2626042.61 | norm 0.4467 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17862 | train 2.804573 | val 3.325837 | tok/s 2608593.80 | norm 0.3327 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17863 | train 3.050307 | val 3.325837 | tok/s 2603525.12 | norm 0.3100 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17864 | train 3.045660 | val 3.325837 | tok/s 2632318.69 | norm 0.4680 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17865 | train 2.966720 | val 3.325837 | tok/s 2629181.47 | norm 0.3714 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17866 | train 2.993598 | val 3.325837 | tok/s 2631177.82 | norm 0.3256 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17867 | train 2.841601 | val 3.325837 | tok/s 2632158.65 | norm 0.3523 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17868 | train 2.931619 | val 3.325837 | tok/s 2631190.77 | norm 0.3669 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17869 | train 2.893045 | val 3.325837 | tok/s 2628908.44 | norm 0.3646 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17870 | train 2.866262 | val 3.325837 | tok/s 2624890.18 | norm 0.3641 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17871 | train 2.894032 | val 3.325837 | tok/s 2628106.81 | norm 0.3808 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17872 | train 2.944139 | val 3.325837 | tok/s 2633529.19 | norm 0.3734 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17873 | train 2.940471 | val 3.325837 | tok/s 2535607.43 | norm 0.3635 | dt 206.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17874 | train 2.929293 | val 3.325837 | tok/s 2628550.62 | norm 0.3521 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17875 | train 2.985829 | val 3.325837 | tok/s 2631115.98 | norm 0.3363 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17876 | train 2.861260 | val 3.325837 | tok/s 2629945.14 | norm 0.3879 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17877 | train 2.889082 | val 3.325837 | tok/s 2626899.70 | norm 0.4142 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17878 | train 2.883824 | val 3.325837 | tok/s 2626792.05 | norm 0.3822 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17879 | train 2.970830 | val 3.325837 | tok/s 2631216.36 | norm 0.3551 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17880 | train 3.029684 | val 3.325837 | tok/s 2628441.91 | norm 0.3691 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17881 | train 2.955995 | val 3.325837 | tok/s 2629097.40 | norm 0.3703 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17882 | train 2.908679 | val 3.325837 | tok/s 2630472.04 | norm 0.3802 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17883 | train 2.922046 | val 3.325837 | tok/s 2624216.47 | norm 0.3894 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17884 | train 3.002906 | val 3.325837 | tok/s 2630878.42 | norm 0.3667 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17885 | train 3.006957 | val 3.325837 | tok/s 2629551.90 | norm 0.3421 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17886 | train 2.909748 | val 3.325837 | tok/s 2624608.44 | norm 0.3705 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17887 | train 2.843539 | val 3.325837 | tok/s 2629945.59 | norm 0.3805 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17888 | train 2.866674 | val 3.325837 | tok/s 2626255.88 | norm 0.3412 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17889 | train 3.132772 | val 3.325837 | tok/s 2627179.23 | norm 0.3596 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17890 | train 2.974229 | val 3.325837 | tok/s 2622662.17 | norm 0.3440 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17891 | train 3.077751 | val 3.325837 | tok/s 2627928.14 | norm 0.3598 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17892 | train 2.851370 | val 3.325837 | tok/s 2620135.62 | norm 0.3834 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17893 | train 3.005808 | val 3.325837 | tok/s 2495771.69 | norm 0.3840 | dt 210.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17894 | train 2.952173 | val 3.325837 | tok/s 2624892.53 | norm 0.4054 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17895 | train 2.914265 | val 3.325837 | tok/s 2484075.87 | norm 0.3841 | dt 211.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17896 | train 3.002395 | val 3.325837 | tok/s 2503762.39 | norm 0.3874 | dt 209.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17897 | train 3.017844 | val 3.325837 | tok/s 2625568.05 | norm 0.4175 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17898 | train 3.030799 | val 3.325837 | tok/s 2628629.43 | norm 0.3826 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17899 | train 2.796841 | val 3.325837 | tok/s 2627451.70 | norm 0.3467 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17900 | train 3.031846 | val 3.325837 | tok/s 2627730.68 | norm 0.3539 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17901 | train 3.102541 | val 3.325837 | tok/s 2628619.71 | norm 0.3672 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17902 | train 2.901601 | val 3.325837 | tok/s 2623174.09 | norm 0.3837 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17903 | train 2.904216 | val 3.325837 | tok/s 2623111.73 | norm 0.4731 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17904 | train 2.886492 | val 3.325837 | tok/s 2629262.65 | norm 0.3392 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17905 | train 3.016743 | val 3.325837 | tok/s 2621296.01 | norm 0.3223 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17906 | train 2.909587 | val 3.325837 | tok/s 2631052.09 | norm 0.3494 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17907 | train 3.094354 | val 3.325837 | tok/s 2628197.13 | norm 0.5549 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17908 | train 2.937753 | val 3.325837 | tok/s 2630244.19 | norm 0.3458 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17909 | train 2.950553 | val 3.325837 | tok/s 2634782.61 | norm 0.3304 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17910 | train 2.857191 | val 3.325837 | tok/s 2630999.91 | norm 0.3619 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17911 | train 2.988081 | val 3.325837 | tok/s 2629101.86 | norm 0.3892 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17912 | train 3.000734 | val 3.325837 | tok/s 2625266.29 | norm 0.3814 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17913 | train 3.049968 | val 3.325837 | tok/s 2630514.05 | norm 0.3347 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17914 | train 2.977629 | val 3.325837 | tok/s 2632943.14 | norm 0.3524 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17915 | train 2.968003 | val 3.325837 | tok/s 2628033.70 | norm 0.3645 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17916 | train 2.979649 | val 3.325837 | tok/s 2630029.87 | norm 0.3505 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17917 | train 2.917205 | val 3.325837 | tok/s 2628602.11 | norm 0.3649 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17918 | train 3.009405 | val 3.325837 | tok/s 2628549.10 | norm 0.3653 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17919 | train 2.985059 | val 3.325837 | tok/s 2632806.22 | norm 0.3670 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17920 | train 3.064740 | val 3.325837 | tok/s 2622431.52 | norm 0.3710 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17921 | train 2.955695 | val 3.325837 | tok/s 2630731.44 | norm 0.3792 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17922 | train 2.974446 | val 3.325837 | tok/s 2632196.74 | norm 0.3476 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17923 | train 2.981837 | val 3.325837 | tok/s 2628147.65 | norm 0.3584 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17924 | train 2.938524 | val 3.325837 | tok/s 2629376.92 | norm 0.3951 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17925 | train 2.943056 | val 3.325837 | tok/s 2623939.71 | norm 0.3681 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17926 | train 3.042630 | val 3.325837 | tok/s 2632734.72 | norm 0.5862 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17927 | train 3.009475 | val 3.325837 | tok/s 2628218.03 | norm 0.3315 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17928 | train 2.932613 | val 3.325837 | tok/s 2631163.79 | norm 0.3150 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17929 | train 3.023163 | val 3.325837 | tok/s 2626447.59 | norm 0.3668 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17930 | train 2.907006 | val 3.325837 | tok/s 2627036.94 | norm 0.3492 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17931 | train 3.081012 | val 3.325837 | tok/s 2631639.62 | norm 0.3596 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17932 | train 2.992976 | val 3.325837 | tok/s 2627868.30 | norm 0.3705 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17933 | train 2.899452 | val 3.325837 | tok/s 2629823.50 | norm 0.3651 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17934 | train 2.923841 | val 3.325837 | tok/s 2627423.16 | norm 0.3745 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17935 | train 3.085341 | val 3.325837 | tok/s 2623580.32 | norm 0.4025 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17936 | train 3.042374 | val 3.325837 | tok/s 2625673.71 | norm 0.3845 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17937 | train 2.824454 | val 3.325837 | tok/s 2588772.37 | norm 0.3510 | dt 202.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17938 | train 2.949288 | val 3.325837 | tok/s 2630117.82 | norm 0.3941 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17939 | train 2.939270 | val 3.325837 | tok/s 2625316.81 | norm 0.4104 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17940 | train 2.961938 | val 3.325837 | tok/s 2629449.20 | norm 0.3611 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17941 | train 2.966681 | val 3.325837 | tok/s 2633469.83 | norm 0.3717 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17942 | train 2.923942 | val 3.325837 | tok/s 2627629.21 | norm 0.3943 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17943 | train 2.947670 | val 3.325837 | tok/s 2628960.81 | norm 0.3892 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17944 | train 3.020470 | val 3.325837 | tok/s 2620950.46 | norm 0.3554 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17945 | train 2.924042 | val 3.325837 | tok/s 2626128.72 | norm 0.3552 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17946 | train 2.939335 | val 3.325837 | tok/s 2624613.29 | norm 0.3819 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17947 | train 3.053317 | val 3.325837 | tok/s 2626415.92 | norm 0.5509 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17948 | train 2.870038 | val 3.325837 | tok/s 2627562.52 | norm 0.3763 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17949 | train 2.963943 | val 3.325837 | tok/s 2625750.70 | norm 0.3181 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17950 | train 2.965235 | val 3.325837 | tok/s 2626238.94 | norm 0.3340 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17951 | train 2.770731 | val 3.325837 | tok/s 2630725.84 | norm 0.3498 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17952 | train 3.018251 | val 3.325837 | tok/s 2630637.66 | norm 0.3807 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17953 | train 2.971146 | val 3.325837 | tok/s 2479239.42 | norm 0.3670 | dt 211.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17954 | train 2.996575 | val 3.325837 | tok/s 2571842.22 | norm 0.3398 | dt 203.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17955 | train 3.005037 | val 3.325837 | tok/s 2625856.55 | norm 0.3593 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17956 | train 2.931431 | val 3.325837 | tok/s 2628200.85 | norm 0.3764 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17957 | train 2.958926 | val 3.325837 | tok/s 2634098.42 | norm 0.3637 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17958 | train 2.929027 | val 3.325837 | tok/s 2632590.60 | norm 0.3590 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17959 | train 3.071602 | val 3.325837 | tok/s 2627775.45 | norm 0.3660 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17960 | train 2.843299 | val 3.325837 | tok/s 2621309.54 | norm 0.3594 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17961 | train 2.991040 | val 3.325837 | tok/s 2629227.10 | norm 0.3492 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17962 | train 3.058865 | val 3.325837 | tok/s 2624668.46 | norm 0.3712 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17963 | train 2.969880 | val 3.325837 | tok/s 2629373.08 | norm 0.3417 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17964 | train 2.920492 | val 3.325837 | tok/s 2631570.88 | norm 0.3290 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17965 | train 3.003003 | val 3.325837 | tok/s 2626805.28 | norm 0.3455 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17966 | train 2.915683 | val 3.325837 | tok/s 2627941.50 | norm 0.3612 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17967 | train 3.013485 | val 3.325837 | tok/s 2521908.04 | norm 0.3510 | dt 207.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17968 | train 2.798375 | val 3.325837 | tok/s 2629773.75 | norm 0.3570 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17969 | train 2.915514 | val 3.325837 | tok/s 2623489.76 | norm 0.3962 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17970 | train 2.949218 | val 3.325837 | tok/s 2630086.83 | norm 0.3895 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17971 | train 2.947311 | val 3.325837 | tok/s 2630036.14 | norm 0.3500 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17972 | train 2.930836 | val 3.325837 | tok/s 2624433.52 | norm 0.3819 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17973 | train 2.966212 | val 3.325837 | tok/s 2632304.85 | norm 0.3738 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17974 | train 2.988233 | val 3.325837 | tok/s 2625929.55 | norm 0.4052 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17975 | train 2.932373 | val 3.325837 | tok/s 2626765.93 | norm 0.3977 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17976 | train 2.958363 | val 3.325837 | tok/s 2628557.56 | norm 0.3807 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17977 | train 2.958022 | val 3.325837 | tok/s 2627913.12 | norm 0.3822 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17978 | train 2.957539 | val 3.325837 | tok/s 2629805.45 | norm 0.3923 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17979 | train 2.882447 | val 3.325837 | tok/s 2628651.22 | norm 0.3805 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17980 | train 2.964881 | val 3.325837 | tok/s 2630149.65 | norm 0.3764 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17981 | train 2.955290 | val 3.325837 | tok/s 2623450.65 | norm 0.3621 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17982 | train 2.965852 | val 3.325837 | tok/s 2626339.98 | norm 0.3619 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17983 | train 2.909673 | val 3.325837 | tok/s 2628261.68 | norm 0.3945 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17984 | train 2.783195 | val 3.325837 | tok/s 2627670.97 | norm 0.3806 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17985 | train 2.951742 | val 3.325837 | tok/s 2625451.62 | norm 0.3844 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17986 | train 3.075529 | val 3.325837 | tok/s 2534113.72 | norm 0.3781 | dt 206.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17987 | train 2.924532 | val 3.325837 | tok/s 2627175.75 | norm 0.3825 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17988 | train 3.041270 | val 3.325837 | tok/s 2630626.83 | norm 0.3761 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17989 | train 3.027606 | val 3.325837 | tok/s 2627977.70 | norm 0.3603 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17990 | train 3.048157 | val 3.325837 | tok/s 2629564.91 | norm 0.3897 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17991 | train 3.061896 | val 3.325837 | tok/s 2624158.58 | norm 0.3961 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17992 | train 2.982184 | val 3.325837 | tok/s 2629430.35 | norm 0.4141 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17993 | train 2.998137 | val 3.325837 | tok/s 2632194.08 | norm 0.3973 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17994 | train 2.981818 | val 3.325837 | tok/s 2626700.27 | norm 0.3771 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17995 | train 2.880349 | val 3.325837 | tok/s 2632599.09 | norm 0.3598 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17996 | train 2.808726 | val 3.325837 | tok/s 2629355.42 | norm 0.3349 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17997 | train 2.852060 | val 3.325837 | tok/s 2630977.50 | norm 0.3422 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17998 | train 2.884155 | val 3.325837 | tok/s 2630737.74 | norm 0.3882 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 17999 | train 2.994550 | val 3.325837 | tok/s 2494699.23 | norm 0.3712 | dt 210.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 18000 | val_loss 3.317983 step 18000 | train 2.946139 | val 3.317983 | tok/s 2615434.80 | norm 0.3594 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18001 | train 2.908082 | val 3.317983 | tok/s 2629538.00 | norm 0.3653 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18002 | train 3.132527 | val 3.317983 | tok/s 2626455.03 | norm 0.4085 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18003 | train 2.934688 | val 3.317983 | tok/s 2628447.35 | norm 0.3984 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18004 | train 2.957139 | val 3.317983 | tok/s 2631096.37 | norm 0.3561 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18005 | train 2.999528 | val 3.317983 | tok/s 2626957.89 | norm 0.3814 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18006 | train 2.914409 | val 3.317983 | tok/s 2631340.56 | norm 0.3758 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18007 | train 2.937219 | val 3.317983 | tok/s 2624582.40 | norm 0.3839 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18008 | train 2.880693 | val 3.317983 | tok/s 2629934.99 | norm 0.4099 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18009 | train 3.008833 | val 3.317983 | tok/s 2626847.01 | norm 0.4080 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18010 | train 2.890446 | val 3.317983 | tok/s 2630278.34 | norm 0.3879 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18011 | train 2.803432 | val 3.317983 | tok/s 2626372.62 | norm 0.4149 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18012 | train 2.981358 | val 3.317983 | tok/s 2631101.26 | norm 0.3715 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18013 | train 3.117042 | val 3.317983 | tok/s 2625061.56 | norm 0.3791 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18014 | train 2.886552 | val 3.317983 | tok/s 2600576.65 | norm 0.3962 | dt 201.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18015 | train 3.042258 | val 3.317983 | tok/s 2609969.82 | norm 0.4159 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18016 | train 2.960234 | val 3.317983 | tok/s 2633762.82 | norm 0.3670 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18017 | train 2.982693 | val 3.317983 | tok/s 2631232.79 | norm 0.3538 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18018 | train 3.030983 | val 3.317983 | tok/s 2628670.36 | norm 0.3847 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18019 | train 3.079824 | val 3.317983 | tok/s 2628369.22 | norm 0.3789 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18020 | train 2.891504 | val 3.317983 | tok/s 2571793.33 | norm 0.3392 | dt 203.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18021 | train 2.959293 | val 3.317983 | tok/s 2628442.90 | norm 0.3584 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18022 | train 2.892985 | val 3.317983 | tok/s 2627325.20 | norm 0.3547 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18023 | train 2.827714 | val 3.317983 | tok/s 2626979.55 | norm 0.3469 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18024 | train 2.973579 | val 3.317983 | tok/s 2624507.66 | norm 0.4062 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18025 | train 2.965183 | val 3.317983 | tok/s 2625663.23 | norm 0.3867 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18026 | train 2.937227 | val 3.317983 | tok/s 2631467.71 | norm 0.3957 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18027 | train 2.924300 | val 3.317983 | tok/s 2629005.38 | norm 0.4151 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18028 | train 3.004416 | val 3.317983 | tok/s 2631468.79 | norm 0.3779 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18029 | train 2.951386 | val 3.317983 | tok/s 2623264.30 | norm 0.3375 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18030 | train 3.009032 | val 3.317983 | tok/s 2624335.32 | norm 0.3551 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18031 | train 2.949973 | val 3.317983 | tok/s 2629885.09 | norm 0.3677 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18032 | train 3.006495 | val 3.317983 | tok/s 2626328.56 | norm 0.3769 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18033 | train 2.940654 | val 3.317983 | tok/s 2628928.57 | norm 0.3725 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18034 | train 3.011880 | val 3.317983 | tok/s 2624424.15 | norm 0.3530 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18035 | train 2.892432 | val 3.317983 | tok/s 2623511.10 | norm 0.3474 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18036 | train 2.904925 | val 3.317983 | tok/s 2629246.35 | norm 0.3942 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18037 | train 2.944389 | val 3.317983 | tok/s 2593224.32 | norm 0.3818 | dt 202.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18038 | train 3.010953 | val 3.317983 | tok/s 2628357.36 | norm 0.3263 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18039 | train 2.995553 | val 3.317983 | tok/s 2622645.87 | norm 0.3358 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18040 | train 2.975810 | val 3.317983 | tok/s 2627308.54 | norm 0.3612 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18041 | train 2.977970 | val 3.317983 | tok/s 2625849.48 | norm 0.3764 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18042 | train 3.065972 | val 3.317983 | tok/s 2628236.63 | norm 0.3747 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18043 | train 3.028270 | val 3.317983 | tok/s 2627770.42 | norm 0.3844 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18044 | train 3.018617 | val 3.317983 | tok/s 2627598.49 | norm 0.4104 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18045 | train 2.963370 | val 3.317983 | tok/s 2627780.43 | norm 0.3738 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18046 | train 2.907033 | val 3.317983 | tok/s 2630183.82 | norm 0.3727 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18047 | train 2.984578 | val 3.317983 | tok/s 2510544.30 | norm 0.3662 | dt 208.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18048 | train 3.110281 | val 3.317983 | tok/s 2620388.23 | norm 0.4042 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18049 | train 2.873993 | val 3.317983 | tok/s 2625202.03 | norm 0.3746 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18050 | train 2.929318 | val 3.317983 | tok/s 2625853.84 | norm 0.3848 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18051 | train 2.870725 | val 3.317983 | tok/s 2631685.10 | norm 0.3692 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18052 | train 2.951488 | val 3.317983 | tok/s 2627467.17 | norm 0.3564 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18053 | train 2.993045 | val 3.317983 | tok/s 2629339.94 | norm 0.4014 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18054 | train 2.973018 | val 3.317983 | tok/s 2628374.88 | norm 0.4129 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18055 | train 3.023035 | val 3.317983 | tok/s 2631657.79 | norm 0.3483 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18056 | train 2.870689 | val 3.317983 | tok/s 2630391.64 | norm 0.3623 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18057 | train 2.871413 | val 3.317983 | tok/s 2632346.75 | norm 0.3768 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18058 | train 2.934926 | val 3.317983 | tok/s 2630589.19 | norm 0.3766 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18059 | train 2.994678 | val 3.317983 | tok/s 2623367.16 | norm 0.3818 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18060 | train 2.960279 | val 3.317983 | tok/s 2630555.60 | norm 0.3337 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18061 | train 2.891627 | val 3.317983 | tok/s 2579431.01 | norm 0.3411 | dt 203.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18062 | train 2.872310 | val 3.317983 | tok/s 2609720.46 | norm 0.3414 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18063 | train 2.904207 | val 3.317983 | tok/s 2628656.66 | norm 0.3482 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18064 | train 2.942370 | val 3.317983 | tok/s 2628016.07 | norm 0.3633 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18065 | train 2.945710 | val 3.317983 | tok/s 2633142.11 | norm 0.3611 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18066 | train 2.951374 | val 3.317983 | tok/s 2629904.54 | norm 0.3812 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18067 | train 2.900054 | val 3.317983 | tok/s 2624251.68 | norm 0.4011 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18068 | train 2.896595 | val 3.317983 | tok/s 2624020.02 | norm 0.3805 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18069 | train 2.916372 | val 3.317983 | tok/s 2623152.05 | norm 0.3788 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18070 | train 2.911947 | val 3.317983 | tok/s 2629398.36 | norm 0.3932 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18071 | train 2.874990 | val 3.317983 | tok/s 2625542.50 | norm 0.3686 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18072 | train 2.827279 | val 3.317983 | tok/s 2626061.11 | norm 0.3598 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18073 | train 2.868130 | val 3.317983 | tok/s 2628516.86 | norm 0.3787 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18074 | train 2.955149 | val 3.317983 | tok/s 2625454.28 | norm 0.4027 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18075 | train 2.868679 | val 3.317983 | tok/s 2628151.76 | norm 0.3897 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18076 | train 2.979917 | val 3.317983 | tok/s 2631272.30 | norm 0.3726 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18077 | train 3.021646 | val 3.317983 | tok/s 2624400.40 | norm 0.3797 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18078 | train 2.990278 | val 3.317983 | tok/s 2624211.57 | norm 0.3857 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18079 | train 2.906373 | val 3.317983 | tok/s 2625334.88 | norm 0.3765 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18080 | train 2.962388 | val 3.317983 | tok/s 2629378.46 | norm 0.3934 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18081 | train 3.002461 | val 3.317983 | tok/s 2627639.36 | norm 0.3666 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18082 | train 3.032430 | val 3.317983 | tok/s 2624946.38 | norm 0.3857 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18083 | train 2.931590 | val 3.317983 | tok/s 2629434.80 | norm 0.4020 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18084 | train 2.947949 | val 3.317983 | tok/s 2624553.81 | norm 0.4081 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18085 | train 2.934559 | val 3.317983 | tok/s 2628921.28 | norm 0.3717 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18086 | train 2.939052 | val 3.317983 | tok/s 2629552.31 | norm 0.3481 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18087 | train 2.912284 | val 3.317983 | tok/s 2629602.44 | norm 0.3668 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18088 | train 2.967958 | val 3.317983 | tok/s 2628084.39 | norm 0.3841 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18089 | train 2.899659 | val 3.317983 | tok/s 2629018.55 | norm 0.3728 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18090 | train 2.934470 | val 3.317983 | tok/s 2627496.95 | norm 0.3679 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18091 | train 2.934341 | val 3.317983 | tok/s 2625346.12 | norm 0.3692 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18092 | train 2.990964 | val 3.317983 | tok/s 2625682.37 | norm 0.3656 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18093 | train 2.908263 | val 3.317983 | tok/s 2631381.66 | norm 0.4007 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18094 | train 2.891992 | val 3.317983 | tok/s 2627199.77 | norm 0.3925 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18095 | train 2.950631 | val 3.317983 | tok/s 2628004.26 | norm 0.4019 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18096 | train 2.957807 | val 3.317983 | tok/s 2627808.05 | norm 0.3743 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18097 | train 3.100598 | val 3.317983 | tok/s 2624547.03 | norm 0.3789 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18098 | train 2.925799 | val 3.317983 | tok/s 2630887.99 | norm 0.3564 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18099 | train 3.036653 | val 3.317983 | tok/s 2625488.79 | norm 0.3807 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18100 | train 2.948086 | val 3.317983 | tok/s 2492252.11 | norm 0.3922 | dt 210.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18101 | train 2.861397 | val 3.317983 | tok/s 2501429.76 | norm 0.3536 | dt 209.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18102 | train 2.900380 | val 3.317983 | tok/s 2629955.66 | norm 0.3549 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18103 | train 2.851480 | val 3.317983 | tok/s 2623802.26 | norm 0.3785 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18104 | train 2.967971 | val 3.317983 | tok/s 2630719.09 | norm 0.3543 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18105 | train 2.986892 | val 3.317983 | tok/s 2629879.08 | norm 0.3424 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18106 | train 2.923064 | val 3.317983 | tok/s 2626349.58 | norm 0.3598 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18107 | train 2.968821 | val 3.317983 | tok/s 2627352.99 | norm 0.3616 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18108 | train 3.011222 | val 3.317983 | tok/s 2628710.87 | norm 0.3585 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18109 | train 2.804747 | val 3.317983 | tok/s 2631091.86 | norm 0.3601 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18110 | train 2.945287 | val 3.317983 | tok/s 2630480.37 | norm 0.3673 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18111 | train 2.946609 | val 3.317983 | tok/s 2622929.39 | norm 0.3781 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18112 | train 2.864731 | val 3.317983 | tok/s 2628671.58 | norm 0.4018 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18113 | train 2.962860 | val 3.317983 | tok/s 2630560.03 | norm 0.3257 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18114 | train 2.876614 | val 3.317983 | tok/s 2627045.06 | norm 0.3406 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18115 | train 2.822952 | val 3.317983 | tok/s 2629646.97 | norm 0.3365 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18116 | train 3.007295 | val 3.317983 | tok/s 2624525.23 | norm 0.3619 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18117 | train 3.082392 | val 3.317983 | tok/s 2630432.78 | norm 0.3688 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18118 | train 2.986470 | val 3.317983 | tok/s 2630216.56 | norm 0.4113 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18119 | train 2.894901 | val 3.317983 | tok/s 2628248.33 | norm 0.3528 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18120 | train 3.077408 | val 3.317983 | tok/s 2633045.65 | norm 0.3355 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18121 | train 2.806330 | val 3.317983 | tok/s 2615928.25 | norm 0.3538 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18122 | train 3.035474 | val 3.317983 | tok/s 2627464.20 | norm 0.3750 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18123 | train 3.040003 | val 3.317983 | tok/s 2631563.68 | norm 0.4244 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18124 | train 2.919556 | val 3.317983 | tok/s 2621357.19 | norm 0.3409 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18125 | train 2.954383 | val 3.317983 | tok/s 2631225.88 | norm 0.3497 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18126 | train 3.075835 | val 3.317983 | tok/s 2628290.35 | norm 0.4446 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18127 | train 2.880794 | val 3.317983 | tok/s 2629992.05 | norm 0.3744 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18128 | train 2.996177 | val 3.317983 | tok/s 2628661.95 | norm 0.3312 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18129 | train 2.920205 | val 3.317983 | tok/s 2626708.42 | norm 0.3814 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18130 | train 2.938823 | val 3.317983 | tok/s 2633560.04 | norm 0.3894 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18131 | train 2.903844 | val 3.317983 | tok/s 2629556.15 | norm 0.3952 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18132 | train 2.998887 | val 3.317983 | tok/s 2627651.24 | norm 0.3998 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18133 | train 3.011752 | val 3.317983 | tok/s 2620231.46 | norm 0.3567 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18134 | train 2.943986 | val 3.317983 | tok/s 2622894.74 | norm 0.3819 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18135 | train 2.951648 | val 3.317983 | tok/s 2630801.51 | norm 0.3619 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18136 | train 3.016475 | val 3.317983 | tok/s 2633283.38 | norm 0.3377 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18137 | train 3.036055 | val 3.317983 | tok/s 2607013.54 | norm 0.3529 | dt 201.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18138 | train 2.916501 | val 3.317983 | tok/s 2624289.93 | norm 0.4083 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18139 | train 2.959614 | val 3.317983 | tok/s 2627331.21 | norm 0.4282 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18140 | train 2.864253 | val 3.317983 | tok/s 2533624.23 | norm 0.3830 | dt 206.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18141 | train 2.952642 | val 3.317983 | tok/s 2632808.55 | norm 0.4180 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18142 | train 2.864669 | val 3.317983 | tok/s 2631518.73 | norm 0.3918 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18143 | train 2.916404 | val 3.317983 | tok/s 2632849.49 | norm 0.3863 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18144 | train 2.828438 | val 3.317983 | tok/s 2626491.24 | norm 0.4069 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18145 | train 2.962390 | val 3.317983 | tok/s 2632164.58 | norm 0.3566 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18146 | train 2.860924 | val 3.317983 | tok/s 2627872.95 | norm 0.3648 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18147 | train 2.903539 | val 3.317983 | tok/s 2630548.96 | norm 0.3710 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18148 | train 2.968276 | val 3.317983 | tok/s 2489923.99 | norm 0.3629 | dt 210.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18149 | train 2.835023 | val 3.317983 | tok/s 2627528.21 | norm 0.3777 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18150 | train 3.036881 | val 3.317983 | tok/s 2628987.43 | norm 0.3659 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18151 | train 2.951838 | val 3.317983 | tok/s 2629795.49 | norm 0.3641 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18152 | train 2.901955 | val 3.317983 | tok/s 2629016.67 | norm 0.3796 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18153 | train 2.926496 | val 3.317983 | tok/s 2631305.92 | norm 0.3787 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18154 | train 2.983758 | val 3.317983 | tok/s 2629241.89 | norm 0.3837 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18155 | train 2.855097 | val 3.317983 | tok/s 2633222.45 | norm 0.3881 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18156 | train 3.020771 | val 3.317983 | tok/s 2625313.19 | norm 0.4130 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18157 | train 2.810182 | val 3.317983 | tok/s 2629219.35 | norm 0.4204 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18158 | train 2.770325 | val 3.317983 | tok/s 2628162.04 | norm 0.4077 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18159 | train 3.156930 | val 3.317983 | tok/s 2626516.60 | norm 0.4150 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18160 | train 2.886484 | val 3.317983 | tok/s 2632873.40 | norm 0.3900 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18161 | train 2.906808 | val 3.317983 | tok/s 2626118.26 | norm 0.3566 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18162 | train 2.901509 | val 3.317983 | tok/s 2633259.50 | norm 0.3881 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18163 | train 2.816542 | val 3.317983 | tok/s 2628340.56 | norm 0.4112 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18164 | train 2.872080 | val 3.317983 | tok/s 2628808.06 | norm 0.3519 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18165 | train 2.941212 | val 3.317983 | tok/s 2631880.34 | norm 0.3203 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18166 | train 2.865739 | val 3.317983 | tok/s 2628584.87 | norm 0.3768 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18167 | train 2.848458 | val 3.317983 | tok/s 2631262.50 | norm 0.3710 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18168 | train 3.041136 | val 3.317983 | tok/s 2633238.63 | norm 0.3554 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18169 | train 3.030110 | val 3.317983 | tok/s 2628839.70 | norm 0.3287 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18170 | train 3.028667 | val 3.317983 | tok/s 2626333.04 | norm 0.3400 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18171 | train 2.897691 | val 3.317983 | tok/s 2622471.98 | norm 0.3598 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18172 | train 2.960199 | val 3.317983 | tok/s 2631231.11 | norm 0.4201 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18173 | train 2.955292 | val 3.317983 | tok/s 2631058.03 | norm 0.3802 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18174 | train 3.185674 | val 3.317983 | tok/s 2629354.84 | norm 0.3880 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18175 | train 3.068283 | val 3.317983 | tok/s 2627067.78 | norm 0.4060 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18176 | train 2.997004 | val 3.317983 | tok/s 2628976.54 | norm 0.4344 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18177 | train 2.876018 | val 3.317983 | tok/s 2628125.70 | norm 0.3714 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18178 | train 2.825266 | val 3.317983 | tok/s 2630618.52 | norm 0.3825 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18179 | train 3.092401 | val 3.317983 | tok/s 2628726.59 | norm 0.4452 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18180 | train 2.889376 | val 3.317983 | tok/s 2624741.93 | norm 0.4214 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18181 | train 2.828828 | val 3.317983 | tok/s 2625598.45 | norm 0.3994 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18182 | train 2.945639 | val 3.317983 | tok/s 2626123.34 | norm 0.3535 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18183 | train 2.921496 | val 3.317983 | tok/s 2624635.23 | norm 0.3764 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18184 | train 2.949465 | val 3.317983 | tok/s 2628749.40 | norm 0.3838 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18185 | train 3.008711 | val 3.317983 | tok/s 2620581.80 | norm 0.3464 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18186 | train 2.867483 | val 3.317983 | tok/s 2593753.37 | norm 0.3340 | dt 202.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18187 | train 2.849867 | val 3.317983 | tok/s 2631106.41 | norm 0.3491 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18188 | train 3.050392 | val 3.317983 | tok/s 2631807.28 | norm 0.3572 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18189 | train 3.059907 | val 3.317983 | tok/s 2629108.81 | norm 0.3581 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18190 | train 3.009804 | val 3.317983 | tok/s 2627406.75 | norm 0.3811 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18191 | train 2.974891 | val 3.317983 | tok/s 2626546.65 | norm 0.4198 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18192 | train 2.898947 | val 3.317983 | tok/s 2547986.98 | norm 0.3940 | dt 205.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18193 | train 2.881951 | val 3.317983 | tok/s 2529285.76 | norm 0.3911 | dt 207.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18194 | train 3.073306 | val 3.317983 | tok/s 2611802.29 | norm 0.3448 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18195 | train 2.911663 | val 3.317983 | tok/s 2626316.20 | norm 0.3735 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18196 | train 3.003266 | val 3.317983 | tok/s 2579654.11 | norm 0.3848 | dt 203.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18197 | train 3.033556 | val 3.317983 | tok/s 2629928.92 | norm 0.3918 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18198 | train 2.933002 | val 3.317983 | tok/s 2633674.70 | norm 0.3851 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18199 | train 2.943750 | val 3.317983 | tok/s 2627032.47 | norm 0.3714 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18200 | train 2.864472 | val 3.317983 | tok/s 2625370.23 | norm 0.4070 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18201 | train 2.940872 | val 3.317983 | tok/s 2629744.63 | norm 0.3891 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18202 | train 2.842859 | val 3.317983 | tok/s 2630087.66 | norm 0.3906 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18203 | train 3.075199 | val 3.317983 | tok/s 2521575.91 | norm 0.3647 | dt 207.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18204 | train 2.821417 | val 3.317983 | tok/s 2629095.29 | norm 0.3696 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18205 | train 2.984046 | val 3.317983 | tok/s 2627512.96 | norm 0.3824 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18206 | train 2.967981 | val 3.317983 | tok/s 2628787.83 | norm 0.3886 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18207 | train 2.959466 | val 3.317983 | tok/s 2632133.40 | norm 0.3560 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18208 | train 2.853125 | val 3.317983 | tok/s 2628767.03 | norm 0.3519 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18209 | train 2.879041 | val 3.317983 | tok/s 2632108.62 | norm 0.3670 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18210 | train 2.983319 | val 3.317983 | tok/s 2623431.78 | norm 0.3596 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18211 | train 2.981127 | val 3.317983 | tok/s 2628188.73 | norm 0.3749 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18212 | train 2.943656 | val 3.317983 | tok/s 2631243.63 | norm 0.3776 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18213 | train 3.123983 | val 3.317983 | tok/s 2628452.88 | norm 0.3911 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18214 | train 3.000973 | val 3.317983 | tok/s 2628330.39 | norm 0.3710 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18215 | train 2.925374 | val 3.317983 | tok/s 2626196.80 | norm 0.3646 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18216 | train 2.952117 | val 3.317983 | tok/s 2630719.69 | norm 0.3842 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18217 | train 3.069664 | val 3.317983 | tok/s 2622441.92 | norm 0.4094 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18218 | train 2.981853 | val 3.317983 | tok/s 2624701.93 | norm 0.3873 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18219 | train 2.970127 | val 3.317983 | tok/s 2630170.85 | norm 0.3776 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18220 | train 2.922217 | val 3.317983 | tok/s 2625158.19 | norm 0.3874 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18221 | train 2.876498 | val 3.317983 | tok/s 2631207.34 | norm 0.3861 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18222 | train 2.952585 | val 3.317983 | tok/s 2625702.07 | norm 0.3962 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18223 | train 3.013339 | val 3.317983 | tok/s 2626518.90 | norm 0.3994 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18224 | train 3.048737 | val 3.317983 | tok/s 2632474.31 | norm 0.3846 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18225 | train 3.060511 | val 3.317983 | tok/s 2629168.64 | norm 0.3865 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18226 | train 2.883253 | val 3.317983 | tok/s 2630119.25 | norm 0.3923 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18227 | train 2.911308 | val 3.317983 | tok/s 2630411.82 | norm 0.4113 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18228 | train 2.945213 | val 3.317983 | tok/s 2626712.55 | norm 0.3699 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18229 | train 3.028155 | val 3.317983 | tok/s 2629301.80 | norm 0.3374 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18230 | train 3.055043 | val 3.317983 | tok/s 2629701.27 | norm 0.3931 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18231 | train 3.003219 | val 3.317983 | tok/s 2626225.19 | norm 0.4249 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18232 | train 3.048674 | val 3.317983 | tok/s 2632525.09 | norm 0.3886 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18233 | train 2.919459 | val 3.317983 | tok/s 2628709.00 | norm 0.3542 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18234 | train 2.946497 | val 3.317983 | tok/s 2625616.11 | norm 0.3728 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18235 | train 3.004242 | val 3.317983 | tok/s 2631283.51 | norm 0.3835 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18236 | train 3.018772 | val 3.317983 | tok/s 2629740.70 | norm 0.3586 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18237 | train 2.938340 | val 3.317983 | tok/s 2621624.38 | norm 0.4125 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18238 | train 2.977983 | val 3.317983 | tok/s 2625393.42 | norm 0.3915 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18239 | train 2.894042 | val 3.317983 | tok/s 2628171.14 | norm 0.3439 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18240 | train 2.917301 | val 3.317983 | tok/s 2626430.13 | norm 0.3620 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18241 | train 2.948460 | val 3.317983 | tok/s 2625981.37 | norm 0.3627 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18242 | train 2.926657 | val 3.317983 | tok/s 2627637.36 | norm 0.3670 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18243 | train 2.921670 | val 3.317983 | tok/s 2625942.72 | norm 0.3937 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18244 | train 3.043009 | val 3.317983 | tok/s 2630754.25 | norm 0.4103 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18245 | train 2.955642 | val 3.317983 | tok/s 2630707.40 | norm 0.3888 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18246 | train 2.878081 | val 3.317983 | tok/s 2627912.94 | norm 0.3462 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18247 | train 2.902377 | val 3.317983 | tok/s 2629900.51 | norm 0.3726 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18248 | train 2.942121 | val 3.317983 | tok/s 2624121.01 | norm 0.3788 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18249 | train 2.959013 | val 3.317983 | tok/s 2629183.19 | norm 0.3488 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 18250 | val_loss 3.317458 step 18250 | train 3.009783 | val 3.317458 | tok/s 2613531.32 | norm 0.3761 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18251 | train 2.950371 | val 3.317458 | tok/s 2623570.98 | norm 0.4161 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18252 | train 3.043741 | val 3.317458 | tok/s 2633664.51 | norm 0.3831 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18253 | train 2.788141 | val 3.317458 | tok/s 2630504.30 | norm 0.3822 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18254 | train 2.874637 | val 3.317458 | tok/s 2629767.55 | norm 0.3749 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18255 | train 2.929334 | val 3.317458 | tok/s 2630871.78 | norm 0.3503 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18256 | train 2.894389 | val 3.317458 | tok/s 2629300.49 | norm 0.3487 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18257 | train 2.943149 | val 3.317458 | tok/s 2628481.43 | norm 0.4066 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18258 | train 2.886509 | val 3.317458 | tok/s 2626212.89 | norm 0.3704 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18259 | train 2.859557 | val 3.317458 | tok/s 2598550.06 | norm 0.3377 | dt 201.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18260 | train 2.834974 | val 3.317458 | tok/s 2453876.92 | norm 0.3426 | dt 213.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18261 | train 2.939997 | val 3.317458 | tok/s 2620445.92 | norm 0.3517 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18262 | train 3.044784 | val 3.317458 | tok/s 2629990.88 | norm 0.3887 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18263 | train 3.065201 | val 3.317458 | tok/s 2619835.25 | norm 0.3887 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18264 | train 3.003563 | val 3.317458 | tok/s 2628430.69 | norm 0.3582 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18265 | train 2.993827 | val 3.317458 | tok/s 2623753.34 | norm 0.3845 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18266 | train 3.065310 | val 3.317458 | tok/s 2628239.08 | norm 0.4005 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18267 | train 2.959345 | val 3.317458 | tok/s 2626999.08 | norm 0.3327 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18268 | train 3.053247 | val 3.317458 | tok/s 2625769.41 | norm 0.3478 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18269 | train 3.051803 | val 3.317458 | tok/s 2627625.88 | norm 0.3500 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18270 | train 2.945051 | val 3.317458 | tok/s 2630028.42 | norm 0.3484 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18271 | train 3.053583 | val 3.317458 | tok/s 2627273.61 | norm 0.3646 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18272 | train 3.094909 | val 3.317458 | tok/s 2632331.88 | norm 0.3815 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18273 | train 2.840192 | val 3.317458 | tok/s 2630932.67 | norm 0.3560 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18274 | train 2.953439 | val 3.317458 | tok/s 2628207.70 | norm 0.3627 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18275 | train 2.861971 | val 3.317458 | tok/s 2631748.18 | norm 0.3519 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18276 | train 3.071303 | val 3.317458 | tok/s 2620253.27 | norm 0.3599 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18277 | train 3.006585 | val 3.317458 | tok/s 2630112.43 | norm 0.3758 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18278 | train 2.795796 | val 3.317458 | tok/s 2627804.91 | norm 0.3460 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18279 | train 2.860089 | val 3.317458 | tok/s 2630161.14 | norm 0.3674 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18280 | train 2.923353 | val 3.317458 | tok/s 2631964.08 | norm 0.3974 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18281 | train 2.940283 | val 3.317458 | tok/s 2631426.09 | norm 0.3845 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18282 | train 2.983513 | val 3.317458 | tok/s 2626373.73 | norm 0.3798 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18283 | train 3.003809 | val 3.317458 | tok/s 2629670.99 | norm 0.3575 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18284 | train 2.837937 | val 3.317458 | tok/s 2624465.10 | norm 0.4075 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18285 | train 2.772523 | val 3.317458 | tok/s 2628703.36 | norm 0.3827 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18286 | train 2.956436 | val 3.317458 | tok/s 2628705.11 | norm 0.3580 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18287 | train 2.881921 | val 3.317458 | tok/s 2628758.09 | norm 0.3559 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18288 | train 2.905914 | val 3.317458 | tok/s 2628631.84 | norm 0.3984 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18289 | train 2.992526 | val 3.317458 | tok/s 2631245.23 | norm 0.3847 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18290 | train 2.967402 | val 3.317458 | tok/s 2624373.62 | norm 0.3871 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18291 | train 2.816593 | val 3.317458 | tok/s 2627461.55 | norm 0.4332 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18292 | train 2.895656 | val 3.317458 | tok/s 2627350.43 | norm 0.3973 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18293 | train 3.011767 | val 3.317458 | tok/s 2629586.39 | norm 0.3814 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18294 | train 2.810932 | val 3.317458 | tok/s 2629877.45 | norm 0.4042 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18295 | train 3.018846 | val 3.317458 | tok/s 2624129.96 | norm 0.3667 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18296 | train 2.764529 | val 3.317458 | tok/s 2626252.66 | norm 0.3559 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18297 | train 2.971047 | val 3.317458 | tok/s 2631995.41 | norm 0.3575 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18298 | train 2.893778 | val 3.317458 | tok/s 2631054.86 | norm 0.3675 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18299 | train 2.935113 | val 3.317458 | tok/s 2626418.10 | norm 0.3840 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18300 | train 2.914352 | val 3.317458 | tok/s 2627286.80 | norm 0.3781 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18301 | train 2.935332 | val 3.317458 | tok/s 2626296.21 | norm 0.3683 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18302 | train 2.983939 | val 3.317458 | tok/s 2630705.67 | norm 0.3680 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18303 | train 2.926632 | val 3.317458 | tok/s 2626639.77 | norm 0.4018 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18304 | train 3.076730 | val 3.317458 | tok/s 2629784.52 | norm 0.3876 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18305 | train 2.981750 | val 3.317458 | tok/s 2618423.97 | norm 0.3562 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18306 | train 2.883403 | val 3.317458 | tok/s 2629880.67 | norm 0.3899 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18307 | train 3.069608 | val 3.317458 | tok/s 2625750.09 | norm 0.3935 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18308 | train 2.891791 | val 3.317458 | tok/s 2630303.90 | norm 0.3696 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18309 | train 2.907992 | val 3.317458 | tok/s 2629943.07 | norm 0.3724 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18310 | train 2.966676 | val 3.317458 | tok/s 2628416.16 | norm 0.3905 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18311 | train 2.907898 | val 3.317458 | tok/s 2627902.39 | norm 0.3928 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18312 | train 2.922291 | val 3.317458 | tok/s 2632969.67 | norm 0.3818 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18313 | train 2.873605 | val 3.317458 | tok/s 2631553.52 | norm 0.3770 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18314 | train 2.910293 | val 3.317458 | tok/s 2627071.31 | norm 0.4037 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18315 | train 2.911424 | val 3.317458 | tok/s 2625922.11 | norm 0.3419 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18316 | train 2.981884 | val 3.317458 | tok/s 2627380.10 | norm 0.3335 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18317 | train 3.027491 | val 3.317458 | tok/s 2630258.39 | norm 0.3567 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18318 | train 2.941996 | val 3.317458 | tok/s 2625348.86 | norm 0.3380 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18319 | train 2.873654 | val 3.317458 | tok/s 2629426.42 | norm 0.3679 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18320 | train 2.991582 | val 3.317458 | tok/s 2627968.03 | norm 0.3812 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18321 | train 2.904403 | val 3.317458 | tok/s 2630874.74 | norm 0.3715 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18322 | train 2.841367 | val 3.317458 | tok/s 2632068.27 | norm 0.3430 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18323 | train 2.982114 | val 3.317458 | tok/s 2622455.35 | norm 0.3477 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18324 | train 3.018430 | val 3.317458 | tok/s 2630505.95 | norm 0.3768 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18325 | train 2.985443 | val 3.317458 | tok/s 2569745.57 | norm 0.3985 | dt 204.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18326 | train 2.989781 | val 3.317458 | tok/s 2571186.77 | norm 0.3703 | dt 203.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18327 | train 2.835286 | val 3.317458 | tok/s 2625316.03 | norm 0.4294 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18328 | train 2.922389 | val 3.317458 | tok/s 2626821.32 | norm 0.4078 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18329 | train 2.876416 | val 3.317458 | tok/s 2628860.43 | norm 0.3593 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18330 | train 3.014757 | val 3.317458 | tok/s 2628683.42 | norm 0.3857 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18331 | train 3.001309 | val 3.317458 | tok/s 2630145.86 | norm 0.3617 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18332 | train 2.979459 | val 3.317458 | tok/s 2622812.07 | norm 0.3758 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18333 | train 2.874171 | val 3.317458 | tok/s 2610595.39 | norm 0.3486 | dt 200.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18334 | train 2.938576 | val 3.317458 | tok/s 2625644.17 | norm 0.3422 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18335 | train 2.908704 | val 3.317458 | tok/s 2628901.73 | norm 0.3652 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18336 | train 2.863385 | val 3.317458 | tok/s 2627385.45 | norm 0.3578 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18337 | train 2.967186 | val 3.317458 | tok/s 2628180.52 | norm 0.3470 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18338 | train 3.000115 | val 3.317458 | tok/s 2626422.27 | norm 0.3621 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18339 | train 2.948175 | val 3.317458 | tok/s 2631093.18 | norm 0.3674 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18340 | train 2.792783 | val 3.317458 | tok/s 2628287.15 | norm 0.4147 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18341 | train 2.979194 | val 3.317458 | tok/s 2626835.41 | norm 0.3631 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18342 | train 2.943875 | val 3.317458 | tok/s 2624359.13 | norm 0.3461 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18343 | train 2.971088 | val 3.317458 | tok/s 2625862.04 | norm 0.3755 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18344 | train 2.993458 | val 3.317458 | tok/s 2633346.08 | norm 0.3785 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18345 | train 2.989614 | val 3.317458 | tok/s 2624967.46 | norm 0.3787 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18346 | train 3.030948 | val 3.317458 | tok/s 2630683.55 | norm 0.4259 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18347 | train 3.010882 | val 3.317458 | tok/s 2627051.18 | norm 0.3661 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18348 | train 2.914950 | val 3.317458 | tok/s 2627380.53 | norm 0.3790 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18349 | train 3.143563 | val 3.317458 | tok/s 2631831.88 | norm 0.4200 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18350 | train 3.031439 | val 3.317458 | tok/s 2629741.65 | norm 0.4160 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18351 | train 2.888758 | val 3.317458 | tok/s 2631070.13 | norm 0.3992 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18352 | train 2.881727 | val 3.317458 | tok/s 2623399.27 | norm 0.3586 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18353 | train 2.905309 | val 3.317458 | tok/s 2629310.60 | norm 0.3613 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18354 | train 2.929129 | val 3.317458 | tok/s 2631471.13 | norm 0.4094 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18355 | train 2.999335 | val 3.317458 | tok/s 2627785.68 | norm 0.3767 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18356 | train 2.952256 | val 3.317458 | tok/s 2629329.59 | norm 0.3537 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18357 | train 3.032141 | val 3.317458 | tok/s 2625525.96 | norm 0.3594 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18358 | train 3.095114 | val 3.317458 | tok/s 2629544.32 | norm 0.3874 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18359 | train 2.868238 | val 3.317458 | tok/s 2633407.65 | norm 0.3532 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18360 | train 2.892970 | val 3.317458 | tok/s 2629888.15 | norm 0.3673 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18361 | train 2.925743 | val 3.317458 | tok/s 2631999.79 | norm 0.3955 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18362 | train 2.938648 | val 3.317458 | tok/s 2625914.48 | norm 0.3584 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18363 | train 2.922928 | val 3.317458 | tok/s 2627692.79 | norm 0.3802 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18364 | train 2.927656 | val 3.317458 | tok/s 2628772.16 | norm 0.4058 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18365 | train 2.829011 | val 3.317458 | tok/s 2631568.55 | norm 0.4038 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18366 | train 2.967803 | val 3.317458 | tok/s 2630834.60 | norm 0.4150 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18367 | train 2.869584 | val 3.317458 | tok/s 2523460.87 | norm 0.4049 | dt 207.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18368 | train 2.989856 | val 3.317458 | tok/s 2594298.58 | norm 0.3739 | dt 202.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18369 | train 3.059740 | val 3.317458 | tok/s 2625147.00 | norm 0.3506 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18370 | train 2.845907 | val 3.317458 | tok/s 2624622.32 | norm 0.3835 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18371 | train 2.965341 | val 3.317458 | tok/s 2628701.41 | norm 0.4045 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18372 | train 2.954088 | val 3.317458 | tok/s 2631158.78 | norm 0.3793 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18373 | train 2.925148 | val 3.317458 | tok/s 2626716.21 | norm 0.3779 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18374 | train 2.943032 | val 3.317458 | tok/s 2628199.83 | norm 0.4175 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18375 | train 2.925437 | val 3.317458 | tok/s 2626568.26 | norm 0.3938 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18376 | train 2.904651 | val 3.317458 | tok/s 2626394.56 | norm 0.3537 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18377 | train 2.890702 | val 3.317458 | tok/s 2623816.06 | norm 0.3573 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18378 | train 3.103608 | val 3.317458 | tok/s 2630171.95 | norm 0.4128 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18379 | train 2.878113 | val 3.317458 | tok/s 2630208.27 | norm 0.3570 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18380 | train 3.028638 | val 3.317458 | tok/s 2623829.26 | norm 0.3731 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18381 | train 3.006222 | val 3.317458 | tok/s 2626285.73 | norm 0.3784 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18382 | train 2.951022 | val 3.317458 | tok/s 2622396.22 | norm 0.3706 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18383 | train 2.866357 | val 3.317458 | tok/s 2626465.31 | norm 0.3777 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18384 | train 2.952708 | val 3.317458 | tok/s 2623672.12 | norm 0.4039 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18385 | train 2.866566 | val 3.317458 | tok/s 2622966.83 | norm 0.3265 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18386 | train 3.004478 | val 3.317458 | tok/s 2624689.34 | norm 0.3609 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18387 | train 3.014161 | val 3.317458 | tok/s 2626173.28 | norm 0.3664 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18388 | train 2.950163 | val 3.317458 | tok/s 2623826.53 | norm 0.3477 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18389 | train 2.930086 | val 3.317458 | tok/s 2625790.32 | norm 0.3534 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18390 | train 3.090938 | val 3.317458 | tok/s 2623448.76 | norm 0.3363 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18391 | train 2.904086 | val 3.317458 | tok/s 2624604.71 | norm 0.3483 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18392 | train 2.889904 | val 3.317458 | tok/s 2628847.55 | norm 0.3350 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18393 | train 2.931624 | val 3.317458 | tok/s 2628219.49 | norm 0.3992 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18394 | train 3.038821 | val 3.317458 | tok/s 2627504.33 | norm 0.3909 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18395 | train 2.844222 | val 3.317458 | tok/s 2628239.20 | norm 0.3921 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18396 | train 2.952599 | val 3.317458 | tok/s 2627613.95 | norm 0.3483 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18397 | train 3.075745 | val 3.317458 | tok/s 2628043.56 | norm 0.3481 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18398 | train 2.972504 | val 3.317458 | tok/s 2627504.91 | norm 0.3723 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18399 | train 2.857209 | val 3.317458 | tok/s 2620684.77 | norm 0.4068 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18400 | train 2.962699 | val 3.317458 | tok/s 2627993.22 | norm 0.3830 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18401 | train 2.943747 | val 3.317458 | tok/s 2631165.39 | norm 0.3449 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18402 | train 2.931335 | val 3.317458 | tok/s 2621309.24 | norm 0.3503 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18403 | train 2.923280 | val 3.317458 | tok/s 2626947.84 | norm 0.3486 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18404 | train 2.902349 | val 3.317458 | tok/s 2623218.73 | norm 0.3362 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18405 | train 2.939137 | val 3.317458 | tok/s 2628142.42 | norm 0.3459 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18406 | train 3.027155 | val 3.317458 | tok/s 2627452.98 | norm 0.3578 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18407 | train 3.069450 | val 3.317458 | tok/s 2628410.15 | norm 0.3325 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18408 | train 2.979029 | val 3.317458 | tok/s 2630904.03 | norm 0.3642 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18409 | train 3.017847 | val 3.317458 | tok/s 2629243.30 | norm 0.4420 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18410 | train 2.935377 | val 3.317458 | tok/s 2626495.72 | norm 0.3749 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18411 | train 2.856798 | val 3.317458 | tok/s 2629583.91 | norm 0.3380 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18412 | train 2.958527 | val 3.317458 | tok/s 2624359.48 | norm 0.3763 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18413 | train 2.929797 | val 3.317458 | tok/s 2629051.30 | norm 0.3770 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18414 | train 2.869459 | val 3.317458 | tok/s 2622872.00 | norm 0.3448 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18415 | train 3.137372 | val 3.317458 | tok/s 2627960.47 | norm 0.3734 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18416 | train 2.996896 | val 3.317458 | tok/s 2630881.09 | norm 0.3893 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18417 | train 2.992325 | val 3.317458 | tok/s 2629902.61 | norm 0.3577 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18418 | train 2.796763 | val 3.317458 | tok/s 2626695.50 | norm 0.3600 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18419 | train 3.053035 | val 3.317458 | tok/s 2623916.12 | norm 0.4027 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18420 | train 2.909470 | val 3.317458 | tok/s 2575443.35 | norm 0.3952 | dt 203.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18421 | train 2.947309 | val 3.317458 | tok/s 2582197.89 | norm 0.3563 | dt 203.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18422 | train 2.940586 | val 3.317458 | tok/s 2625860.06 | norm 0.4069 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18423 | train 2.968385 | val 3.317458 | tok/s 2632751.89 | norm 0.3639 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18424 | train 3.034019 | val 3.317458 | tok/s 2624204.77 | norm 0.4115 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18425 | train 2.924645 | val 3.317458 | tok/s 2630409.83 | norm 0.3789 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18426 | train 3.090743 | val 3.317458 | tok/s 2631386.36 | norm 0.3612 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18427 | train 2.932560 | val 3.317458 | tok/s 2624150.73 | norm 0.3597 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18428 | train 2.852073 | val 3.317458 | tok/s 2629651.47 | norm 0.3727 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18429 | train 2.983498 | val 3.317458 | tok/s 2632007.43 | norm 0.3698 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18430 | train 2.928504 | val 3.317458 | tok/s 2623831.23 | norm 0.3666 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18431 | train 2.996868 | val 3.317458 | tok/s 2634041.95 | norm 0.3626 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18432 | train 2.913396 | val 3.317458 | tok/s 2626026.42 | norm 0.3582 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18433 | train 2.921360 | val 3.317458 | tok/s 2629052.74 | norm 0.3788 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18434 | train 2.860972 | val 3.317458 | tok/s 2628128.55 | norm 0.3928 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18435 | train 2.860830 | val 3.317458 | tok/s 2630978.55 | norm 0.3955 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18436 | train 2.785810 | val 3.317458 | tok/s 2630440.67 | norm 0.4064 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18437 | train 2.868543 | val 3.317458 | tok/s 2626058.15 | norm 0.3849 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18438 | train 3.019443 | val 3.317458 | tok/s 2632050.16 | norm 0.3557 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18439 | train 3.064348 | val 3.317458 | tok/s 2630923.00 | norm 0.4000 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18440 | train 2.899787 | val 3.317458 | tok/s 2632429.51 | norm 0.3550 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18441 | train 2.911411 | val 3.317458 | tok/s 2628261.05 | norm 0.3661 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18442 | train 3.007910 | val 3.317458 | tok/s 2623946.83 | norm 0.3908 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18443 | train 2.954659 | val 3.317458 | tok/s 2631425.80 | norm 0.4025 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18444 | train 2.842772 | val 3.317458 | tok/s 2626080.94 | norm 0.3688 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18445 | train 3.134081 | val 3.317458 | tok/s 2628185.44 | norm 0.4527 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18446 | train 2.865538 | val 3.317458 | tok/s 2621391.82 | norm 0.3849 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18447 | train 2.801422 | val 3.317458 | tok/s 2627280.34 | norm 0.3545 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18448 | train 3.009425 | val 3.317458 | tok/s 2634040.22 | norm 0.3861 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18449 | train 2.865234 | val 3.317458 | tok/s 2625027.23 | norm 0.4139 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18450 | train 2.892858 | val 3.317458 | tok/s 2627030.82 | norm 0.3783 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18451 | train 2.893644 | val 3.317458 | tok/s 2625589.14 | norm 0.3927 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18452 | train 3.007965 | val 3.317458 | tok/s 2608456.67 | norm 0.4115 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18453 | train 3.004264 | val 3.317458 | tok/s 2630918.37 | norm 0.3891 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18454 | train 2.941537 | val 3.317458 | tok/s 2625684.98 | norm 0.3589 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18455 | train 2.920954 | val 3.317458 | tok/s 2625242.40 | norm 0.3758 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18456 | train 2.910525 | val 3.317458 | tok/s 2626765.64 | norm 0.3968 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18457 | train 3.027000 | val 3.317458 | tok/s 2629641.46 | norm 0.3572 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18458 | train 2.940858 | val 3.317458 | tok/s 2627882.41 | norm 0.3714 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18459 | train 3.017811 | val 3.317458 | tok/s 2628304.94 | norm 0.3422 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18460 | train 3.091305 | val 3.317458 | tok/s 2628679.71 | norm 0.3277 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18461 | train 3.056142 | val 3.317458 | tok/s 2625644.91 | norm 0.3550 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18462 | train 2.998312 | val 3.317458 | tok/s 2629016.69 | norm 0.3639 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18463 | train 2.986881 | val 3.317458 | tok/s 2629047.93 | norm 0.4473 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18464 | train 2.946972 | val 3.317458 | tok/s 2629026.33 | norm 0.3835 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18465 | train 2.849205 | val 3.317458 | tok/s 2629445.90 | norm 0.3425 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18466 | train 2.989948 | val 3.317458 | tok/s 2624885.44 | norm 0.3423 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18467 | train 2.817578 | val 3.317458 | tok/s 2629146.41 | norm 0.4001 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18468 | train 2.985353 | val 3.317458 | tok/s 2622834.89 | norm 0.4298 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18469 | train 2.792713 | val 3.317458 | tok/s 2628332.54 | norm 0.3963 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18470 | train 2.995214 | val 3.317458 | tok/s 2629340.48 | norm 0.3922 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18471 | train 2.971327 | val 3.317458 | tok/s 2627847.63 | norm 0.3747 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18472 | train 2.995570 | val 3.317458 | tok/s 2627897.05 | norm 0.3815 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18473 | train 2.939654 | val 3.317458 | tok/s 2621487.03 | norm 0.3985 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18474 | train 2.938815 | val 3.317458 | tok/s 2625968.86 | norm 0.4214 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18475 | train 2.979442 | val 3.317458 | tok/s 2629866.06 | norm 0.3679 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18476 | train 2.864001 | val 3.317458 | tok/s 2626562.45 | norm 0.3826 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18477 | train 2.918298 | val 3.317458 | tok/s 2624183.93 | norm 0.4041 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18478 | train 3.001445 | val 3.317458 | tok/s 2632288.60 | norm 0.3835 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18479 | train 2.983632 | val 3.317458 | tok/s 2626796.79 | norm 0.3826 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18480 | train 3.025411 | val 3.317458 | tok/s 2632147.50 | norm 0.4014 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18481 | train 2.851503 | val 3.317458 | tok/s 2495936.26 | norm 0.3451 | dt 210.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18482 | train 2.860167 | val 3.317458 | tok/s 2620362.99 | norm 0.3486 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18483 | train 2.900593 | val 3.317458 | tok/s 2625652.58 | norm 0.3628 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18484 | train 3.076394 | val 3.317458 | tok/s 2626496.47 | norm 0.4247 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18485 | train 2.968088 | val 3.317458 | tok/s 2629774.35 | norm 0.3915 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18486 | train 3.034618 | val 3.317458 | tok/s 2628927.66 | norm 0.3396 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18487 | train 3.047969 | val 3.317458 | tok/s 2629049.82 | norm 0.3673 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18488 | train 3.084177 | val 3.317458 | tok/s 2631793.77 | norm 0.3450 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18489 | train 2.888096 | val 3.317458 | tok/s 2626141.25 | norm 0.3486 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18490 | train 2.968582 | val 3.317458 | tok/s 2632582.58 | norm 0.3857 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18491 | train 2.899550 | val 3.317458 | tok/s 2620093.86 | norm 0.4038 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18492 | train 2.914131 | val 3.317458 | tok/s 2628257.75 | norm 0.3909 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18493 | train 2.940775 | val 3.317458 | tok/s 2609166.01 | norm 0.3953 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18494 | train 2.888211 | val 3.317458 | tok/s 2627745.22 | norm 0.3559 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18495 | train 2.948466 | val 3.317458 | tok/s 2629315.37 | norm 0.3407 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18496 | train 2.991279 | val 3.317458 | tok/s 2626168.56 | norm 0.3714 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18497 | train 2.949543 | val 3.317458 | tok/s 2630138.88 | norm 0.3869 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18498 | train 2.931504 | val 3.317458 | tok/s 2631112.56 | norm 0.3605 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18499 | train 2.921150 | val 3.317458 | tok/s 2625493.00 | norm 0.3797 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 18500 | val_loss 3.315466 step 18500 | train 3.017075 | val 3.315466 | tok/s 2616231.61 | norm 0.3962 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18501 | train 2.889150 | val 3.315466 | tok/s 2629732.64 | norm 0.3749 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18502 | train 2.929227 | val 3.315466 | tok/s 2630886.96 | norm 0.3581 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18503 | train 2.943198 | val 3.315466 | tok/s 2628247.96 | norm 0.3482 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18504 | train 2.891817 | val 3.315466 | tok/s 2616914.24 | norm 0.3665 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18505 | train 2.992561 | val 3.315466 | tok/s 2627295.30 | norm 0.3905 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18506 | train 2.892189 | val 3.315466 | tok/s 2631263.61 | norm 0.3518 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18507 | train 2.803948 | val 3.315466 | tok/s 2627887.85 | norm 0.3681 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18508 | train 2.914396 | val 3.315466 | tok/s 2630229.24 | norm 0.3697 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18509 | train 3.021658 | val 3.315466 | tok/s 2615453.67 | norm 0.3207 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18510 | train 2.905958 | val 3.315466 | tok/s 2591668.94 | norm 0.3065 | dt 202.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18511 | train 2.944108 | val 3.315466 | tok/s 2618766.29 | norm 0.3350 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18512 | train 2.961822 | val 3.315466 | tok/s 2624400.72 | norm 0.3867 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18513 | train 3.015454 | val 3.315466 | tok/s 2626926.61 | norm 0.3983 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18514 | train 2.998797 | val 3.315466 | tok/s 2624979.02 | norm 0.3881 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18515 | train 2.901598 | val 3.315466 | tok/s 2629017.45 | norm 0.3532 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18516 | train 2.916043 | val 3.315466 | tok/s 2622812.24 | norm 0.3810 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18517 | train 2.913896 | val 3.315466 | tok/s 2624790.85 | norm 0.3893 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18518 | train 2.856014 | val 3.315466 | tok/s 2626387.31 | norm 0.3998 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18519 | train 3.060907 | val 3.315466 | tok/s 2455544.17 | norm 0.3859 | dt 213.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18520 | train 2.977888 | val 3.315466 | tok/s 2628582.91 | norm 0.3963 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18521 | train 2.743167 | val 3.315466 | tok/s 2630146.39 | norm 0.3682 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18522 | train 2.887686 | val 3.315466 | tok/s 2627158.06 | norm 0.3671 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18523 | train 3.001179 | val 3.315466 | tok/s 2630882.80 | norm 0.3489 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18524 | train 2.910320 | val 3.315466 | tok/s 2629550.45 | norm 0.3611 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18525 | train 2.851045 | val 3.315466 | tok/s 2627677.13 | norm 0.3755 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18526 | train 2.936941 | val 3.315466 | tok/s 2631046.48 | norm 0.3553 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18527 | train 2.907535 | val 3.315466 | tok/s 2598686.22 | norm 0.3663 | dt 201.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18528 | train 2.890855 | val 3.315466 | tok/s 2631783.54 | norm 0.3821 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18529 | train 2.929223 | val 3.315466 | tok/s 2628298.45 | norm 0.3968 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18530 | train 2.878824 | val 3.315466 | tok/s 2625970.93 | norm 0.4083 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18531 | train 2.965838 | val 3.315466 | tok/s 2626657.79 | norm 0.3977 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18532 | train 3.023264 | val 3.315466 | tok/s 2629678.76 | norm 0.3916 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18533 | train 2.847535 | val 3.315466 | tok/s 2628589.63 | norm 0.4394 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18534 | train 2.945614 | val 3.315466 | tok/s 2629216.74 | norm 0.3897 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18535 | train 2.947079 | val 3.315466 | tok/s 2628968.13 | norm 0.3823 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18536 | train 2.889228 | val 3.315466 | tok/s 2627023.54 | norm 0.3452 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18537 | train 3.032169 | val 3.315466 | tok/s 2627814.30 | norm 0.3681 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18538 | train 2.887107 | val 3.315466 | tok/s 2626993.93 | norm 0.3749 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18539 | train 3.117369 | val 3.315466 | tok/s 2630765.31 | norm 0.3715 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18540 | train 2.975113 | val 3.315466 | tok/s 2630117.23 | norm 0.3580 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18541 | train 2.889664 | val 3.315466 | tok/s 2630546.04 | norm 0.3562 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18542 | train 2.878229 | val 3.315466 | tok/s 2622143.30 | norm 0.3647 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18543 | train 3.041306 | val 3.315466 | tok/s 2630798.88 | norm 0.3416 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18544 | train 2.938402 | val 3.315466 | tok/s 2627712.44 | norm 0.3608 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18545 | train 2.955191 | val 3.315466 | tok/s 2630894.30 | norm 0.3752 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18546 | train 2.929483 | val 3.315466 | tok/s 2633274.82 | norm 0.4024 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18547 | train 2.977332 | val 3.315466 | tok/s 2625364.43 | norm 0.3988 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18548 | train 3.062427 | val 3.315466 | tok/s 2634497.13 | norm 0.4374 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18549 | train 2.986879 | val 3.315466 | tok/s 2627554.64 | norm 0.4176 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18550 | train 2.863229 | val 3.315466 | tok/s 2630670.42 | norm 0.4012 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18551 | train 3.010222 | val 3.315466 | tok/s 2630823.00 | norm 0.3848 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18552 | train 2.982389 | val 3.315466 | tok/s 2629820.42 | norm 0.3887 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18553 | train 2.962363 | val 3.315466 | tok/s 2630438.33 | norm 0.3570 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18554 | train 3.148446 | val 3.315466 | tok/s 2630973.97 | norm 0.4502 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18555 | train 2.923460 | val 3.315466 | tok/s 2605434.39 | norm 0.3942 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18556 | train 2.943603 | val 3.315466 | tok/s 2630017.98 | norm 0.8592 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18557 | train 2.960210 | val 3.315466 | tok/s 2617013.05 | norm 0.3806 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18558 | train 2.950551 | val 3.315466 | tok/s 2633264.89 | norm 0.3793 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18559 | train 2.984608 | val 3.315466 | tok/s 2629864.26 | norm 0.3776 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18560 | train 3.049945 | val 3.315466 | tok/s 2631963.71 | norm 0.3738 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18561 | train 2.941521 | val 3.315466 | tok/s 2626269.07 | norm 0.3560 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18562 | train 2.966658 | val 3.315466 | tok/s 2627924.88 | norm 0.3936 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18563 | train 2.866949 | val 3.315466 | tok/s 2629366.30 | norm 0.3516 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18564 | train 2.860464 | val 3.315466 | tok/s 2627095.53 | norm 0.3694 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18565 | train 2.937198 | val 3.315466 | tok/s 2631370.30 | norm 0.3827 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18566 | train 2.936610 | val 3.315466 | tok/s 2617773.14 | norm 0.3351 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18567 | train 3.069148 | val 3.315466 | tok/s 2624709.06 | norm 0.3503 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18568 | train 2.884037 | val 3.315466 | tok/s 2632059.16 | norm 0.3398 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18569 | train 2.889617 | val 3.315466 | tok/s 2628343.66 | norm 0.3463 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18570 | train 3.108844 | val 3.315466 | tok/s 2626779.10 | norm 0.4445 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18571 | train 2.997939 | val 3.315466 | tok/s 2619721.52 | norm 0.3588 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18572 | train 2.890248 | val 3.315466 | tok/s 2621330.13 | norm 0.3768 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18573 | train 2.829546 | val 3.315466 | tok/s 2626707.47 | norm 0.3924 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18574 | train 2.900469 | val 3.315466 | tok/s 2628014.66 | norm 0.3732 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18575 | train 3.126355 | val 3.315466 | tok/s 2629131.64 | norm 0.3991 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18576 | train 2.980634 | val 3.315466 | tok/s 2624104.48 | norm 0.3527 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18577 | train 2.927064 | val 3.315466 | tok/s 2628023.08 | norm 0.3731 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18578 | train 2.884611 | val 3.315466 | tok/s 2627967.16 | norm 0.3757 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18579 | train 2.988313 | val 3.315466 | tok/s 2629586.30 | norm 0.3605 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18580 | train 2.960489 | val 3.315466 | tok/s 2633315.67 | norm 0.3764 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18581 | train 2.997202 | val 3.315466 | tok/s 2630391.84 | norm 0.3894 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18582 | train 3.091432 | val 3.315466 | tok/s 2630931.48 | norm 0.3803 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18583 | train 2.943268 | val 3.315466 | tok/s 2630852.43 | norm 0.3682 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18584 | train 2.917300 | val 3.315466 | tok/s 2629295.45 | norm 0.3507 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18585 | train 2.969779 | val 3.315466 | tok/s 2631981.53 | norm 0.3349 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18586 | train 3.024384 | val 3.315466 | tok/s 2627877.76 | norm 0.3585 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18587 | train 3.028608 | val 3.315466 | tok/s 2628946.06 | norm 0.4092 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18588 | train 2.942323 | val 3.315466 | tok/s 2631410.80 | norm 0.3616 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18589 | train 2.989226 | val 3.315466 | tok/s 2627956.78 | norm 0.3693 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18590 | train 3.035225 | val 3.315466 | tok/s 2627626.44 | norm 0.4800 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18591 | train 2.786349 | val 3.315466 | tok/s 2626502.20 | norm 0.3735 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18592 | train 2.869676 | val 3.315466 | tok/s 2626326.94 | norm 0.3579 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18593 | train 2.886546 | val 3.315466 | tok/s 2634333.53 | norm 0.4284 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18594 | train 2.964053 | val 3.315466 | tok/s 2627825.34 | norm 0.3648 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18595 | train 2.963458 | val 3.315466 | tok/s 2631355.06 | norm 0.3593 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18596 | train 2.908612 | val 3.315466 | tok/s 2626966.48 | norm 0.3883 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18597 | train 3.079326 | val 3.315466 | tok/s 2563167.62 | norm 0.3770 | dt 204.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18598 | train 3.065839 | val 3.315466 | tok/s 2594791.61 | norm 0.4909 | dt 202.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18599 | train 2.960773 | val 3.315466 | tok/s 2565146.21 | norm 0.3901 | dt 204.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18600 | train 2.954187 | val 3.315466 | tok/s 2628756.87 | norm 0.3609 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18601 | train 2.857672 | val 3.315466 | tok/s 2627905.41 | norm 0.3840 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18602 | train 2.910279 | val 3.315466 | tok/s 2624617.68 | norm 0.3910 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18603 | train 3.063973 | val 3.315466 | tok/s 2625646.27 | norm 0.3842 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18604 | train 2.997742 | val 3.315466 | tok/s 2625180.97 | norm 0.3967 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18605 | train 3.027695 | val 3.315466 | tok/s 2628144.08 | norm 0.3702 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18606 | train 2.921240 | val 3.315466 | tok/s 2626948.32 | norm 0.3507 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18607 | train 3.055297 | val 3.315466 | tok/s 2628362.41 | norm 0.4470 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18608 | train 2.860818 | val 3.315466 | tok/s 2626715.03 | norm 0.3293 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18609 | train 3.098120 | val 3.315466 | tok/s 2624025.68 | norm 0.4727 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18610 | train 2.966687 | val 3.315466 | tok/s 2628915.37 | norm 0.4251 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18611 | train 2.913536 | val 3.315466 | tok/s 2627435.03 | norm 0.3569 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18612 | train 2.953048 | val 3.315466 | tok/s 2629218.23 | norm 0.3464 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18613 | train 3.060099 | val 3.315466 | tok/s 2625552.64 | norm 0.3645 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18614 | train 2.941299 | val 3.315466 | tok/s 2573207.19 | norm 0.3815 | dt 203.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18615 | train 3.011910 | val 3.315466 | tok/s 2630127.56 | norm 0.3813 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18616 | train 2.829390 | val 3.315466 | tok/s 2625834.77 | norm 0.3814 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18617 | train 2.887463 | val 3.315466 | tok/s 2628626.27 | norm 0.3752 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18618 | train 2.867065 | val 3.315466 | tok/s 2620420.13 | norm 0.3859 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18619 | train 2.870758 | val 3.315466 | tok/s 2624067.33 | norm 0.3744 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18620 | train 2.953812 | val 3.315466 | tok/s 2629290.48 | norm 0.3686 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18621 | train 2.943883 | val 3.315466 | tok/s 2627809.97 | norm 0.3995 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18622 | train 2.915267 | val 3.315466 | tok/s 2625332.74 | norm 0.3908 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18623 | train 3.021009 | val 3.315466 | tok/s 2626474.97 | norm 0.3758 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18624 | train 2.917945 | val 3.315466 | tok/s 2627021.13 | norm 0.4029 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18625 | train 3.063650 | val 3.315466 | tok/s 2634170.75 | norm 0.4179 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18626 | train 3.038560 | val 3.315466 | tok/s 2634187.39 | norm 0.4054 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18627 | train 2.902762 | val 3.315466 | tok/s 2626650.34 | norm 0.3562 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18628 | train 2.932156 | val 3.315466 | tok/s 2624394.21 | norm 0.3630 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18629 | train 3.018606 | val 3.315466 | tok/s 2628132.63 | norm 0.3668 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18630 | train 2.939115 | val 3.315466 | tok/s 2629081.44 | norm 0.3744 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18631 | train 2.975837 | val 3.315466 | tok/s 2627424.09 | norm 0.3732 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18632 | train 3.034136 | val 3.315466 | tok/s 2630172.61 | norm 0.3780 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18633 | train 2.906980 | val 3.315466 | tok/s 2604395.86 | norm 0.3808 | dt 201.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18634 | train 2.902947 | val 3.315466 | tok/s 2626845.71 | norm 0.3891 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18635 | train 2.900165 | val 3.315466 | tok/s 2628123.91 | norm 0.3670 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18636 | train 3.105698 | val 3.315466 | tok/s 2627394.73 | norm 0.3612 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18637 | train 2.910717 | val 3.315466 | tok/s 2631836.45 | norm 0.3411 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18638 | train 2.937589 | val 3.315466 | tok/s 2629386.37 | norm 0.3299 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18639 | train 2.928640 | val 3.315466 | tok/s 2627482.30 | norm 0.3450 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18640 | train 2.959320 | val 3.315466 | tok/s 2631007.13 | norm 0.3779 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18641 | train 2.997847 | val 3.315466 | tok/s 2610829.34 | norm 0.3445 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18642 | train 2.875938 | val 3.315466 | tok/s 2630957.46 | norm 0.3522 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18643 | train 2.937219 | val 3.315466 | tok/s 2628302.96 | norm 0.3787 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18644 | train 2.872487 | val 3.315466 | tok/s 2627003.22 | norm 0.3368 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18645 | train 2.933469 | val 3.315466 | tok/s 2630774.55 | norm 0.3595 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18646 | train 2.916201 | val 3.315466 | tok/s 2627345.43 | norm 0.4004 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18647 | train 3.028795 | val 3.315466 | tok/s 2627495.90 | norm 0.4038 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18648 | train 2.992259 | val 3.315466 | tok/s 2630834.99 | norm 0.3993 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18649 | train 3.057996 | val 3.315466 | tok/s 2631453.81 | norm 0.3655 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18650 | train 2.944796 | val 3.315466 | tok/s 2630450.99 | norm 0.4413 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18651 | train 3.080264 | val 3.315466 | tok/s 2626768.02 | norm 0.4225 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18652 | train 3.259509 | val 3.315466 | tok/s 2631026.72 | norm 0.5250 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18653 | train 3.029524 | val 3.315466 | tok/s 2623376.16 | norm 0.4392 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18654 | train 3.053474 | val 3.315466 | tok/s 2628787.61 | norm 0.4326 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18655 | train 3.009997 | val 3.315466 | tok/s 2629588.74 | norm 0.3493 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18656 | train 2.929518 | val 3.315466 | tok/s 2626438.15 | norm 0.3404 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18657 | train 3.051938 | val 3.315466 | tok/s 2628270.24 | norm 0.3994 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18658 | train 2.989142 | val 3.315466 | tok/s 2629908.50 | norm 0.4101 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18659 | train 2.943090 | val 3.315466 | tok/s 2630824.19 | norm 0.3929 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18660 | train 2.958622 | val 3.315466 | tok/s 2629134.33 | norm 0.3628 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18661 | train 3.118198 | val 3.315466 | tok/s 2623167.63 | norm 0.3572 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18662 | train 2.968536 | val 3.315466 | tok/s 2631328.16 | norm 0.3790 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18663 | train 3.006423 | val 3.315466 | tok/s 2628869.00 | norm 0.3866 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18664 | train 2.967145 | val 3.315466 | tok/s 2629658.75 | norm 0.3740 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18665 | train 2.795112 | val 3.315466 | tok/s 2627898.80 | norm 0.3514 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18666 | train 2.973809 | val 3.315466 | tok/s 2608696.71 | norm 0.3736 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18667 | train 3.012339 | val 3.315466 | tok/s 2629865.58 | norm 0.3741 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18668 | train 2.914373 | val 3.315466 | tok/s 2630010.13 | norm 0.3562 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18669 | train 2.955316 | val 3.315466 | tok/s 2630876.92 | norm 0.3563 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18670 | train 2.972247 | val 3.315466 | tok/s 2625795.70 | norm 0.3836 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18671 | train 3.012405 | val 3.315466 | tok/s 2627545.47 | norm 0.3905 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18672 | train 2.971564 | val 3.315466 | tok/s 2633344.65 | norm 0.3571 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18673 | train 2.980932 | val 3.315466 | tok/s 2630836.40 | norm 0.3375 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18674 | train 2.957214 | val 3.315466 | tok/s 2631164.76 | norm 0.3581 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18675 | train 2.999015 | val 3.315466 | tok/s 2628590.63 | norm 0.3636 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18676 | train 2.916993 | val 3.315466 | tok/s 2623139.49 | norm 0.3664 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18677 | train 2.960770 | val 3.315466 | tok/s 2631365.51 | norm 0.3871 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18678 | train 3.024218 | val 3.315466 | tok/s 2624144.78 | norm 0.3433 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18679 | train 2.988742 | val 3.315466 | tok/s 2626449.86 | norm 0.3486 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18680 | train 2.787799 | val 3.315466 | tok/s 2627535.94 | norm 0.3600 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18681 | train 2.932178 | val 3.315466 | tok/s 2629914.29 | norm 0.3919 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18682 | train 3.060150 | val 3.315466 | tok/s 2630481.02 | norm 0.4334 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18683 | train 2.942618 | val 3.315466 | tok/s 2628061.80 | norm 0.4030 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18684 | train 2.954701 | val 3.315466 | tok/s 2626526.16 | norm 0.3620 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18685 | train 2.911305 | val 3.315466 | tok/s 2631286.25 | norm 0.4199 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18686 | train 2.944517 | val 3.315466 | tok/s 2628438.78 | norm 0.4035 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18687 | train 2.790029 | val 3.315466 | tok/s 2632471.75 | norm 0.4277 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18688 | train 2.960191 | val 3.315466 | tok/s 2628007.84 | norm 0.4478 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18689 | train 2.936541 | val 3.315466 | tok/s 2628453.16 | norm 0.3695 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18690 | train 2.875363 | val 3.315466 | tok/s 2623285.17 | norm 0.4076 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18691 | train 3.163054 | val 3.315466 | tok/s 2622226.89 | norm 0.4543 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18692 | train 2.945338 | val 3.315466 | tok/s 2628705.29 | norm 0.3841 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18693 | train 3.018235 | val 3.315466 | tok/s 2628510.05 | norm 0.3600 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18694 | train 2.977640 | val 3.315466 | tok/s 2626424.09 | norm 0.3910 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18695 | train 2.998219 | val 3.315466 | tok/s 2627942.07 | norm 0.3927 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18696 | train 2.849915 | val 3.315466 | tok/s 2629900.81 | norm 0.3832 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18697 | train 2.977231 | val 3.315466 | tok/s 2630043.36 | norm 0.3510 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18698 | train 2.908181 | val 3.315466 | tok/s 2630448.05 | norm 0.3485 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18699 | train 3.058683 | val 3.315466 | tok/s 2629325.45 | norm 0.3593 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18700 | train 2.908926 | val 3.315466 | tok/s 2626996.60 | norm 0.3513 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18701 | train 2.927486 | val 3.315466 | tok/s 2628861.63 | norm 0.3555 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18702 | train 2.939418 | val 3.315466 | tok/s 2633141.09 | norm 0.3532 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18703 | train 2.947941 | val 3.315466 | tok/s 2629995.35 | norm 0.3516 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18704 | train 2.931690 | val 3.315466 | tok/s 2521174.24 | norm 0.3708 | dt 207.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18705 | train 2.921486 | val 3.315466 | tok/s 2621546.57 | norm 0.3457 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18706 | train 2.998082 | val 3.315466 | tok/s 2630152.30 | norm 0.3443 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18707 | train 2.921316 | val 3.315466 | tok/s 2630034.57 | norm 0.3547 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18708 | train 2.801250 | val 3.315466 | tok/s 2629794.50 | norm 0.3726 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18709 | train 2.921521 | val 3.315466 | tok/s 2627382.63 | norm 0.3820 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18710 | train 2.942444 | val 3.315466 | tok/s 2626899.46 | norm 0.3815 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18711 | train 2.853705 | val 3.315466 | tok/s 2621774.04 | norm 0.3895 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18712 | train 3.123351 | val 3.315466 | tok/s 2629066.40 | norm 0.3751 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18713 | train 2.961801 | val 3.315466 | tok/s 2627935.81 | norm 0.3988 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18714 | train 3.002698 | val 3.315466 | tok/s 2616299.59 | norm 0.4137 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18715 | train 2.968795 | val 3.315466 | tok/s 2625228.69 | norm 0.3734 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18716 | train 2.919545 | val 3.315466 | tok/s 2627243.12 | norm 0.3668 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18717 | train 3.049712 | val 3.315466 | tok/s 2626212.83 | norm 0.3825 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18718 | train 2.969022 | val 3.315466 | tok/s 2627594.86 | norm 0.3481 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18719 | train 2.894605 | val 3.315466 | tok/s 2628057.59 | norm 0.3806 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18720 | train 2.943918 | val 3.315466 | tok/s 2628838.29 | norm 0.4368 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18721 | train 2.932853 | val 3.315466 | tok/s 2630108.36 | norm 0.4158 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18722 | train 2.897689 | val 3.315466 | tok/s 2631017.46 | norm 0.3823 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18723 | train 3.002172 | val 3.315466 | tok/s 2629317.30 | norm 0.3864 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18724 | train 3.035312 | val 3.315466 | tok/s 2626888.53 | norm 0.3830 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18725 | train 2.865907 | val 3.315466 | tok/s 2625388.60 | norm 0.3429 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18726 | train 3.115391 | val 3.315466 | tok/s 2631949.57 | norm 0.3520 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18727 | train 2.962620 | val 3.315466 | tok/s 2627836.36 | norm 0.3786 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18728 | train 2.921113 | val 3.315466 | tok/s 2626826.02 | norm 0.3508 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18729 | train 2.937911 | val 3.315466 | tok/s 2625691.77 | norm 0.3639 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18730 | train 2.838408 | val 3.315466 | tok/s 2625681.04 | norm 0.3761 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18731 | train 2.928789 | val 3.315466 | tok/s 2626000.92 | norm 0.3660 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18732 | train 2.862228 | val 3.315466 | tok/s 2629016.35 | norm 0.3992 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18733 | train 2.889238 | val 3.315466 | tok/s 2628477.80 | norm 0.3739 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18734 | train 3.033151 | val 3.315466 | tok/s 2625849.09 | norm 0.3656 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18735 | train 2.789032 | val 3.315466 | tok/s 2627002.60 | norm 0.3832 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18736 | train 3.094282 | val 3.315466 | tok/s 2627949.59 | norm 0.3942 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18737 | train 2.964290 | val 3.315466 | tok/s 2627884.54 | norm 0.3768 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18738 | train 3.037976 | val 3.315466 | tok/s 2627893.59 | norm 0.3799 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18739 | train 2.958475 | val 3.315466 | tok/s 2624115.59 | norm 0.3824 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18740 | train 2.934741 | val 3.315466 | tok/s 2625039.22 | norm 0.3845 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18741 | train 2.948342 | val 3.315466 | tok/s 2626643.43 | norm 0.4041 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18742 | train 2.874834 | val 3.315466 | tok/s 2627072.40 | norm 0.5090 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18743 | train 3.171183 | val 3.315466 | tok/s 2625366.29 | norm 0.4574 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18744 | train 3.031447 | val 3.315466 | tok/s 2628675.63 | norm 0.4031 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18745 | train 2.921209 | val 3.315466 | tok/s 2629519.01 | norm 0.3435 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18746 | train 3.083140 | val 3.315466 | tok/s 2631233.78 | norm 0.3590 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18747 | train 2.995714 | val 3.315466 | tok/s 2634312.46 | norm 0.3767 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18748 | train 2.924788 | val 3.315466 | tok/s 2634771.75 | norm 0.4009 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18749 | train 2.963795 | val 3.315466 | tok/s 2627796.01 | norm 0.3788 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 18750 | val_loss 3.136270 step 18750 | train 2.904478 | val 3.136270 | tok/s 2605839.88 | norm 0.3485 | dt 201.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18751 | train 2.912917 | val 3.136270 | tok/s 2628149.82 | norm 0.3510 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18752 | train 2.878640 | val 3.136270 | tok/s 2630559.31 | norm 0.3737 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18753 | train 2.927983 | val 3.136270 | tok/s 2622410.59 | norm 0.3953 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18754 | train 3.219348 | val 3.136270 | tok/s 2627038.65 | norm 0.4278 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18755 | train 2.941834 | val 3.136270 | tok/s 2626063.39 | norm 0.3438 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18756 | train 2.974830 | val 3.136270 | tok/s 2629485.01 | norm 0.3683 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18757 | train 2.933356 | val 3.136270 | tok/s 2627889.41 | norm 0.3789 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18758 | train 2.932567 | val 3.136270 | tok/s 2623485.70 | norm 0.3622 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18759 | train 2.982741 | val 3.136270 | tok/s 2629765.85 | norm 0.3979 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18760 | train 2.949399 | val 3.136270 | tok/s 2627768.59 | norm 0.3749 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18761 | train 2.966548 | val 3.136270 | tok/s 2609654.28 | norm 0.3608 | dt 200.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18762 | train 2.978287 | val 3.136270 | tok/s 2623666.99 | norm 0.4072 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18763 | train 2.935160 | val 3.136270 | tok/s 2625566.96 | norm 0.3630 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18764 | train 3.043777 | val 3.136270 | tok/s 2632337.72 | norm 0.3758 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18765 | train 2.885656 | val 3.136270 | tok/s 2630104.47 | norm 0.3743 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18766 | train 2.863059 | val 3.136270 | tok/s 2631216.68 | norm 0.3889 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18767 | train 2.878822 | val 3.136270 | tok/s 2625189.16 | norm 0.3770 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18768 | train 3.010093 | val 3.136270 | tok/s 2628307.17 | norm 0.3480 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18769 | train 2.958743 | val 3.136270 | tok/s 2629516.48 | norm 0.3484 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18770 | train 2.911819 | val 3.136270 | tok/s 2625387.32 | norm 0.3374 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18771 | train 2.968243 | val 3.136270 | tok/s 2624563.49 | norm 0.3596 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18772 | train 2.904624 | val 3.136270 | tok/s 2621461.56 | norm 0.3621 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18773 | train 3.053268 | val 3.136270 | tok/s 2611797.39 | norm 0.3580 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18774 | train 3.046506 | val 3.136270 | tok/s 2628900.50 | norm 0.3896 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18775 | train 2.821028 | val 3.136270 | tok/s 2627795.86 | norm 0.3904 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18776 | train 3.018087 | val 3.136270 | tok/s 2623420.81 | norm 0.3491 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18777 | train 2.969131 | val 3.136270 | tok/s 2608285.89 | norm 0.3175 | dt 201.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18778 | train 2.920349 | val 3.136270 | tok/s 2627246.13 | norm 0.3447 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18779 | train 3.317398 | val 3.136270 | tok/s 2630260.00 | norm 0.3955 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18780 | train 2.931392 | val 3.136270 | tok/s 2630053.73 | norm 0.4766 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18781 | train 3.009898 | val 3.136270 | tok/s 2633884.39 | norm 0.3549 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18782 | train 2.931463 | val 3.136270 | tok/s 2624251.07 | norm 0.3648 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18783 | train 2.911715 | val 3.136270 | tok/s 2585000.50 | norm 0.3834 | dt 202.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18784 | train 2.990065 | val 3.136270 | tok/s 2629295.53 | norm 0.3830 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18785 | train 2.924189 | val 3.136270 | tok/s 2629404.91 | norm 0.3694 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18786 | train 2.989385 | val 3.136270 | tok/s 2630656.56 | norm 0.3562 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18787 | train 3.025358 | val 3.136270 | tok/s 2624038.52 | norm 0.3800 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18788 | train 3.057980 | val 3.136270 | tok/s 2629724.71 | norm 0.4206 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18789 | train 2.890074 | val 3.136270 | tok/s 2632017.16 | norm 0.4132 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18790 | train 2.825073 | val 3.136270 | tok/s 2628704.08 | norm 0.3780 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18791 | train 3.030961 | val 3.136270 | tok/s 2628572.16 | norm 0.3904 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18792 | train 2.973054 | val 3.136270 | tok/s 2630926.26 | norm 0.3841 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18793 | train 2.996179 | val 3.136270 | tok/s 2625243.89 | norm 0.3549 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18794 | train 2.949514 | val 3.136270 | tok/s 2629023.40 | norm 0.3408 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18795 | train 3.011980 | val 3.136270 | tok/s 2629929.89 | norm 0.3585 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18796 | train 3.051856 | val 3.136270 | tok/s 2630897.13 | norm 0.4896 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18797 | train 2.876563 | val 3.136270 | tok/s 2625196.84 | norm 0.3643 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18798 | train 2.894467 | val 3.136270 | tok/s 2626045.31 | norm 0.3311 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18799 | train 3.111335 | val 3.136270 | tok/s 2630315.57 | norm 0.4608 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18800 | train 2.870709 | val 3.136270 | tok/s 2630174.76 | norm 0.3733 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18801 | train 2.928804 | val 3.136270 | tok/s 2625211.60 | norm 0.3128 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18802 | train 2.750939 | val 3.136270 | tok/s 2626751.92 | norm 0.3822 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18803 | train 2.883839 | val 3.136270 | tok/s 2629155.47 | norm 0.4131 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18804 | train 2.909625 | val 3.136270 | tok/s 2629061.44 | norm 0.3656 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18805 | train 2.852215 | val 3.136270 | tok/s 2618590.71 | norm 0.3529 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18806 | train 2.931028 | val 3.136270 | tok/s 2628939.80 | norm 0.3842 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18807 | train 2.991515 | val 3.136270 | tok/s 2622491.96 | norm 0.3641 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18808 | train 2.919832 | val 3.136270 | tok/s 2623186.11 | norm 0.3375 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18809 | train 2.840118 | val 3.136270 | tok/s 2625146.23 | norm 0.3793 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18810 | train 2.884705 | val 3.136270 | tok/s 2626915.67 | norm 0.3848 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18811 | train 2.876629 | val 3.136270 | tok/s 2626208.75 | norm 0.4415 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18812 | train 2.962107 | val 3.136270 | tok/s 2629626.74 | norm 0.3962 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18813 | train 2.998657 | val 3.136270 | tok/s 2631581.66 | norm 0.3545 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18814 | train 3.033221 | val 3.136270 | tok/s 2632096.52 | norm 0.3629 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18815 | train 2.993805 | val 3.136270 | tok/s 2626478.45 | norm 0.3663 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18816 | train 2.966957 | val 3.136270 | tok/s 2629234.02 | norm 0.3905 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18817 | train 2.942544 | val 3.136270 | tok/s 2629613.87 | norm 0.3754 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18818 | train 3.044750 | val 3.136270 | tok/s 2632084.04 | norm 0.4080 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18819 | train 2.949053 | val 3.136270 | tok/s 2631661.87 | norm 0.3751 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18820 | train 2.974536 | val 3.136270 | tok/s 2534773.83 | norm 0.3921 | dt 206.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18821 | train 2.949322 | val 3.136270 | tok/s 2630834.82 | norm 0.4049 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18822 | train 2.975992 | val 3.136270 | tok/s 2628339.06 | norm 0.3884 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18823 | train 2.835693 | val 3.136270 | tok/s 2623718.95 | norm 0.3474 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18824 | train 2.886339 | val 3.136270 | tok/s 2617208.26 | norm 0.3544 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18825 | train 2.991165 | val 3.136270 | tok/s 2622271.90 | norm 0.3606 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18826 | train 2.955194 | val 3.136270 | tok/s 2631935.40 | norm 0.3503 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18827 | train 2.867046 | val 3.136270 | tok/s 2626313.61 | norm 0.3527 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18828 | train 2.913914 | val 3.136270 | tok/s 2630874.82 | norm 0.3760 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18829 | train 2.972555 | val 3.136270 | tok/s 2627346.88 | norm 0.3579 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18830 | train 3.013708 | val 3.136270 | tok/s 2625609.93 | norm 0.3642 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18831 | train 2.959792 | val 3.136270 | tok/s 2628486.50 | norm 0.3809 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18832 | train 2.964658 | val 3.136270 | tok/s 2629716.69 | norm 0.3336 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18833 | train 2.847708 | val 3.136270 | tok/s 2627587.60 | norm 0.3677 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18834 | train 2.860807 | val 3.136270 | tok/s 2625863.81 | norm 0.3616 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18835 | train 2.949282 | val 3.136270 | tok/s 2624206.90 | norm 0.3226 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18836 | train 3.027738 | val 3.136270 | tok/s 2628388.75 | norm 0.3663 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18837 | train 2.951528 | val 3.136270 | tok/s 2626385.26 | norm 0.3594 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18838 | train 2.900815 | val 3.136270 | tok/s 2628981.83 | norm 0.3808 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18839 | train 2.887291 | val 3.136270 | tok/s 2628876.36 | norm 0.4080 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18840 | train 2.945648 | val 3.136270 | tok/s 2629131.79 | norm 0.4171 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18841 | train 2.991883 | val 3.136270 | tok/s 2630978.20 | norm 0.3957 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18842 | train 3.284625 | val 3.136270 | tok/s 2630945.75 | norm 0.4697 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18843 | train 2.966022 | val 3.136270 | tok/s 2626725.46 | norm 0.3280 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18844 | train 2.930210 | val 3.136270 | tok/s 2628351.18 | norm 0.3360 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18845 | train 2.962986 | val 3.136270 | tok/s 2628284.16 | norm 0.3782 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18846 | train 3.043957 | val 3.136270 | tok/s 2629570.78 | norm 0.3757 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18847 | train 2.896777 | val 3.136270 | tok/s 2627141.96 | norm 0.4002 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18848 | train 2.901746 | val 3.136270 | tok/s 2628361.51 | norm 0.3734 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18849 | train 2.937849 | val 3.136270 | tok/s 2625824.09 | norm 0.3535 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18850 | train 2.919365 | val 3.136270 | tok/s 2625182.18 | norm 0.3393 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18851 | train 2.923333 | val 3.136270 | tok/s 2630235.78 | norm 0.3691 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18852 | train 2.885430 | val 3.136270 | tok/s 2628617.86 | norm 0.3648 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18853 | train 2.954894 | val 3.136270 | tok/s 2629749.56 | norm 0.3490 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18854 | train 2.953906 | val 3.136270 | tok/s 2623383.76 | norm 0.3678 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18855 | train 3.134240 | val 3.136270 | tok/s 2622007.07 | norm 0.3648 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18856 | train 2.991431 | val 3.136270 | tok/s 2620671.41 | norm 0.3797 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18857 | train 2.895608 | val 3.136270 | tok/s 2628584.90 | norm 0.3777 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18858 | train 2.910925 | val 3.136270 | tok/s 2630922.75 | norm 0.3786 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18859 | train 2.903521 | val 3.136270 | tok/s 2628593.53 | norm 0.3700 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18860 | train 2.824090 | val 3.136270 | tok/s 2626242.89 | norm 0.3776 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18861 | train 3.006787 | val 3.136270 | tok/s 2630209.10 | norm 0.3765 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18862 | train 2.958605 | val 3.136270 | tok/s 2626258.38 | norm 0.3341 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18863 | train 2.909370 | val 3.136270 | tok/s 2628879.72 | norm 0.3918 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18864 | train 2.911387 | val 3.136270 | tok/s 2628583.33 | norm 0.3954 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18865 | train 2.920305 | val 3.136270 | tok/s 2628727.16 | norm 0.3533 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18866 | train 3.016458 | val 3.136270 | tok/s 2629975.35 | norm 0.3776 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18867 | train 2.938714 | val 3.136270 | tok/s 2597524.14 | norm 0.3710 | dt 201.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18868 | train 2.865472 | val 3.136270 | tok/s 2629378.58 | norm 0.3773 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18869 | train 2.904644 | val 3.136270 | tok/s 2626245.74 | norm 0.3871 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18870 | train 2.935327 | val 3.136270 | tok/s 2625693.89 | norm 0.3745 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18871 | train 3.066773 | val 3.136270 | tok/s 2630526.63 | norm 0.3689 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18872 | train 2.943704 | val 3.136270 | tok/s 2590993.25 | norm 0.3552 | dt 202.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18873 | train 3.039392 | val 3.136270 | tok/s 2631246.81 | norm 0.3464 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18874 | train 2.945619 | val 3.136270 | tok/s 2628979.68 | norm 0.3268 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18875 | train 3.072125 | val 3.136270 | tok/s 2632236.94 | norm 0.3204 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18876 | train 2.892592 | val 3.136270 | tok/s 2625706.51 | norm 0.3395 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18877 | train 2.944606 | val 3.136270 | tok/s 2565270.28 | norm 0.3277 | dt 204.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18878 | train 2.907809 | val 3.136270 | tok/s 2632564.53 | norm 0.3501 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18879 | train 2.847896 | val 3.136270 | tok/s 2628262.40 | norm 0.3531 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18880 | train 2.969875 | val 3.136270 | tok/s 2631901.54 | norm 0.3370 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18881 | train 3.066339 | val 3.136270 | tok/s 2630606.89 | norm 0.3627 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18882 | train 2.919273 | val 3.136270 | tok/s 2629338.28 | norm 0.4114 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18883 | train 3.042033 | val 3.136270 | tok/s 2628635.28 | norm 0.3773 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18884 | train 2.987605 | val 3.136270 | tok/s 2627746.82 | norm 0.3931 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18885 | train 2.994176 | val 3.136270 | tok/s 2622745.68 | norm 0.4213 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18886 | train 2.920596 | val 3.136270 | tok/s 2630752.69 | norm 0.3936 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18887 | train 2.918694 | val 3.136270 | tok/s 2626196.46 | norm 0.4249 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18888 | train 2.955248 | val 3.136270 | tok/s 2632069.52 | norm 0.4136 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18889 | train 3.096212 | val 3.136270 | tok/s 2627114.49 | norm 0.3704 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18890 | train 2.909035 | val 3.136270 | tok/s 2628995.71 | norm 0.3725 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18891 | train 2.954470 | val 3.136270 | tok/s 2624345.75 | norm 0.4002 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18892 | train 2.942315 | val 3.136270 | tok/s 2625495.91 | norm 0.3920 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18893 | train 2.950519 | val 3.136270 | tok/s 2628382.06 | norm 0.3539 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18894 | train 2.903353 | val 3.136270 | tok/s 2629322.18 | norm 0.3386 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18895 | train 2.947074 | val 3.136270 | tok/s 2631614.04 | norm 0.3851 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18896 | train 3.044761 | val 3.136270 | tok/s 2632065.48 | norm 0.3601 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18897 | train 3.035037 | val 3.136270 | tok/s 2631530.99 | norm 0.3628 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18898 | train 2.991295 | val 3.136270 | tok/s 2631850.44 | norm 0.3599 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18899 | train 2.987153 | val 3.136270 | tok/s 2626639.35 | norm 0.3638 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18900 | train 2.824123 | val 3.136270 | tok/s 2629287.07 | norm 0.3910 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18901 | train 2.900871 | val 3.136270 | tok/s 2624566.34 | norm 0.3842 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18902 | train 2.854500 | val 3.136270 | tok/s 2626559.04 | norm 0.3804 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18903 | train 2.968706 | val 3.136270 | tok/s 2630887.90 | norm 0.3959 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18904 | train 2.945039 | val 3.136270 | tok/s 2626604.80 | norm 0.3975 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18905 | train 3.046397 | val 3.136270 | tok/s 2624293.43 | norm 0.4128 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18906 | train 2.945463 | val 3.136270 | tok/s 2627751.40 | norm 0.3441 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18907 | train 2.939185 | val 3.136270 | tok/s 2628135.82 | norm 0.3423 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18908 | train 2.962018 | val 3.136270 | tok/s 2630380.15 | norm 0.3796 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18909 | train 2.861490 | val 3.136270 | tok/s 2630548.09 | norm 0.4025 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18910 | train 3.029957 | val 3.136270 | tok/s 2629022.35 | norm 0.3930 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18911 | train 2.931583 | val 3.136270 | tok/s 2623147.50 | norm 0.3855 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18912 | train 2.916161 | val 3.136270 | tok/s 2626520.16 | norm 0.3456 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18913 | train 3.018227 | val 3.136270 | tok/s 2576292.40 | norm 0.3352 | dt 203.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18914 | train 2.922107 | val 3.136270 | tok/s 2627242.87 | norm 0.3453 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18915 | train 3.037423 | val 3.136270 | tok/s 2628575.69 | norm 0.3831 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18916 | train 2.942256 | val 3.136270 | tok/s 2627661.95 | norm 0.4055 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18917 | train 2.950263 | val 3.136270 | tok/s 2630157.30 | norm 0.3720 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18918 | train 2.945360 | val 3.136270 | tok/s 2632243.62 | norm 0.3626 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18919 | train 2.887308 | val 3.136270 | tok/s 2626202.26 | norm 0.3857 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18920 | train 2.884725 | val 3.136270 | tok/s 2627965.01 | norm 0.3844 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18921 | train 2.840482 | val 3.136270 | tok/s 2628468.51 | norm 0.3850 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18922 | train 3.075301 | val 3.136270 | tok/s 2629696.21 | norm 0.3917 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18923 | train 3.022336 | val 3.136270 | tok/s 2626238.76 | norm 0.3622 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18924 | train 2.915709 | val 3.136270 | tok/s 2626219.73 | norm 0.3686 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18925 | train 2.986497 | val 3.136270 | tok/s 2626229.99 | norm 0.3788 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18926 | train 2.981462 | val 3.136270 | tok/s 2626378.51 | norm 0.3957 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18927 | train 2.973368 | val 3.136270 | tok/s 2627513.23 | norm 0.3828 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18928 | train 2.873063 | val 3.136270 | tok/s 2463162.88 | norm 0.3947 | dt 212.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18929 | train 2.934988 | val 3.136270 | tok/s 2627182.01 | norm 0.3768 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18930 | train 2.911264 | val 3.136270 | tok/s 2631260.01 | norm 0.3516 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18931 | train 2.933307 | val 3.136270 | tok/s 2627100.27 | norm 0.4081 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18932 | train 2.938174 | val 3.136270 | tok/s 2629476.26 | norm 0.4006 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18933 | train 2.897274 | val 3.136270 | tok/s 2628508.41 | norm 0.3801 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18934 | train 3.014155 | val 3.136270 | tok/s 2626368.09 | norm 0.3782 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18935 | train 3.196289 | val 3.136270 | tok/s 2630521.78 | norm 0.4030 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18936 | train 2.943142 | val 3.136270 | tok/s 2629003.96 | norm 0.3318 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18937 | train 2.859327 | val 3.136270 | tok/s 2629995.11 | norm 0.3141 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18938 | train 2.924634 | val 3.136270 | tok/s 2621351.52 | norm 0.3623 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18939 | train 3.103224 | val 3.136270 | tok/s 2623817.10 | norm 0.4148 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18940 | train 2.927909 | val 3.136270 | tok/s 2630261.50 | norm 0.3656 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18941 | train 3.058122 | val 3.136270 | tok/s 2631091.46 | norm 0.3467 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18942 | train 2.923209 | val 3.136270 | tok/s 2633382.85 | norm 0.3440 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18943 | train 2.863448 | val 3.136270 | tok/s 2627216.29 | norm 0.3503 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18944 | train 2.892827 | val 3.136270 | tok/s 2627194.98 | norm 0.3503 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18945 | train 3.060154 | val 3.136270 | tok/s 2632098.71 | norm 0.3552 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18946 | train 2.840972 | val 3.136270 | tok/s 2629361.75 | norm 0.3878 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18947 | train 2.984198 | val 3.136270 | tok/s 2628203.71 | norm 0.3896 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18948 | train 2.853217 | val 3.136270 | tok/s 2629830.55 | norm 0.4092 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18949 | train 2.933472 | val 3.136270 | tok/s 2622106.21 | norm 0.3763 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18950 | train 2.857309 | val 3.136270 | tok/s 2628399.45 | norm 0.3758 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18951 | train 2.882426 | val 3.136270 | tok/s 2630981.08 | norm 0.3733 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18952 | train 2.919436 | val 3.136270 | tok/s 2631083.17 | norm 0.3448 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18953 | train 2.935195 | val 3.136270 | tok/s 2628453.74 | norm 0.3563 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18954 | train 2.881532 | val 3.136270 | tok/s 2625941.66 | norm 0.3637 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18955 | train 2.936530 | val 3.136270 | tok/s 2622840.15 | norm 0.3713 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18956 | train 3.045016 | val 3.136270 | tok/s 2629859.68 | norm 0.3811 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18957 | train 3.020580 | val 3.136270 | tok/s 2628959.23 | norm 0.4011 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18958 | train 2.986054 | val 3.136270 | tok/s 2625085.61 | norm 0.3925 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18959 | train 2.920668 | val 3.136270 | tok/s 2630736.71 | norm 0.3928 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18960 | train 2.881499 | val 3.136270 | tok/s 2631963.49 | norm 0.3726 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18961 | train 3.014891 | val 3.136270 | tok/s 2612561.15 | norm 0.3795 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18962 | train 3.066669 | val 3.136270 | tok/s 2629619.59 | norm 0.3844 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18963 | train 2.881459 | val 3.136270 | tok/s 2624642.46 | norm 0.3653 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18964 | train 2.945018 | val 3.136270 | tok/s 2629680.47 | norm 0.3668 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18965 | train 2.926155 | val 3.136270 | tok/s 2628981.71 | norm 0.3759 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18966 | train 2.998370 | val 3.136270 | tok/s 2623980.04 | norm 0.4062 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18967 | train 2.971195 | val 3.136270 | tok/s 2630984.73 | norm 0.3711 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18968 | train 2.923903 | val 3.136270 | tok/s 2626046.48 | norm 0.3571 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18969 | train 2.926265 | val 3.136270 | tok/s 2626233.41 | norm 0.3712 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18970 | train 3.019755 | val 3.136270 | tok/s 2628913.26 | norm 0.3788 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18971 | train 2.919542 | val 3.136270 | tok/s 2628462.07 | norm 0.3873 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18972 | train 3.022222 | val 3.136270 | tok/s 2629005.91 | norm 0.3918 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18973 | train 2.874136 | val 3.136270 | tok/s 2632127.29 | norm 0.3489 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18974 | train 2.947628 | val 3.136270 | tok/s 2629348.05 | norm 0.3143 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18975 | train 2.993131 | val 3.136270 | tok/s 2634510.05 | norm 0.3438 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18976 | train 2.951253 | val 3.136270 | tok/s 2627039.82 | norm 0.3494 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18977 | train 3.034912 | val 3.136270 | tok/s 2633022.15 | norm 0.3638 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18978 | train 2.887345 | val 3.136270 | tok/s 2631960.76 | norm 0.3737 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18979 | train 2.922383 | val 3.136270 | tok/s 2629643.59 | norm 0.3625 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18980 | train 2.943145 | val 3.136270 | tok/s 2623658.71 | norm 0.3681 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18981 | train 2.903010 | val 3.136270 | tok/s 2628051.63 | norm 0.3481 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18982 | train 2.916827 | val 3.136270 | tok/s 2629502.48 | norm 0.3689 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18983 | train 2.963022 | val 3.136270 | tok/s 2628146.31 | norm 0.3719 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18984 | train 2.921010 | val 3.136270 | tok/s 2627612.87 | norm 0.3862 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18985 | train 3.004062 | val 3.136270 | tok/s 2599652.33 | norm 0.3708 | dt 201.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18986 | train 2.924753 | val 3.136270 | tok/s 2623899.13 | norm 0.3649 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18987 | train 2.847948 | val 3.136270 | tok/s 2628589.54 | norm 0.3714 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18988 | train 2.898515 | val 3.136270 | tok/s 2627571.37 | norm 0.3762 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18989 | train 2.834546 | val 3.136270 | tok/s 2630829.23 | norm 0.3818 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18990 | train 3.021934 | val 3.136270 | tok/s 2629043.94 | norm 0.3538 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18991 | train 2.903828 | val 3.136270 | tok/s 2631822.10 | norm 0.3597 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18992 | train 2.969759 | val 3.136270 | tok/s 2631018.39 | norm 0.3851 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18993 | train 2.878113 | val 3.136270 | tok/s 2630943.18 | norm 0.3803 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18994 | train 3.026995 | val 3.136270 | tok/s 2505087.55 | norm 0.3944 | dt 209.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18995 | train 2.896792 | val 3.136270 | tok/s 2624309.38 | norm 0.3841 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18996 | train 2.946816 | val 3.136270 | tok/s 2623301.44 | norm 0.3664 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18997 | train 2.889020 | val 3.136270 | tok/s 2625490.09 | norm 0.3962 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18998 | train 2.828660 | val 3.136270 | tok/s 2623248.99 | norm 0.3677 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 18999 | train 3.037342 | val 3.136270 | tok/s 2629217.94 | norm 0.3728 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 19000 | val_loss 3.228252 step 19000 | train 2.975568 | val 3.228252 | tok/s 2612582.38 | norm 0.3990 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19001 | train 2.921456 | val 3.228252 | tok/s 2622703.49 | norm 0.3924 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19002 | train 2.889381 | val 3.228252 | tok/s 2626682.99 | norm 0.3866 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19003 | train 3.007204 | val 3.228252 | tok/s 2549796.70 | norm 0.3626 | dt 205.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19004 | train 3.012380 | val 3.228252 | tok/s 2627947.89 | norm 0.3532 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19005 | train 2.763035 | val 3.228252 | tok/s 2628230.86 | norm 0.3826 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19006 | train 2.960033 | val 3.228252 | tok/s 2626069.35 | norm 0.3993 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19007 | train 2.885904 | val 3.228252 | tok/s 2626171.14 | norm 0.3982 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19008 | train 2.798645 | val 3.228252 | tok/s 2630509.26 | norm 0.3781 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19009 | train 2.878267 | val 3.228252 | tok/s 2629320.81 | norm 0.3939 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19010 | train 2.878783 | val 3.228252 | tok/s 2628071.85 | norm 0.4199 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19011 | train 2.920043 | val 3.228252 | tok/s 2621249.95 | norm 0.3708 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19012 | train 2.978839 | val 3.228252 | tok/s 2627140.95 | norm 0.3603 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19013 | train 3.119642 | val 3.228252 | tok/s 2629751.63 | norm 0.4507 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19014 | train 3.064962 | val 3.228252 | tok/s 2627243.09 | norm 0.3958 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19015 | train 2.923421 | val 3.228252 | tok/s 2628799.27 | norm 0.3164 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19016 | train 2.819689 | val 3.228252 | tok/s 2627234.00 | norm 0.3441 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19017 | train 3.018952 | val 3.228252 | tok/s 2628181.97 | norm 0.3447 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19018 | train 3.019425 | val 3.228252 | tok/s 2630818.09 | norm 0.4054 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19019 | train 2.964921 | val 3.228252 | tok/s 2629774.98 | norm 0.4405 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19020 | train 2.895440 | val 3.228252 | tok/s 2629307.08 | norm 0.3559 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19021 | train 2.924349 | val 3.228252 | tok/s 2623461.27 | norm 0.3830 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19022 | train 2.922042 | val 3.228252 | tok/s 2625729.90 | norm 0.3736 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19023 | train 2.915808 | val 3.228252 | tok/s 2631543.06 | norm 0.3679 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19024 | train 2.930897 | val 3.228252 | tok/s 2540588.07 | norm 0.3546 | dt 206.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19025 | train 2.917251 | val 3.228252 | tok/s 2628795.41 | norm 0.3953 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19026 | train 3.023349 | val 3.228252 | tok/s 2624100.21 | norm 0.4452 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19027 | train 2.962709 | val 3.228252 | tok/s 2625936.00 | norm 0.3694 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19028 | train 3.046371 | val 3.228252 | tok/s 2632885.76 | norm 0.4006 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19029 | train 2.898218 | val 3.228252 | tok/s 2624185.90 | norm 0.4002 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19030 | train 2.903923 | val 3.228252 | tok/s 2628555.53 | norm 0.3551 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19031 | train 2.939940 | val 3.228252 | tok/s 2627851.31 | norm 0.3924 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19032 | train 3.008304 | val 3.228252 | tok/s 2627045.18 | norm 0.3700 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19033 | train 2.991950 | val 3.228252 | tok/s 2627982.16 | norm 0.3659 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19034 | train 2.899882 | val 3.228252 | tok/s 2628731.91 | norm 0.3587 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19035 | train 2.964721 | val 3.228252 | tok/s 2630013.91 | norm 0.3741 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19036 | train 2.949806 | val 3.228252 | tok/s 2626936.25 | norm 0.3701 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19037 | train 2.925631 | val 3.228252 | tok/s 2633546.17 | norm 0.3542 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19038 | train 2.992474 | val 3.228252 | tok/s 2547650.03 | norm 0.3722 | dt 205.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19039 | train 3.105491 | val 3.228252 | tok/s 2520288.64 | norm 0.3852 | dt 208.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19040 | train 2.929392 | val 3.228252 | tok/s 2612500.08 | norm 0.3453 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19041 | train 2.903541 | val 3.228252 | tok/s 2631860.14 | norm 0.3274 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19042 | train 2.956138 | val 3.228252 | tok/s 2629358.60 | norm 0.3822 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19043 | train 2.968435 | val 3.228252 | tok/s 2624141.96 | norm 0.3953 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19044 | train 2.871359 | val 3.228252 | tok/s 2625149.76 | norm 0.3870 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19045 | train 2.869866 | val 3.228252 | tok/s 2632445.00 | norm 0.3655 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19046 | train 2.865242 | val 3.228252 | tok/s 2519494.89 | norm 0.3484 | dt 208.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19047 | train 2.839079 | val 3.228252 | tok/s 2571248.27 | norm 0.3551 | dt 203.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19048 | train 2.888472 | val 3.228252 | tok/s 2621848.69 | norm 0.3672 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19049 | train 3.013205 | val 3.228252 | tok/s 2627616.09 | norm 0.3559 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19050 | train 2.819969 | val 3.228252 | tok/s 2626415.19 | norm 0.3430 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19051 | train 2.955135 | val 3.228252 | tok/s 2626413.21 | norm 0.3585 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19052 | train 2.941214 | val 3.228252 | tok/s 2629412.35 | norm 0.3527 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19053 | train 2.931321 | val 3.228252 | tok/s 2627375.79 | norm 0.3565 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19054 | train 3.045552 | val 3.228252 | tok/s 2627577.85 | norm 0.3925 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19055 | train 3.144068 | val 3.228252 | tok/s 2629185.45 | norm 0.4141 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19056 | train 2.970572 | val 3.228252 | tok/s 2626534.66 | norm 0.3585 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19057 | train 2.962894 | val 3.228252 | tok/s 2624634.05 | norm 0.3537 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19058 | train 2.901552 | val 3.228252 | tok/s 2626261.30 | norm 0.4000 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19059 | train 3.090213 | val 3.228252 | tok/s 2624558.37 | norm 0.4008 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19060 | train 2.874468 | val 3.228252 | tok/s 2627873.76 | norm 0.3386 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19061 | train 2.991395 | val 3.228252 | tok/s 2628801.34 | norm 0.3590 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19062 | train 2.923626 | val 3.228252 | tok/s 2631000.73 | norm 0.3815 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19063 | train 2.878862 | val 3.228252 | tok/s 2628469.80 | norm 0.3837 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19064 | train 2.801877 | val 3.228252 | tok/s 2630463.05 | norm 0.4248 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19065 | train 2.920904 | val 3.228252 | tok/s 2630052.21 | norm 0.3602 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19066 | train 2.897134 | val 3.228252 | tok/s 2625022.30 | norm 0.3326 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19067 | train 2.991189 | val 3.228252 | tok/s 2630463.46 | norm 0.3347 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19068 | train 3.001657 | val 3.228252 | tok/s 2627025.92 | norm 0.3668 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19069 | train 2.869481 | val 3.228252 | tok/s 2629975.56 | norm 0.3746 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19070 | train 2.835300 | val 3.228252 | tok/s 2628998.61 | norm 0.3568 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19071 | train 2.883140 | val 3.228252 | tok/s 2626307.49 | norm 0.3549 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19072 | train 2.879355 | val 3.228252 | tok/s 2627778.29 | norm 0.3313 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19073 | train 2.930012 | val 3.228252 | tok/s 2628888.86 | norm 0.3666 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19074 | train 2.921826 | val 3.228252 | tok/s 2632710.92 | norm 0.3686 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19075 | train 3.079277 | val 3.228252 | tok/s 2632634.53 | norm 0.3722 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19076 | train 3.021378 | val 3.228252 | tok/s 2625844.21 | norm 0.3663 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19077 | train 2.941080 | val 3.228252 | tok/s 2630421.39 | norm 0.3701 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19078 | train 3.155336 | val 3.228252 | tok/s 2626558.93 | norm 0.3556 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19079 | train 3.085371 | val 3.228252 | tok/s 2628542.07 | norm 0.3553 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19080 | train 2.922350 | val 3.228252 | tok/s 2631231.72 | norm 0.3464 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19081 | train 2.926049 | val 3.228252 | tok/s 2629989.45 | norm 0.3489 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19082 | train 2.959614 | val 3.228252 | tok/s 2627982.28 | norm 0.3815 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19083 | train 2.986762 | val 3.228252 | tok/s 2628237.99 | norm 0.3889 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19084 | train 3.069911 | val 3.228252 | tok/s 2627610.59 | norm 0.4076 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19085 | train 3.008237 | val 3.228252 | tok/s 2577074.55 | norm 0.4244 | dt 203.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19086 | train 3.201491 | val 3.228252 | tok/s 2580257.78 | norm 0.4242 | dt 203.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19087 | train 2.878856 | val 3.228252 | tok/s 2630974.97 | norm 0.3845 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19088 | train 2.888421 | val 3.228252 | tok/s 2625050.00 | norm 0.4059 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19089 | train 2.990703 | val 3.228252 | tok/s 2625382.88 | norm 0.4109 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19090 | train 2.905099 | val 3.228252 | tok/s 2623108.24 | norm 0.4715 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19091 | train 3.003048 | val 3.228252 | tok/s 2623594.74 | norm 0.3965 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19092 | train 2.859976 | val 3.228252 | tok/s 2630965.92 | norm 0.3215 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19093 | train 3.082911 | val 3.228252 | tok/s 2494857.07 | norm 0.3565 | dt 210.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19094 | train 2.928930 | val 3.228252 | tok/s 2632716.60 | norm 0.3551 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19095 | train 2.926555 | val 3.228252 | tok/s 2635050.31 | norm 0.3503 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19096 | train 3.129177 | val 3.228252 | tok/s 2631669.09 | norm 0.3761 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19097 | train 2.821923 | val 3.228252 | tok/s 2630270.91 | norm 0.3681 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19098 | train 2.995032 | val 3.228252 | tok/s 2629781.68 | norm 0.3729 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19099 | train 2.858300 | val 3.228252 | tok/s 2623515.63 | norm 0.3587 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19100 | train 2.997824 | val 3.228252 | tok/s 2627544.46 | norm 0.3484 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19101 | train 2.891786 | val 3.228252 | tok/s 2630899.90 | norm 0.3513 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19102 | train 2.999283 | val 3.228252 | tok/s 2632736.06 | norm 0.3841 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19103 | train 2.920155 | val 3.228252 | tok/s 2630997.94 | norm 0.3742 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19104 | train 2.890284 | val 3.228252 | tok/s 2631781.18 | norm 0.3593 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19105 | train 3.212302 | val 3.228252 | tok/s 2620824.94 | norm 0.3850 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19106 | train 3.006618 | val 3.228252 | tok/s 2624262.49 | norm 0.3917 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19107 | train 2.928627 | val 3.228252 | tok/s 2630522.27 | norm 0.3789 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19108 | train 3.018684 | val 3.228252 | tok/s 2624628.53 | norm 0.3859 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19109 | train 3.008976 | val 3.228252 | tok/s 2627792.33 | norm 0.3997 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19110 | train 3.035809 | val 3.228252 | tok/s 2622715.83 | norm 0.4077 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19111 | train 2.794016 | val 3.228252 | tok/s 2624942.95 | norm 0.3777 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19112 | train 3.005546 | val 3.228252 | tok/s 2631029.70 | norm 0.3576 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19113 | train 2.919060 | val 3.228252 | tok/s 2631102.96 | norm 0.3550 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19114 | train 2.925775 | val 3.228252 | tok/s 2627673.05 | norm 0.3967 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19115 | train 2.892674 | val 3.228252 | tok/s 2623724.69 | norm 0.3941 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19116 | train 2.948342 | val 3.228252 | tok/s 2621077.32 | norm 0.3424 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19117 | train 2.986006 | val 3.228252 | tok/s 2626131.26 | norm 0.3695 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19118 | train 2.851904 | val 3.228252 | tok/s 2625973.77 | norm 0.3799 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19119 | train 3.001738 | val 3.228252 | tok/s 2628110.78 | norm 0.4047 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19120 | train 2.902380 | val 3.228252 | tok/s 2628464.06 | norm 0.3859 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19121 | train 2.915918 | val 3.228252 | tok/s 2629728.42 | norm 0.3594 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19122 | train 2.921691 | val 3.228252 | tok/s 2629951.69 | norm 0.3800 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19123 | train 3.001120 | val 3.228252 | tok/s 2624563.11 | norm 0.3975 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19124 | train 2.931388 | val 3.228252 | tok/s 2632512.12 | norm 0.3640 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19125 | train 3.041263 | val 3.228252 | tok/s 2628319.44 | norm 0.3905 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19126 | train 2.830826 | val 3.228252 | tok/s 2628393.11 | norm 0.3861 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19127 | train 3.072039 | val 3.228252 | tok/s 2622836.52 | norm 0.4074 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19128 | train 2.960304 | val 3.228252 | tok/s 2618823.02 | norm 0.3828 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19129 | train 2.852406 | val 3.228252 | tok/s 2628314.84 | norm 0.3449 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19130 | train 2.775644 | val 3.228252 | tok/s 2628160.69 | norm 0.3798 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19131 | train 2.939011 | val 3.228252 | tok/s 2627220.46 | norm 0.3918 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19132 | train 2.894673 | val 3.228252 | tok/s 2621439.91 | norm 0.3971 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19133 | train 3.047452 | val 3.228252 | tok/s 2623275.94 | norm 0.3894 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19134 | train 2.997526 | val 3.228252 | tok/s 2626768.31 | norm 0.3388 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19135 | train 3.149976 | val 3.228252 | tok/s 2625502.61 | norm 0.3563 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19136 | train 2.954674 | val 3.228252 | tok/s 2534699.92 | norm 0.4298 | dt 206.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19137 | train 2.947597 | val 3.228252 | tok/s 2607037.05 | norm 0.3455 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19138 | train 2.946176 | val 3.228252 | tok/s 2626506.44 | norm 0.3174 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19139 | train 2.878019 | val 3.228252 | tok/s 2627989.00 | norm 0.3760 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19140 | train 2.860785 | val 3.228252 | tok/s 2622382.33 | norm 0.3872 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19141 | train 2.942518 | val 3.228252 | tok/s 2626809.30 | norm 0.3775 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19142 | train 3.041065 | val 3.228252 | tok/s 2624963.16 | norm 0.3588 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19143 | train 2.968776 | val 3.228252 | tok/s 2463331.23 | norm 0.3755 | dt 212.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19144 | train 2.974989 | val 3.228252 | tok/s 2626835.49 | norm 0.3588 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19145 | train 3.084037 | val 3.228252 | tok/s 2628458.06 | norm 0.4053 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19146 | train 2.820110 | val 3.228252 | tok/s 2629682.42 | norm 0.3676 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19147 | train 2.919475 | val 3.228252 | tok/s 2627930.71 | norm 0.3832 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19148 | train 3.061772 | val 3.228252 | tok/s 2626277.63 | norm 0.4143 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19149 | train 3.174698 | val 3.228252 | tok/s 2632426.97 | norm 0.4012 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19150 | train 2.935806 | val 3.228252 | tok/s 2626612.05 | norm 0.3874 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19151 | train 2.935480 | val 3.228252 | tok/s 2626427.02 | norm 0.3867 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19152 | train 2.955253 | val 3.228252 | tok/s 2627219.32 | norm 0.3946 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19153 | train 2.979673 | val 3.228252 | tok/s 2628217.41 | norm 0.3665 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19154 | train 3.039302 | val 3.228252 | tok/s 2631353.22 | norm 0.3489 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19155 | train 3.037777 | val 3.228252 | tok/s 2628750.49 | norm 0.4045 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19156 | train 2.976493 | val 3.228252 | tok/s 2624326.94 | norm 0.3912 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19157 | train 3.026131 | val 3.228252 | tok/s 2623823.09 | norm 0.4138 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19158 | train 2.788778 | val 3.228252 | tok/s 2624963.84 | norm 0.3824 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19159 | train 2.780721 | val 3.228252 | tok/s 2627991.39 | norm 0.3664 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19160 | train 2.901028 | val 3.228252 | tok/s 2628476.67 | norm 0.3901 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19161 | train 2.884279 | val 3.228252 | tok/s 2630144.03 | norm 0.3516 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19162 | train 2.847291 | val 3.228252 | tok/s 2624188.55 | norm 0.3493 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19163 | train 2.884428 | val 3.228252 | tok/s 2628800.64 | norm 0.3803 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19164 | train 2.978572 | val 3.228252 | tok/s 2629757.82 | norm 0.3830 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19165 | train 2.970562 | val 3.228252 | tok/s 2626037.11 | norm 0.3316 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19166 | train 3.115693 | val 3.228252 | tok/s 2631898.44 | norm 0.4878 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19167 | train 2.936003 | val 3.228252 | tok/s 2630041.47 | norm 0.3669 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19168 | train 3.039212 | val 3.228252 | tok/s 2630037.01 | norm 0.3622 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19169 | train 2.937620 | val 3.228252 | tok/s 2631503.12 | norm 0.3831 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19170 | train 2.961201 | val 3.228252 | tok/s 2628595.80 | norm 0.3602 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19171 | train 3.110341 | val 3.228252 | tok/s 2628702.22 | norm 0.3916 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19172 | train 2.997634 | val 3.228252 | tok/s 2627158.62 | norm 0.3871 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19173 | train 2.914764 | val 3.228252 | tok/s 2628260.51 | norm 0.3594 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19174 | train 2.996312 | val 3.228252 | tok/s 2628409.17 | norm 0.3722 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19175 | train 2.887929 | val 3.228252 | tok/s 2626264.34 | norm 0.3757 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19176 | train 2.988822 | val 3.228252 | tok/s 2630256.49 | norm 0.3656 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19177 | train 2.960086 | val 3.228252 | tok/s 2626139.19 | norm 0.3822 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19178 | train 3.002163 | val 3.228252 | tok/s 2628896.98 | norm 0.3778 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19179 | train 3.081519 | val 3.228252 | tok/s 2628856.94 | norm 0.4431 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19180 | train 2.821753 | val 3.228252 | tok/s 2621088.39 | norm 0.3989 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19181 | train 2.908623 | val 3.228252 | tok/s 2629110.22 | norm 0.3776 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19182 | train 2.966065 | val 3.228252 | tok/s 2624368.80 | norm 0.3718 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19183 | train 2.890437 | val 3.228252 | tok/s 2627954.70 | norm 0.3798 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19184 | train 2.902195 | val 3.228252 | tok/s 2627690.26 | norm 0.4082 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19185 | train 2.926371 | val 3.228252 | tok/s 2628116.01 | norm 0.3752 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19186 | train 2.958991 | val 3.228252 | tok/s 2628846.74 | norm 0.3519 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19187 | train 2.929730 | val 3.228252 | tok/s 2626964.68 | norm 0.3868 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19188 | train 2.845315 | val 3.228252 | tok/s 2626071.93 | norm 0.3729 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19189 | train 2.948852 | val 3.228252 | tok/s 2625195.56 | norm 0.3677 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19190 | train 2.997946 | val 3.228252 | tok/s 2624776.80 | norm 0.3850 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19191 | train 2.923641 | val 3.228252 | tok/s 2627478.21 | norm 0.3634 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19192 | train 2.874122 | val 3.228252 | tok/s 2618074.54 | norm 0.3456 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19193 | train 2.918415 | val 3.228252 | tok/s 2440420.57 | norm 0.3631 | dt 214.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19194 | train 2.855160 | val 3.228252 | tok/s 2624876.75 | norm 0.3879 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19195 | train 2.895474 | val 3.228252 | tok/s 2630111.60 | norm 0.3896 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19196 | train 2.854813 | val 3.228252 | tok/s 2630149.25 | norm 0.3720 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19197 | train 3.195115 | val 3.228252 | tok/s 2628443.91 | norm 0.3848 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19198 | train 2.995554 | val 3.228252 | tok/s 2627288.69 | norm 0.3656 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19199 | train 3.006735 | val 3.228252 | tok/s 2628521.37 | norm 0.3287 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19200 | train 3.009653 | val 3.228252 | tok/s 2549673.99 | norm 0.3350 | dt 205.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19201 | train 2.824634 | val 3.228252 | tok/s 2631017.43 | norm 0.3481 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19202 | train 2.914324 | val 3.228252 | tok/s 2602737.11 | norm 0.4129 | dt 201.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19203 | train 2.788235 | val 3.228252 | tok/s 2627602.48 | norm 0.4096 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19204 | train 3.031473 | val 3.228252 | tok/s 2622108.53 | norm 0.3981 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19205 | train 2.996061 | val 3.228252 | tok/s 2626846.76 | norm 0.3859 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19206 | train 2.851430 | val 3.228252 | tok/s 2627677.37 | norm 0.3864 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19207 | train 2.951189 | val 3.228252 | tok/s 2629883.93 | norm 0.3892 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19208 | train 2.973696 | val 3.228252 | tok/s 2629270.97 | norm 0.3705 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19209 | train 2.833613 | val 3.228252 | tok/s 2633068.92 | norm 0.3792 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19210 | train 2.961176 | val 3.228252 | tok/s 2628350.20 | norm 0.3710 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19211 | train 2.889070 | val 3.228252 | tok/s 2632338.10 | norm 0.3762 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19212 | train 2.892622 | val 3.228252 | tok/s 2621574.63 | norm 0.3981 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19213 | train 2.902554 | val 3.228252 | tok/s 2634155.19 | norm 0.3831 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19214 | train 3.019181 | val 3.228252 | tok/s 2631923.41 | norm 0.3928 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19215 | train 2.885934 | val 3.228252 | tok/s 2628014.53 | norm 0.3793 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19216 | train 3.043942 | val 3.228252 | tok/s 2628289.23 | norm 0.3681 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19217 | train 2.978661 | val 3.228252 | tok/s 2624344.01 | norm 0.3687 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19218 | train 2.972988 | val 3.228252 | tok/s 2626925.39 | norm 0.3492 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19219 | train 2.886900 | val 3.228252 | tok/s 2624309.80 | norm 0.3715 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19220 | train 2.977504 | val 3.228252 | tok/s 2626874.93 | norm 0.3708 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19221 | train 2.900559 | val 3.228252 | tok/s 2627791.90 | norm 0.3599 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19222 | train 3.106236 | val 3.228252 | tok/s 2620537.79 | norm 0.3695 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19223 | train 2.995058 | val 3.228252 | tok/s 2629472.95 | norm 0.3565 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19224 | train 3.004473 | val 3.228252 | tok/s 2632107.27 | norm 0.3559 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19225 | train 2.973711 | val 3.228252 | tok/s 2632060.48 | norm 0.3551 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19226 | train 2.978965 | val 3.228252 | tok/s 2628475.26 | norm 0.3263 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19227 | train 2.916423 | val 3.228252 | tok/s 2627369.20 | norm 0.3543 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19228 | train 2.923932 | val 3.228252 | tok/s 2632306.45 | norm 0.3640 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19229 | train 2.891398 | val 3.228252 | tok/s 2630674.79 | norm 0.3817 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19230 | train 2.842250 | val 3.228252 | tok/s 2626600.44 | norm 0.3805 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19231 | train 3.218902 | val 3.228252 | tok/s 2631731.99 | norm 0.4273 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19232 | train 2.934189 | val 3.228252 | tok/s 2626898.26 | norm 0.3991 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19233 | train 2.978637 | val 3.228252 | tok/s 2630911.41 | norm 0.3357 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19234 | train 3.013020 | val 3.228252 | tok/s 2613139.48 | norm 0.3650 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19235 | train 2.886149 | val 3.228252 | tok/s 2631369.86 | norm 0.3835 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19236 | train 2.983543 | val 3.228252 | tok/s 2625538.86 | norm 0.4214 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19237 | train 2.862357 | val 3.228252 | tok/s 2625769.26 | norm 0.3877 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19238 | train 2.941540 | val 3.228252 | tok/s 2629285.82 | norm 0.3716 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19239 | train 2.897679 | val 3.228252 | tok/s 2630290.37 | norm 0.3775 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19240 | train 3.069413 | val 3.228252 | tok/s 2626774.24 | norm 0.3432 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19241 | train 3.133016 | val 3.228252 | tok/s 2620260.23 | norm 0.3611 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19242 | train 2.969493 | val 3.228252 | tok/s 2622755.98 | norm 0.3648 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19243 | train 2.881143 | val 3.228252 | tok/s 2625662.60 | norm 0.3666 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19244 | train 2.805628 | val 3.228252 | tok/s 2621876.69 | norm 0.3483 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19245 | train 2.935541 | val 3.228252 | tok/s 2624593.82 | norm 0.3499 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19246 | train 2.935099 | val 3.228252 | tok/s 2619056.44 | norm 0.3635 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19247 | train 3.002742 | val 3.228252 | tok/s 2623546.00 | norm 0.3556 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19248 | train 2.939703 | val 3.228252 | tok/s 2630494.44 | norm 0.3490 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19249 | train 2.900053 | val 3.228252 | tok/s 2628674.01 | norm 0.3761 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 19250 | val_loss 3.265834 step 19250 | train 2.932233 | val 3.265834 | tok/s 2452027.36 | norm 0.3619 | dt 213.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19251 | train 2.848279 | val 3.265834 | tok/s 2605611.57 | norm 0.3797 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19252 | train 2.975450 | val 3.265834 | tok/s 2614168.02 | norm 0.3895 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19253 | train 2.900241 | val 3.265834 | tok/s 2628565.52 | norm 0.3660 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19254 | train 3.031387 | val 3.265834 | tok/s 2627755.23 | norm 0.3759 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19255 | train 2.851702 | val 3.265834 | tok/s 2628495.84 | norm 0.3774 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19256 | train 2.953496 | val 3.265834 | tok/s 2626748.33 | norm 0.3891 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19257 | train 3.015693 | val 3.265834 | tok/s 2481603.74 | norm 0.3821 | dt 211.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19258 | train 2.963873 | val 3.265834 | tok/s 2625865.75 | norm 0.3732 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19259 | train 2.899515 | val 3.265834 | tok/s 2627025.38 | norm 0.3649 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19260 | train 2.913909 | val 3.265834 | tok/s 2624815.63 | norm 0.4083 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19261 | train 3.012672 | val 3.265834 | tok/s 2626945.47 | norm 0.4029 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19262 | train 2.873116 | val 3.265834 | tok/s 2625654.54 | norm 0.3853 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19263 | train 2.922794 | val 3.265834 | tok/s 2623672.84 | norm 0.3731 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19264 | train 2.876256 | val 3.265834 | tok/s 2622030.07 | norm 0.3824 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19265 | train 3.027651 | val 3.265834 | tok/s 2623217.85 | norm 0.4211 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19266 | train 2.956518 | val 3.265834 | tok/s 2630130.20 | norm 0.3665 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19267 | train 2.968842 | val 3.265834 | tok/s 2624811.41 | norm 0.3829 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19268 | train 2.886395 | val 3.265834 | tok/s 2628035.28 | norm 0.4128 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19269 | train 2.836176 | val 3.265834 | tok/s 2625435.76 | norm 0.3659 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19270 | train 3.158939 | val 3.265834 | tok/s 2627345.62 | norm 0.4531 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19271 | train 2.897172 | val 3.265834 | tok/s 2628112.78 | norm 0.3763 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19272 | train 3.008834 | val 3.265834 | tok/s 2629222.96 | norm 0.3339 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19273 | train 2.957664 | val 3.265834 | tok/s 2628507.16 | norm 0.3808 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19274 | train 2.903069 | val 3.265834 | tok/s 2628656.91 | norm 0.3709 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19275 | train 2.941550 | val 3.265834 | tok/s 2626568.81 | norm 0.3504 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19276 | train 3.021329 | val 3.265834 | tok/s 2623945.07 | norm 0.3481 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19277 | train 2.866109 | val 3.265834 | tok/s 2625747.70 | norm 0.3668 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19278 | train 2.970065 | val 3.265834 | tok/s 2628210.33 | norm 0.3612 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19279 | train 2.828463 | val 3.265834 | tok/s 2625625.23 | norm 0.3747 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19280 | train 2.884713 | val 3.265834 | tok/s 2627908.16 | norm 0.3848 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19281 | train 3.167066 | val 3.265834 | tok/s 2629606.19 | norm 0.4054 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19282 | train 2.923544 | val 3.265834 | tok/s 2630792.39 | norm 0.3562 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19283 | train 2.928673 | val 3.265834 | tok/s 2628895.48 | norm 0.3698 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19284 | train 2.897818 | val 3.265834 | tok/s 2622884.62 | norm 0.3498 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19285 | train 2.939634 | val 3.265834 | tok/s 2622022.19 | norm 0.3459 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19286 | train 2.938676 | val 3.265834 | tok/s 2634111.91 | norm 0.3590 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19287 | train 2.904835 | val 3.265834 | tok/s 2627746.74 | norm 0.3470 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19288 | train 2.870443 | val 3.265834 | tok/s 2627630.58 | norm 0.3912 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19289 | train 2.886170 | val 3.265834 | tok/s 2621704.02 | norm 0.3823 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19290 | train 2.871221 | val 3.265834 | tok/s 2607006.32 | norm 0.3465 | dt 201.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19291 | train 2.951688 | val 3.265834 | tok/s 2628988.06 | norm 0.3651 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19292 | train 3.007158 | val 3.265834 | tok/s 2630461.97 | norm 0.3839 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19293 | train 2.778379 | val 3.265834 | tok/s 2629322.84 | norm 0.3933 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19294 | train 2.932527 | val 3.265834 | tok/s 2624662.94 | norm 0.4005 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19295 | train 2.994984 | val 3.265834 | tok/s 2614202.48 | norm 0.3600 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19296 | train 2.869449 | val 3.265834 | tok/s 2628681.39 | norm 0.4244 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19297 | train 3.006658 | val 3.265834 | tok/s 2628134.77 | norm 0.3584 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19298 | train 2.790344 | val 3.265834 | tok/s 2629596.81 | norm 0.3507 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19299 | train 2.933355 | val 3.265834 | tok/s 2629183.91 | norm 0.3862 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19300 | train 2.954042 | val 3.265834 | tok/s 2623436.70 | norm 0.3958 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19301 | train 2.882676 | val 3.265834 | tok/s 2629285.87 | norm 0.3705 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19302 | train 2.919878 | val 3.265834 | tok/s 2522917.74 | norm 0.3669 | dt 207.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19303 | train 2.971687 | val 3.265834 | tok/s 2605579.55 | norm 0.3720 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19304 | train 2.894450 | val 3.265834 | tok/s 2625598.54 | norm 0.3677 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19305 | train 2.919420 | val 3.265834 | tok/s 2621432.52 | norm 0.3622 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19306 | train 2.933085 | val 3.265834 | tok/s 2627520.11 | norm 0.3620 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19307 | train 2.915352 | val 3.265834 | tok/s 2622311.84 | norm 0.3918 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19308 | train 2.956633 | val 3.265834 | tok/s 2621529.47 | norm 0.3864 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19309 | train 3.063118 | val 3.265834 | tok/s 2619013.52 | norm 0.4094 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19310 | train 2.955178 | val 3.265834 | tok/s 2626724.79 | norm 0.4439 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19311 | train 2.894529 | val 3.265834 | tok/s 2629140.30 | norm 0.3431 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19312 | train 2.966040 | val 3.265834 | tok/s 2623731.00 | norm 0.3317 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19313 | train 3.007754 | val 3.265834 | tok/s 2627379.88 | norm 0.3842 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19314 | train 2.900252 | val 3.265834 | tok/s 2624311.52 | norm 0.3393 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19315 | train 2.865399 | val 3.265834 | tok/s 2625887.39 | norm 0.3741 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19316 | train 2.972997 | val 3.265834 | tok/s 2623791.59 | norm 0.4024 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19317 | train 2.829509 | val 3.265834 | tok/s 2624694.74 | norm 0.3894 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19318 | train 2.849545 | val 3.265834 | tok/s 2588855.63 | norm 0.3845 | dt 202.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19319 | train 2.812991 | val 3.265834 | tok/s 2620550.83 | norm 0.3756 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19320 | train 2.900200 | val 3.265834 | tok/s 2578130.01 | norm 0.3849 | dt 203.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19321 | train 3.083931 | val 3.265834 | tok/s 2617667.51 | norm 0.4202 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19322 | train 2.997760 | val 3.265834 | tok/s 2617957.11 | norm 0.3939 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19323 | train 2.903567 | val 3.265834 | tok/s 2615681.13 | norm 0.3610 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19324 | train 2.831247 | val 3.265834 | tok/s 2627075.08 | norm 0.3510 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19325 | train 2.899607 | val 3.265834 | tok/s 2623849.28 | norm 0.3634 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19326 | train 2.893504 | val 3.265834 | tok/s 2623196.48 | norm 0.3434 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19327 | train 2.874993 | val 3.265834 | tok/s 2625165.71 | norm 0.3855 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19328 | train 2.980246 | val 3.265834 | tok/s 2626444.69 | norm 0.3513 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19329 | train 2.879808 | val 3.265834 | tok/s 2624897.40 | norm 0.3328 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19330 | train 2.898397 | val 3.265834 | tok/s 2628259.95 | norm 0.3823 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19331 | train 2.980368 | val 3.265834 | tok/s 2630992.92 | norm 0.4386 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19332 | train 2.917210 | val 3.265834 | tok/s 2627383.05 | norm 0.4052 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19333 | train 2.934070 | val 3.265834 | tok/s 2627905.26 | norm 0.3795 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19334 | train 2.900176 | val 3.265834 | tok/s 2630367.93 | norm 0.3772 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19335 | train 2.903594 | val 3.265834 | tok/s 2627346.81 | norm 0.3459 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19336 | train 3.040899 | val 3.265834 | tok/s 2633819.69 | norm 0.3416 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19337 | train 2.906475 | val 3.265834 | tok/s 2627512.63 | norm 0.3533 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19338 | train 2.999889 | val 3.265834 | tok/s 2626288.02 | norm 0.3269 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19339 | train 2.832216 | val 3.265834 | tok/s 2630206.49 | norm 0.3350 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19340 | train 3.050694 | val 3.265834 | tok/s 2631814.85 | norm 0.3760 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19341 | train 2.871718 | val 3.265834 | tok/s 2628554.62 | norm 0.4055 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19342 | train 2.875497 | val 3.265834 | tok/s 2626632.27 | norm 0.3436 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19343 | train 2.954456 | val 3.265834 | tok/s 2628539.40 | norm 0.3546 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19344 | train 2.948005 | val 3.265834 | tok/s 2628336.97 | norm 0.3249 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19345 | train 3.023649 | val 3.265834 | tok/s 2629064.68 | norm 0.3319 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19346 | train 2.913543 | val 3.265834 | tok/s 2624065.99 | norm 0.3539 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19347 | train 2.915884 | val 3.265834 | tok/s 2628230.01 | norm 0.3680 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19348 | train 2.876645 | val 3.265834 | tok/s 2630804.12 | norm 0.3688 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19349 | train 2.884169 | val 3.265834 | tok/s 2629652.07 | norm 0.3975 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19350 | train 3.021899 | val 3.265834 | tok/s 2629065.50 | norm 0.3874 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19351 | train 2.877815 | val 3.265834 | tok/s 2629598.70 | norm 0.3471 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19352 | train 2.974365 | val 3.265834 | tok/s 2627838.27 | norm 0.3688 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19353 | train 2.928359 | val 3.265834 | tok/s 2631795.14 | norm 0.3737 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19354 | train 2.967124 | val 3.265834 | tok/s 2630474.55 | norm 0.3639 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19355 | train 2.975061 | val 3.265834 | tok/s 2631369.87 | norm 0.3685 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19356 | train 2.974767 | val 3.265834 | tok/s 2624708.62 | norm 0.3920 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19357 | train 2.988928 | val 3.265834 | tok/s 2626199.95 | norm 0.4440 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19358 | train 2.938735 | val 3.265834 | tok/s 2629411.55 | norm 0.3823 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19359 | train 2.925774 | val 3.265834 | tok/s 2628631.88 | norm 0.3504 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19360 | train 3.070378 | val 3.265834 | tok/s 2533946.75 | norm 0.3617 | dt 206.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19361 | train 3.055290 | val 3.265834 | tok/s 2609581.63 | norm 0.3658 | dt 200.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19362 | train 3.076126 | val 3.265834 | tok/s 2627279.47 | norm 0.3534 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19363 | train 2.913369 | val 3.265834 | tok/s 2625149.93 | norm 0.3474 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19364 | train 2.874617 | val 3.265834 | tok/s 2624556.19 | norm 0.3680 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19365 | train 2.948292 | val 3.265834 | tok/s 2628883.66 | norm 0.3721 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19366 | train 3.040951 | val 3.265834 | tok/s 2627565.96 | norm 0.3700 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19367 | train 2.973450 | val 3.265834 | tok/s 2626156.88 | norm 0.3667 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19368 | train 2.969881 | val 3.265834 | tok/s 2629942.30 | norm 0.3730 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19369 | train 2.880777 | val 3.265834 | tok/s 2628201.92 | norm 0.3812 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19370 | train 2.846396 | val 3.265834 | tok/s 2630170.63 | norm 0.3654 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19371 | train 2.906048 | val 3.265834 | tok/s 2629527.74 | norm 0.3243 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19372 | train 2.956493 | val 3.265834 | tok/s 2628251.18 | norm 0.3361 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19373 | train 3.012576 | val 3.265834 | tok/s 2633507.79 | norm 0.3609 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19374 | train 2.874952 | val 3.265834 | tok/s 2629354.92 | norm 0.3691 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19375 | train 2.971205 | val 3.265834 | tok/s 2627529.80 | norm 0.3598 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19376 | train 2.944985 | val 3.265834 | tok/s 2627839.47 | norm 0.3510 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19377 | train 2.861316 | val 3.265834 | tok/s 2628565.20 | norm 0.3301 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19378 | train 2.877241 | val 3.265834 | tok/s 2629348.05 | norm 0.3496 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19379 | train 3.055917 | val 3.265834 | tok/s 2625844.26 | norm 0.3814 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19380 | train 2.922885 | val 3.265834 | tok/s 2630343.62 | norm 0.3863 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19381 | train 2.904708 | val 3.265834 | tok/s 2621787.54 | norm 0.3704 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19382 | train 2.947755 | val 3.265834 | tok/s 2629789.96 | norm 0.3958 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19383 | train 2.859824 | val 3.265834 | tok/s 2628981.27 | norm 0.3796 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19384 | train 3.032744 | val 3.265834 | tok/s 2548221.94 | norm 0.3491 | dt 205.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19385 | train 2.871169 | val 3.265834 | tok/s 2629708.69 | norm 0.3669 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19386 | train 2.965286 | val 3.265834 | tok/s 2631344.95 | norm 0.3995 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19387 | train 2.932570 | val 3.265834 | tok/s 2629139.41 | norm 0.3814 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19388 | train 2.944585 | val 3.265834 | tok/s 2622498.70 | norm 0.3694 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19389 | train 3.002144 | val 3.265834 | tok/s 2620589.46 | norm 0.3771 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19390 | train 2.901365 | val 3.265834 | tok/s 2632822.48 | norm 0.3443 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19391 | train 2.821784 | val 3.265834 | tok/s 2629412.62 | norm 0.3647 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19392 | train 2.923011 | val 3.265834 | tok/s 2628326.03 | norm 0.3887 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19393 | train 2.966642 | val 3.265834 | tok/s 2624621.21 | norm 0.3803 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19394 | train 3.067596 | val 3.265834 | tok/s 2625313.90 | norm 0.3883 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19395 | train 2.907151 | val 3.265834 | tok/s 2632839.54 | norm 0.3585 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19396 | train 2.917845 | val 3.265834 | tok/s 2628602.21 | norm 0.3627 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19397 | train 2.976803 | val 3.265834 | tok/s 2631274.78 | norm 0.4122 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19398 | train 2.926425 | val 3.265834 | tok/s 2626255.06 | norm 0.3722 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19399 | train 2.967534 | val 3.265834 | tok/s 2625222.88 | norm 0.3629 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19400 | train 3.125414 | val 3.265834 | tok/s 2629019.75 | norm 0.4143 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19401 | train 2.949302 | val 3.265834 | tok/s 2625198.33 | norm 0.3639 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19402 | train 2.937659 | val 3.265834 | tok/s 2628790.26 | norm 0.3715 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19403 | train 2.824051 | val 3.265834 | tok/s 2623127.98 | norm 0.3891 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19404 | train 2.977461 | val 3.265834 | tok/s 2625942.41 | norm 0.3813 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19405 | train 3.031240 | val 3.265834 | tok/s 2628700.38 | norm 0.3311 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19406 | train 2.827966 | val 3.265834 | tok/s 2625800.59 | norm 0.3669 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19407 | train 2.908129 | val 3.265834 | tok/s 2559074.51 | norm 0.3812 | dt 204.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19408 | train 2.928416 | val 3.265834 | tok/s 2624229.33 | norm 0.3453 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19409 | train 2.983496 | val 3.265834 | tok/s 2625969.25 | norm 0.4078 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19410 | train 2.938010 | val 3.265834 | tok/s 2590890.77 | norm 0.3739 | dt 202.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19411 | train 2.924681 | val 3.265834 | tok/s 2627954.75 | norm 0.3367 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19412 | train 2.937545 | val 3.265834 | tok/s 2629763.73 | norm 0.3821 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19413 | train 3.006968 | val 3.265834 | tok/s 2626732.00 | norm 0.3763 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19414 | train 2.936652 | val 3.265834 | tok/s 2627241.53 | norm 0.3720 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19415 | train 2.943146 | val 3.265834 | tok/s 2634174.61 | norm 0.3782 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19416 | train 2.919470 | val 3.265834 | tok/s 2632258.54 | norm 0.3783 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19417 | train 2.999865 | val 3.265834 | tok/s 2632210.81 | norm 0.3395 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19418 | train 2.983191 | val 3.265834 | tok/s 2628393.02 | norm 0.3437 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19419 | train 3.004072 | val 3.265834 | tok/s 2625696.81 | norm 0.3757 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19420 | train 2.874055 | val 3.265834 | tok/s 2630280.28 | norm 0.3876 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19421 | train 3.025891 | val 3.265834 | tok/s 2580276.14 | norm 0.3655 | dt 203.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19422 | train 2.909190 | val 3.265834 | tok/s 2630998.87 | norm 0.3322 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19423 | train 2.855426 | val 3.265834 | tok/s 2626874.45 | norm 0.3698 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19424 | train 2.966552 | val 3.265834 | tok/s 2624150.79 | norm 0.3743 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19425 | train 3.053064 | val 3.265834 | tok/s 2629799.74 | norm 0.4020 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19426 | train 2.962675 | val 3.265834 | tok/s 2616857.55 | norm 0.3657 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19427 | train 2.989141 | val 3.265834 | tok/s 2630578.47 | norm 0.3627 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19428 | train 3.009058 | val 3.265834 | tok/s 2624537.90 | norm 0.3837 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19429 | train 2.975088 | val 3.265834 | tok/s 2628330.10 | norm 0.3983 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19430 | train 2.995533 | val 3.265834 | tok/s 2626532.11 | norm 0.4085 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19431 | train 2.872388 | val 3.265834 | tok/s 2622360.32 | norm 0.3830 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19432 | train 2.892979 | val 3.265834 | tok/s 2628093.89 | norm 0.3584 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19433 | train 2.918545 | val 3.265834 | tok/s 2624931.68 | norm 0.3355 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19434 | train 2.820934 | val 3.265834 | tok/s 2629803.76 | norm 0.3243 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19435 | train 2.940433 | val 3.265834 | tok/s 2631408.13 | norm 0.3864 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19436 | train 2.954360 | val 3.265834 | tok/s 2628432.23 | norm 0.3500 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19437 | train 2.883721 | val 3.265834 | tok/s 2630590.92 | norm 0.3095 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19438 | train 2.904229 | val 3.265834 | tok/s 2627589.22 | norm 0.3646 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19439 | train 2.957084 | val 3.265834 | tok/s 2628159.60 | norm 0.3923 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19440 | train 3.077188 | val 3.265834 | tok/s 2630446.54 | norm 0.4433 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19441 | train 2.889715 | val 3.265834 | tok/s 2625392.34 | norm 0.3346 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19442 | train 2.858495 | val 3.265834 | tok/s 2631746.96 | norm 0.3303 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19443 | train 2.774975 | val 3.265834 | tok/s 2629329.72 | norm 0.3622 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19444 | train 2.905576 | val 3.265834 | tok/s 2632686.88 | norm 0.3749 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19445 | train 2.964978 | val 3.265834 | tok/s 2620969.50 | norm 0.3634 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19446 | train 2.845942 | val 3.265834 | tok/s 2626976.72 | norm 0.3525 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19447 | train 2.921005 | val 3.265834 | tok/s 2629760.99 | norm 0.3693 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19448 | train 3.000688 | val 3.265834 | tok/s 2627411.67 | norm 0.3603 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19449 | train 2.985488 | val 3.265834 | tok/s 2631099.62 | norm 0.3377 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19450 | train 2.926891 | val 3.265834 | tok/s 2626346.76 | norm 0.3698 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19451 | train 2.895023 | val 3.265834 | tok/s 2629090.55 | norm 0.3955 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19452 | train 2.939861 | val 3.265834 | tok/s 2627994.56 | norm 0.3976 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19453 | train 2.937679 | val 3.265834 | tok/s 2624430.79 | norm 0.3728 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19454 | train 2.802611 | val 3.265834 | tok/s 2632429.84 | norm 0.3499 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19455 | train 2.910450 | val 3.265834 | tok/s 2618704.17 | norm 0.4092 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19456 | train 2.795583 | val 3.265834 | tok/s 2628360.96 | norm 0.3701 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19457 | train 2.928668 | val 3.265834 | tok/s 2631478.34 | norm 0.3568 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19458 | train 2.910673 | val 3.265834 | tok/s 2630163.53 | norm 0.3548 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19459 | train 2.917021 | val 3.265834 | tok/s 2628838.80 | norm 0.3547 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19460 | train 2.913010 | val 3.265834 | tok/s 2629214.46 | norm 0.3793 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19461 | train 2.898441 | val 3.265834 | tok/s 2445515.91 | norm 0.3524 | dt 214.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19462 | train 2.914690 | val 3.265834 | tok/s 2614381.86 | norm 0.3550 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19463 | train 3.037332 | val 3.265834 | tok/s 2628053.51 | norm 0.3767 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19464 | train 2.975885 | val 3.265834 | tok/s 2629994.78 | norm 0.3887 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19465 | train 2.835449 | val 3.265834 | tok/s 2625621.21 | norm 0.3615 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19466 | train 2.959064 | val 3.265834 | tok/s 2629007.55 | norm 0.3963 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19467 | train 2.976310 | val 3.265834 | tok/s 2630004.29 | norm 0.3910 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19468 | train 2.946375 | val 3.265834 | tok/s 2529744.98 | norm 0.3865 | dt 207.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19469 | train 2.992799 | val 3.265834 | tok/s 2627711.65 | norm 0.3748 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19470 | train 2.978523 | val 3.265834 | tok/s 2625453.35 | norm 0.3615 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19471 | train 2.917445 | val 3.265834 | tok/s 2629253.15 | norm 0.3742 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19472 | train 2.880470 | val 3.265834 | tok/s 2626185.39 | norm 0.3692 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19473 | train 3.040742 | val 3.265834 | tok/s 2630406.65 | norm 0.3465 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19474 | train 3.006503 | val 3.265834 | tok/s 2633246.70 | norm 0.3394 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19475 | train 2.922840 | val 3.265834 | tok/s 2629095.52 | norm 0.3782 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19476 | train 2.839290 | val 3.265834 | tok/s 2631273.91 | norm 0.4052 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19477 | train 2.969310 | val 3.265834 | tok/s 2630923.07 | norm 0.3805 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19478 | train 3.080611 | val 3.265834 | tok/s 2624170.27 | norm 0.3746 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19479 | train 3.006179 | val 3.265834 | tok/s 2632354.31 | norm 0.3934 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19480 | train 2.976572 | val 3.265834 | tok/s 2615720.10 | norm 0.3817 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19481 | train 2.942967 | val 3.265834 | tok/s 2629174.90 | norm 0.3944 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19482 | train 2.982816 | val 3.265834 | tok/s 2634270.16 | norm 0.3951 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19483 | train 2.845537 | val 3.265834 | tok/s 2622714.97 | norm 0.3965 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19484 | train 3.000514 | val 3.265834 | tok/s 2632022.15 | norm 0.4126 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19485 | train 2.920075 | val 3.265834 | tok/s 2627142.23 | norm 0.3849 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19486 | train 2.954558 | val 3.265834 | tok/s 2628631.50 | norm 0.3941 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19487 | train 2.891840 | val 3.265834 | tok/s 2624900.38 | norm 0.4504 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19488 | train 2.987385 | val 3.265834 | tok/s 2624113.25 | norm 0.3966 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19489 | train 2.940086 | val 3.265834 | tok/s 2630479.12 | norm 0.3511 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19490 | train 2.993305 | val 3.265834 | tok/s 2628439.90 | norm 0.3804 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19491 | train 2.880112 | val 3.265834 | tok/s 2629402.28 | norm 0.3984 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19492 | train 3.060034 | val 3.265834 | tok/s 2623218.15 | norm 0.3793 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19493 | train 2.962089 | val 3.265834 | tok/s 2625313.48 | norm 0.3373 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19494 | train 2.994168 | val 3.265834 | tok/s 2633201.75 | norm 0.3896 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19495 | train 2.912258 | val 3.265834 | tok/s 2632320.86 | norm 0.3804 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19496 | train 2.923748 | val 3.265834 | tok/s 2631920.25 | norm 0.3348 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19497 | train 2.936689 | val 3.265834 | tok/s 2628260.13 | norm 0.3674 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19498 | train 2.994597 | val 3.265834 | tok/s 2624287.11 | norm 0.3641 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19499 | train 2.892751 | val 3.265834 | tok/s 2628872.94 | norm 0.3867 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 19500 | val_loss 3.283141 step 19500 | train 2.897318 | val 3.283141 | tok/s 2553621.73 | norm 0.3889 | dt 205.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19501 | train 2.965726 | val 3.283141 | tok/s 2628014.58 | norm 0.3809 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19502 | train 2.959542 | val 3.283141 | tok/s 2630860.94 | norm 0.3663 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19503 | train 2.783928 | val 3.283141 | tok/s 2629586.72 | norm 0.3337 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19504 | train 3.168565 | val 3.283141 | tok/s 2629260.99 | norm 0.4328 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19505 | train 2.969497 | val 3.283141 | tok/s 2627194.15 | norm 0.5009 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19506 | train 2.978286 | val 3.283141 | tok/s 2626663.21 | norm 0.3969 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19507 | train 3.060658 | val 3.283141 | tok/s 2630426.40 | norm 0.4203 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19508 | train 2.943771 | val 3.283141 | tok/s 2624310.29 | norm 0.3665 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19509 | train 3.004616 | val 3.283141 | tok/s 2623275.35 | norm 0.3557 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19510 | train 2.862968 | val 3.283141 | tok/s 2627895.25 | norm 0.3459 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19511 | train 2.924866 | val 3.283141 | tok/s 2626249.19 | norm 0.4110 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19512 | train 2.848790 | val 3.283141 | tok/s 2629312.27 | norm 0.4437 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19513 | train 2.901215 | val 3.283141 | tok/s 2625509.95 | norm 0.4092 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19514 | train 3.024479 | val 3.283141 | tok/s 2627769.47 | norm 0.3830 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19515 | train 2.914084 | val 3.283141 | tok/s 2627648.56 | norm 0.3382 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19516 | train 3.009902 | val 3.283141 | tok/s 2632090.74 | norm 0.3603 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19517 | train 2.952349 | val 3.283141 | tok/s 2629915.96 | norm 0.3351 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19518 | train 3.002477 | val 3.283141 | tok/s 2625222.58 | norm 0.3411 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19519 | train 2.908187 | val 3.283141 | tok/s 2629401.33 | norm 0.3780 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19520 | train 2.988987 | val 3.283141 | tok/s 2632235.43 | norm 0.3967 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19521 | train 2.835217 | val 3.283141 | tok/s 2632102.16 | norm 0.4085 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19522 | train 2.845956 | val 3.283141 | tok/s 2629712.56 | norm 0.4047 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19523 | train 3.069092 | val 3.283141 | tok/s 2620472.93 | norm 0.4090 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19524 | train 3.064086 | val 3.283141 | tok/s 2628102.09 | norm 0.4168 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19525 | train 2.943944 | val 3.283141 | tok/s 2626733.53 | norm 0.3785 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19526 | train 2.887482 | val 3.283141 | tok/s 2626219.79 | norm 0.3647 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19527 | train 2.899423 | val 3.283141 | tok/s 2627837.11 | norm 0.3741 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19528 | train 2.918198 | val 3.283141 | tok/s 2623895.49 | norm 0.3791 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19529 | train 3.086825 | val 3.283141 | tok/s 2628251.46 | norm 0.4299 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19530 | train 3.014588 | val 3.283141 | tok/s 2629500.88 | norm 0.3885 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19531 | train 3.005974 | val 3.283141 | tok/s 2622135.72 | norm 0.3419 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19532 | train 2.967369 | val 3.283141 | tok/s 2479919.83 | norm 0.3739 | dt 211.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19533 | train 2.912610 | val 3.283141 | tok/s 2473099.84 | norm 0.4059 | dt 212.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19534 | train 2.846542 | val 3.283141 | tok/s 2626840.94 | norm 0.3901 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19535 | train 2.851418 | val 3.283141 | tok/s 2630468.99 | norm 0.4062 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19536 | train 2.862640 | val 3.283141 | tok/s 2625766.84 | norm 0.4281 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19537 | train 3.071960 | val 3.283141 | tok/s 2627703.08 | norm 0.4403 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19538 | train 3.072019 | val 3.283141 | tok/s 2629392.43 | norm 0.3801 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19539 | train 2.847151 | val 3.283141 | tok/s 2627685.47 | norm 0.3457 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19540 | train 2.926668 | val 3.283141 | tok/s 2632173.20 | norm 0.3702 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19541 | train 3.204426 | val 3.283141 | tok/s 2623926.28 | norm 0.3865 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19542 | train 2.947155 | val 3.283141 | tok/s 2632756.73 | norm 0.3761 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19543 | train 2.989417 | val 3.283141 | tok/s 2630356.46 | norm 0.3532 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19544 | train 2.965030 | val 3.283141 | tok/s 2628655.01 | norm 0.3551 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19545 | train 2.957266 | val 3.283141 | tok/s 2623975.15 | norm 0.3846 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19546 | train 2.973142 | val 3.283141 | tok/s 2618200.98 | norm 0.3867 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19547 | train 2.926134 | val 3.283141 | tok/s 2623949.68 | norm 0.3885 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19548 | train 2.871970 | val 3.283141 | tok/s 2624913.36 | norm 0.4014 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19549 | train 2.882556 | val 3.283141 | tok/s 2629572.39 | norm 0.4288 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19550 | train 2.951405 | val 3.283141 | tok/s 2627873.99 | norm 0.3681 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19551 | train 2.899490 | val 3.283141 | tok/s 2622683.19 | norm 0.3149 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19552 | train 2.914626 | val 3.283141 | tok/s 2628322.18 | norm 0.3270 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19553 | train 2.916550 | val 3.283141 | tok/s 2612208.77 | norm 0.3372 | dt 200.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19554 | train 2.935450 | val 3.283141 | tok/s 2629325.43 | norm 0.3484 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19555 | train 2.849019 | val 3.283141 | tok/s 2627387.96 | norm 0.3495 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19556 | train 2.920598 | val 3.283141 | tok/s 2606012.46 | norm 0.3787 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19557 | train 2.893072 | val 3.283141 | tok/s 2633314.63 | norm 0.3765 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19558 | train 2.824235 | val 3.283141 | tok/s 2625867.00 | norm 0.4410 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19559 | train 3.000674 | val 3.283141 | tok/s 2628541.16 | norm 0.3920 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19560 | train 2.919764 | val 3.283141 | tok/s 2629789.31 | norm 0.3684 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19561 | train 2.964818 | val 3.283141 | tok/s 2630119.05 | norm 0.3614 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19562 | train 2.941049 | val 3.283141 | tok/s 2631028.36 | norm 0.3684 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19563 | train 2.835428 | val 3.283141 | tok/s 2629292.53 | norm 0.3737 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19564 | train 2.909124 | val 3.283141 | tok/s 2630017.71 | norm 0.3672 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19565 | train 2.887380 | val 3.283141 | tok/s 2627052.80 | norm 0.3786 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19566 | train 2.946718 | val 3.283141 | tok/s 2629413.17 | norm 0.3898 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19567 | train 2.838705 | val 3.283141 | tok/s 2630561.96 | norm 0.3818 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19568 | train 2.939431 | val 3.283141 | tok/s 2631430.53 | norm 0.3634 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19569 | train 2.921671 | val 3.283141 | tok/s 2630315.80 | norm 0.3636 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19570 | train 2.974938 | val 3.283141 | tok/s 2629727.57 | norm 0.3771 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19571 | train 2.968519 | val 3.283141 | tok/s 2627254.14 | norm 0.3595 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19572 | train 2.855707 | val 3.283141 | tok/s 2628607.07 | norm 0.3874 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19573 | train 2.988388 | val 3.283141 | tok/s 2627280.22 | norm 0.3787 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19574 | train 2.821143 | val 3.283141 | tok/s 2630882.67 | norm 0.3494 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19575 | train 2.929844 | val 3.283141 | tok/s 2537434.91 | norm 0.3514 | dt 206.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19576 | train 2.727941 | val 3.283141 | tok/s 2466801.57 | norm 0.3474 | dt 212.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19577 | train 2.980594 | val 3.283141 | tok/s 2632477.61 | norm 0.3681 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19578 | train 2.971506 | val 3.283141 | tok/s 2629208.02 | norm 0.3656 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19579 | train 2.896735 | val 3.283141 | tok/s 2624666.55 | norm 0.3616 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19580 | train 2.829382 | val 3.283141 | tok/s 2624168.22 | norm 0.3472 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19581 | train 2.888253 | val 3.283141 | tok/s 2626129.18 | norm 0.3494 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19582 | train 2.991447 | val 3.283141 | tok/s 2624579.97 | norm 0.3697 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19583 | train 3.060517 | val 3.283141 | tok/s 2624411.52 | norm 0.3820 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19584 | train 3.077577 | val 3.283141 | tok/s 2628216.63 | norm 0.3941 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19585 | train 2.946232 | val 3.283141 | tok/s 2629520.86 | norm 0.3481 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19586 | train 2.924244 | val 3.283141 | tok/s 2633014.14 | norm 0.3501 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19587 | train 2.937133 | val 3.283141 | tok/s 2629210.47 | norm 0.4112 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19588 | train 2.985163 | val 3.283141 | tok/s 2623142.83 | norm 0.3954 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19589 | train 2.890423 | val 3.283141 | tok/s 2628016.52 | norm 0.3991 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19590 | train 3.065891 | val 3.283141 | tok/s 2627122.23 | norm 0.4074 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19591 | train 3.014389 | val 3.283141 | tok/s 2630682.10 | norm 0.3724 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19592 | train 2.967962 | val 3.283141 | tok/s 2623076.16 | norm 0.3969 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19593 | train 3.018700 | val 3.283141 | tok/s 2628378.24 | norm 0.4146 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19594 | train 2.889444 | val 3.283141 | tok/s 2627661.01 | norm 0.3582 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19595 | train 2.960644 | val 3.283141 | tok/s 2627159.64 | norm 0.3312 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19596 | train 2.996097 | val 3.283141 | tok/s 2625602.14 | norm 0.3562 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19597 | train 2.838143 | val 3.283141 | tok/s 2618664.71 | norm 0.3564 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19598 | train 2.917312 | val 3.283141 | tok/s 2626116.34 | norm 0.3490 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19599 | train 3.023147 | val 3.283141 | tok/s 2628782.79 | norm 0.3790 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19600 | train 2.963697 | val 3.283141 | tok/s 2628183.74 | norm 0.3865 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19601 | train 2.966979 | val 3.283141 | tok/s 2631319.12 | norm 0.3872 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19602 | train 2.883663 | val 3.283141 | tok/s 2623682.38 | norm 0.3753 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19603 | train 2.964799 | val 3.283141 | tok/s 2627047.99 | norm 0.3733 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19604 | train 3.000631 | val 3.283141 | tok/s 2629721.24 | norm 0.3721 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19605 | train 2.970149 | val 3.283141 | tok/s 2625965.94 | norm 0.3755 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19606 | train 2.942585 | val 3.283141 | tok/s 2629114.27 | norm 0.3757 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19607 | train 2.874251 | val 3.283141 | tok/s 2623218.19 | norm 0.3739 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19608 | train 3.039420 | val 3.283141 | tok/s 2623415.63 | norm 0.4316 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19609 | train 2.990700 | val 3.283141 | tok/s 2629088.61 | norm 0.4089 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19610 | train 2.950496 | val 3.283141 | tok/s 2630541.75 | norm 0.3683 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19611 | train 2.907375 | val 3.283141 | tok/s 2633454.85 | norm 0.3754 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19612 | train 2.899132 | val 3.283141 | tok/s 2628803.91 | norm 0.3637 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19613 | train 2.949385 | val 3.283141 | tok/s 2626684.11 | norm 0.3487 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19614 | train 2.993366 | val 3.283141 | tok/s 2628596.75 | norm 0.3726 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19615 | train 2.954460 | val 3.283141 | tok/s 2625162.39 | norm 0.3684 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19616 | train 2.978688 | val 3.283141 | tok/s 2624280.28 | norm 0.4356 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19617 | train 2.929077 | val 3.283141 | tok/s 2620376.73 | norm 0.4251 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19618 | train 2.889120 | val 3.283141 | tok/s 2608497.65 | norm 0.3883 | dt 200.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19619 | train 2.946041 | val 3.283141 | tok/s 2628780.29 | norm 0.4089 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19620 | train 2.972112 | val 3.283141 | tok/s 2627412.88 | norm 0.3845 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19621 | train 2.983361 | val 3.283141 | tok/s 2626193.49 | norm 0.3715 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19622 | train 3.080222 | val 3.283141 | tok/s 2625067.21 | norm 0.3648 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19623 | train 2.896673 | val 3.283141 | tok/s 2626027.49 | norm 0.3397 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19624 | train 3.002585 | val 3.283141 | tok/s 2439718.80 | norm 0.3690 | dt 214.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19625 | train 2.968095 | val 3.283141 | tok/s 2623731.42 | norm 0.3550 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19626 | train 3.058164 | val 3.283141 | tok/s 2625385.74 | norm 0.3041 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19627 | train 2.912997 | val 3.283141 | tok/s 2622359.06 | norm 0.3198 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19628 | train 2.934767 | val 3.283141 | tok/s 2622297.33 | norm 0.3686 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19629 | train 3.076040 | val 3.283141 | tok/s 2622365.31 | norm 0.3903 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19630 | train 2.885569 | val 3.283141 | tok/s 2622823.76 | norm 0.3702 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19631 | train 2.901474 | val 3.283141 | tok/s 2623628.49 | norm 0.3531 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19632 | train 3.024665 | val 3.283141 | tok/s 2626059.97 | norm 0.4105 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19633 | train 2.986246 | val 3.283141 | tok/s 2625128.66 | norm 0.3592 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19634 | train 2.875428 | val 3.283141 | tok/s 2625263.45 | norm 0.3348 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19635 | train 2.932322 | val 3.283141 | tok/s 2620528.95 | norm 0.3351 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19636 | train 2.882008 | val 3.283141 | tok/s 2625932.34 | norm 0.4032 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19637 | train 2.787301 | val 3.283141 | tok/s 2627327.67 | norm 0.4068 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19638 | train 2.902517 | val 3.283141 | tok/s 2627463.88 | norm 0.3679 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19639 | train 3.022965 | val 3.283141 | tok/s 2606420.89 | norm 0.3377 | dt 201.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19640 | train 3.047421 | val 3.283141 | tok/s 2613250.27 | norm 0.4723 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19641 | train 3.110874 | val 3.283141 | tok/s 2627435.12 | norm 0.4032 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19642 | train 3.012847 | val 3.283141 | tok/s 2626634.39 | norm 0.3646 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19643 | train 2.848057 | val 3.283141 | tok/s 2623490.01 | norm 0.3650 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19644 | train 2.938666 | val 3.283141 | tok/s 2623242.55 | norm 0.3578 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19645 | train 3.002400 | val 3.283141 | tok/s 2618410.66 | norm 0.3578 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19646 | train 2.913743 | val 3.283141 | tok/s 2627172.45 | norm 0.3569 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19647 | train 2.992684 | val 3.283141 | tok/s 2620373.46 | norm 0.3380 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19648 | train 2.774170 | val 3.283141 | tok/s 2624565.87 | norm 0.3447 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19649 | train 3.034956 | val 3.283141 | tok/s 2623699.98 | norm 0.3848 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19650 | train 2.903242 | val 3.283141 | tok/s 2621355.15 | norm 0.3657 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19651 | train 2.932287 | val 3.283141 | tok/s 2622771.33 | norm 0.3850 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19652 | train 3.033265 | val 3.283141 | tok/s 2625806.33 | norm 0.3697 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19653 | train 2.995522 | val 3.283141 | tok/s 2628202.00 | norm 0.3450 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19654 | train 2.965078 | val 3.283141 | tok/s 2625090.66 | norm 0.3590 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19655 | train 3.098801 | val 3.283141 | tok/s 2622007.72 | norm 0.3693 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19656 | train 2.918588 | val 3.283141 | tok/s 2624595.24 | norm 0.3606 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19657 | train 2.882354 | val 3.283141 | tok/s 2623856.20 | norm 0.3998 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19658 | train 2.938927 | val 3.283141 | tok/s 2625142.46 | norm 0.3816 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19659 | train 3.022291 | val 3.283141 | tok/s 2624731.23 | norm 0.3615 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19660 | train 2.865653 | val 3.283141 | tok/s 2624250.61 | norm 0.3467 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19661 | train 3.056376 | val 3.283141 | tok/s 2626838.42 | norm 0.3676 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19662 | train 2.887798 | val 3.283141 | tok/s 2627792.89 | norm 0.4063 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19663 | train 2.911527 | val 3.283141 | tok/s 2628854.14 | norm 0.3654 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19664 | train 2.925036 | val 3.283141 | tok/s 2625781.16 | norm 0.3717 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19665 | train 2.989182 | val 3.283141 | tok/s 2624561.89 | norm 0.3525 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19666 | train 3.075610 | val 3.283141 | tok/s 2629296.55 | norm 0.3561 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19667 | train 2.871683 | val 3.283141 | tok/s 2628570.98 | norm 0.3704 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19668 | train 2.971086 | val 3.283141 | tok/s 2626485.04 | norm 0.3486 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19669 | train 2.981680 | val 3.283141 | tok/s 2624607.49 | norm 0.3552 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19670 | train 2.918649 | val 3.283141 | tok/s 2626909.82 | norm 0.3794 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19671 | train 2.899599 | val 3.283141 | tok/s 2626516.33 | norm 0.3891 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19672 | train 2.904277 | val 3.283141 | tok/s 2514780.28 | norm 0.3840 | dt 208.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19673 | train 3.151413 | val 3.283141 | tok/s 2628874.35 | norm 0.4027 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19674 | train 3.082204 | val 3.283141 | tok/s 2628679.58 | norm 0.3941 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19675 | train 2.880876 | val 3.283141 | tok/s 2626796.58 | norm 0.3584 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19676 | train 2.842225 | val 3.283141 | tok/s 2624614.42 | norm 0.3872 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19677 | train 2.893190 | val 3.283141 | tok/s 2620602.10 | norm 0.3863 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19678 | train 2.955989 | val 3.283141 | tok/s 2631701.32 | norm 0.3594 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19679 | train 2.879726 | val 3.283141 | tok/s 2626823.57 | norm 0.3862 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19680 | train 3.002277 | val 3.283141 | tok/s 2624167.59 | norm 0.3740 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19681 | train 2.819345 | val 3.283141 | tok/s 2629662.30 | norm 0.3570 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19682 | train 3.020090 | val 3.283141 | tok/s 2613814.47 | norm 0.3745 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19683 | train 2.943995 | val 3.283141 | tok/s 2622541.38 | norm 0.3882 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19684 | train 2.833679 | val 3.283141 | tok/s 2622192.50 | norm 0.3650 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19685 | train 2.847539 | val 3.283141 | tok/s 2623968.27 | norm 0.3326 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19686 | train 2.866280 | val 3.283141 | tok/s 2620735.63 | norm 0.3669 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19687 | train 2.914704 | val 3.283141 | tok/s 2618579.39 | norm 0.3598 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19688 | train 2.873700 | val 3.283141 | tok/s 2625026.47 | norm 0.3832 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19689 | train 2.856771 | val 3.283141 | tok/s 2624581.11 | norm 0.3924 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19690 | train 2.861574 | val 3.283141 | tok/s 2624977.34 | norm 0.3600 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19691 | train 2.984169 | val 3.283141 | tok/s 2626060.97 | norm 0.3763 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19692 | train 2.916705 | val 3.283141 | tok/s 2623692.73 | norm 0.3486 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19693 | train 2.972122 | val 3.283141 | tok/s 2630559.40 | norm 0.3320 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19694 | train 2.935514 | val 3.283141 | tok/s 2626416.01 | norm 0.3571 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19695 | train 2.947257 | val 3.283141 | tok/s 2458470.42 | norm 0.3683 | dt 213.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19696 | train 3.020184 | val 3.283141 | tok/s 2612431.66 | norm 0.3881 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19697 | train 2.910168 | val 3.283141 | tok/s 2624353.48 | norm 0.3737 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19698 | train 2.893877 | val 3.283141 | tok/s 2632120.71 | norm 0.3774 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19699 | train 3.054859 | val 3.283141 | tok/s 2623978.42 | norm 0.3975 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19700 | train 2.926105 | val 3.283141 | tok/s 2627559.24 | norm 0.3774 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19701 | train 2.908142 | val 3.283141 | tok/s 2623871.19 | norm 0.3511 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19702 | train 2.983972 | val 3.283141 | tok/s 2617618.92 | norm 0.3817 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19703 | train 2.897203 | val 3.283141 | tok/s 2631395.12 | norm 0.3733 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19704 | train 2.909121 | val 3.283141 | tok/s 2559725.07 | norm 0.3830 | dt 204.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19705 | train 2.900339 | val 3.283141 | tok/s 2628060.78 | norm 0.4053 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19706 | train 2.981405 | val 3.283141 | tok/s 2600054.67 | norm 0.3667 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19707 | train 3.081891 | val 3.283141 | tok/s 2628547.17 | norm 0.4045 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19708 | train 2.954211 | val 3.283141 | tok/s 2631954.59 | norm 0.3760 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19709 | train 2.911825 | val 3.283141 | tok/s 2617154.40 | norm 0.3486 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19710 | train 3.031592 | val 3.283141 | tok/s 2628839.08 | norm 0.3594 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19711 | train 2.964883 | val 3.283141 | tok/s 2624327.43 | norm 0.3598 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19712 | train 3.020994 | val 3.283141 | tok/s 2625282.42 | norm 0.3785 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19713 | train 3.076195 | val 3.283141 | tok/s 2623219.12 | norm 0.3570 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19714 | train 2.822599 | val 3.283141 | tok/s 2620105.70 | norm 0.3656 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19715 | train 2.915894 | val 3.283141 | tok/s 2625303.07 | norm 0.3622 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19716 | train 2.959969 | val 3.283141 | tok/s 2624944.80 | norm 0.3587 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19717 | train 2.911917 | val 3.283141 | tok/s 2630609.66 | norm 0.3579 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19718 | train 3.018041 | val 3.283141 | tok/s 2621176.77 | norm 0.3752 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19719 | train 2.831388 | val 3.283141 | tok/s 2624033.36 | norm 0.3685 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19720 | train 2.964736 | val 3.283141 | tok/s 2628998.41 | norm 0.3514 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19721 | train 2.821619 | val 3.283141 | tok/s 2628343.90 | norm 0.3615 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19722 | train 3.121558 | val 3.283141 | tok/s 2628573.16 | norm 0.4028 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19723 | train 2.968691 | val 3.283141 | tok/s 2622596.47 | norm 0.4483 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19724 | train 2.880826 | val 3.283141 | tok/s 2623772.63 | norm 0.3486 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19725 | train 2.943365 | val 3.283141 | tok/s 2628095.59 | norm 0.3695 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19726 | train 2.938308 | val 3.283141 | tok/s 2628858.15 | norm 0.3691 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19727 | train 2.853923 | val 3.283141 | tok/s 2626417.78 | norm 0.3681 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19728 | train 2.821351 | val 3.283141 | tok/s 2627011.44 | norm 0.3649 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19729 | train 2.897539 | val 3.283141 | tok/s 2622687.52 | norm 0.3547 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19730 | train 2.823290 | val 3.283141 | tok/s 2630477.24 | norm 0.3752 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19731 | train 2.951119 | val 3.283141 | tok/s 2623582.26 | norm 0.4379 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19732 | train 3.087461 | val 3.283141 | tok/s 2593069.58 | norm 0.3767 | dt 202.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19733 | train 2.924489 | val 3.283141 | tok/s 2621127.93 | norm 0.3563 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19734 | train 3.013000 | val 3.283141 | tok/s 2626812.51 | norm 0.3760 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19735 | train 2.999369 | val 3.283141 | tok/s 2628850.40 | norm 0.3611 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19736 | train 2.727396 | val 3.283141 | tok/s 2531593.64 | norm 0.3496 | dt 207.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19737 | train 2.902349 | val 3.283141 | tok/s 2525258.53 | norm 0.3723 | dt 207.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19738 | train 2.897285 | val 3.283141 | tok/s 2557549.37 | norm 0.3935 | dt 205.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19739 | train 2.963501 | val 3.283141 | tok/s 2625073.17 | norm 0.3608 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19740 | train 2.996716 | val 3.283141 | tok/s 2624866.24 | norm 0.3547 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19741 | train 2.984228 | val 3.283141 | tok/s 2620381.43 | norm 0.3448 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19742 | train 2.981611 | val 3.283141 | tok/s 2628070.27 | norm 0.3996 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19743 | train 2.868559 | val 3.283141 | tok/s 2625781.41 | norm 0.3884 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19744 | train 2.984095 | val 3.283141 | tok/s 2626748.25 | norm 0.3474 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19745 | train 3.033062 | val 3.283141 | tok/s 2620889.34 | norm 0.3618 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19746 | train 2.989124 | val 3.283141 | tok/s 2620500.24 | norm 0.3811 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19747 | train 3.085018 | val 3.283141 | tok/s 2624763.55 | norm 0.3894 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19748 | train 3.052685 | val 3.283141 | tok/s 2625663.77 | norm 0.3650 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19749 | train 3.015965 | val 3.283141 | tok/s 2627414.45 | norm 0.3559 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 19750 | val_loss 3.284945 step 19750 | train 3.124727 | val 3.284945 | tok/s 2580526.11 | norm 0.3317 | dt 203.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19751 | train 3.003145 | val 3.284945 | tok/s 2614943.10 | norm 0.3546 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19752 | train 2.839193 | val 3.284945 | tok/s 2620624.20 | norm 0.3344 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19753 | train 2.963846 | val 3.284945 | tok/s 2627437.48 | norm 0.3234 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19754 | train 2.920494 | val 3.284945 | tok/s 2630526.46 | norm 0.3498 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19755 | train 3.016813 | val 3.284945 | tok/s 2623195.88 | norm 0.3416 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19756 | train 3.008558 | val 3.284945 | tok/s 2627839.80 | norm 0.3614 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19757 | train 2.911108 | val 3.284945 | tok/s 2591703.94 | norm 0.4079 | dt 202.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19758 | train 2.960982 | val 3.284945 | tok/s 2627219.46 | norm 0.3741 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19759 | train 3.028413 | val 3.284945 | tok/s 2625605.81 | norm 0.3713 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19760 | train 2.954009 | val 3.284945 | tok/s 2629064.68 | norm 0.3709 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19761 | train 2.966474 | val 3.284945 | tok/s 2624068.23 | norm 0.3715 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19762 | train 2.924158 | val 3.284945 | tok/s 2622217.70 | norm 0.3777 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19763 | train 2.909712 | val 3.284945 | tok/s 2627283.13 | norm 0.3748 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19764 | train 3.096084 | val 3.284945 | tok/s 2631007.82 | norm 0.3848 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19765 | train 2.975520 | val 3.284945 | tok/s 2597668.88 | norm 0.4033 | dt 201.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19766 | train 2.869090 | val 3.284945 | tok/s 2629708.53 | norm 0.4022 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19767 | train 2.936952 | val 3.284945 | tok/s 2620187.71 | norm 0.3876 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19768 | train 3.049819 | val 3.284945 | tok/s 2630684.86 | norm 0.3825 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19769 | train 2.874529 | val 3.284945 | tok/s 2628644.66 | norm 0.4109 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19770 | train 2.928446 | val 3.284945 | tok/s 2624432.64 | norm 0.3826 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19771 | train 2.971309 | val 3.284945 | tok/s 2493994.71 | norm 0.3859 | dt 210.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19772 | train 2.880912 | val 3.284945 | tok/s 2626137.90 | norm 0.3629 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19773 | train 2.849508 | val 3.284945 | tok/s 2626987.24 | norm 0.3223 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19774 | train 3.044629 | val 3.284945 | tok/s 2614318.22 | norm 0.3681 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19775 | train 2.961117 | val 3.284945 | tok/s 2628670.82 | norm 0.3513 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19776 | train 2.890409 | val 3.284945 | tok/s 2625771.80 | norm 0.3284 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19777 | train 2.927402 | val 3.284945 | tok/s 2625208.46 | norm 0.3474 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19778 | train 3.025352 | val 3.284945 | tok/s 2616020.23 | norm 0.3691 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19779 | train 2.955954 | val 3.284945 | tok/s 2587787.77 | norm 0.3864 | dt 202.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19780 | train 2.933161 | val 3.284945 | tok/s 2625642.55 | norm 0.3736 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19781 | train 3.008499 | val 3.284945 | tok/s 2623244.70 | norm 0.3601 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19782 | train 2.900503 | val 3.284945 | tok/s 2623635.09 | norm 0.3678 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19783 | train 3.125788 | val 3.284945 | tok/s 2618430.26 | norm 0.4121 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19784 | train 2.975835 | val 3.284945 | tok/s 2620268.39 | norm 0.3796 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19785 | train 2.981440 | val 3.284945 | tok/s 2629484.63 | norm 0.3735 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19786 | train 2.944837 | val 3.284945 | tok/s 2570520.26 | norm 0.3866 | dt 203.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19787 | train 2.894994 | val 3.284945 | tok/s 2626775.37 | norm 0.3612 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19788 | train 2.930728 | val 3.284945 | tok/s 2624654.09 | norm 0.3698 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19789 | train 3.053418 | val 3.284945 | tok/s 2625118.12 | norm 0.3849 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19790 | train 2.948141 | val 3.284945 | tok/s 2627258.77 | norm 0.3979 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19791 | train 2.858990 | val 3.284945 | tok/s 2624486.79 | norm 0.4138 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19792 | train 3.128223 | val 3.284945 | tok/s 2624744.25 | norm 0.4592 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19793 | train 2.935992 | val 3.284945 | tok/s 2623078.13 | norm 0.3688 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19794 | train 2.932479 | val 3.284945 | tok/s 2621011.66 | norm 0.3549 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19795 | train 2.990762 | val 3.284945 | tok/s 2627577.93 | norm 0.3949 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19796 | train 2.821226 | val 3.284945 | tok/s 2620306.28 | norm 0.3840 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19797 | train 2.902586 | val 3.284945 | tok/s 2522791.15 | norm 0.3304 | dt 207.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19798 | train 2.851140 | val 3.284945 | tok/s 2609967.18 | norm 0.3274 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19799 | train 2.939642 | val 3.284945 | tok/s 2622734.04 | norm 0.3466 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19800 | train 2.874467 | val 3.284945 | tok/s 2628331.58 | norm 0.3545 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19801 | train 2.948240 | val 3.284945 | tok/s 2622323.08 | norm 0.3914 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19802 | train 2.879940 | val 3.284945 | tok/s 2629711.53 | norm 0.4303 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19803 | train 2.952573 | val 3.284945 | tok/s 2623965.32 | norm 0.4026 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19804 | train 2.947665 | val 3.284945 | tok/s 2574206.97 | norm 0.3862 | dt 203.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19805 | train 2.888204 | val 3.284945 | tok/s 2599301.75 | norm 0.4048 | dt 201.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19806 | train 2.862093 | val 3.284945 | tok/s 2619502.46 | norm 0.3891 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19807 | train 2.790000 | val 3.284945 | tok/s 2626459.40 | norm 0.3403 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19808 | train 2.991023 | val 3.284945 | tok/s 2625661.09 | norm 0.3697 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19809 | train 3.022071 | val 3.284945 | tok/s 2627380.09 | norm 0.3555 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19810 | train 2.901996 | val 3.284945 | tok/s 2625334.71 | norm 0.3429 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19811 | train 2.891685 | val 3.284945 | tok/s 2606080.98 | norm 0.3626 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19812 | train 2.986161 | val 3.284945 | tok/s 2627262.80 | norm 0.3863 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19813 | train 2.870848 | val 3.284945 | tok/s 2626399.75 | norm 0.3786 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19814 | train 2.979943 | val 3.284945 | tok/s 2624073.14 | norm 0.3559 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19815 | train 3.007869 | val 3.284945 | tok/s 2627180.88 | norm 0.3411 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19816 | train 2.971690 | val 3.284945 | tok/s 2624681.90 | norm 0.3554 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19817 | train 2.904946 | val 3.284945 | tok/s 2608273.10 | norm 0.3613 | dt 201.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19818 | train 2.821274 | val 3.284945 | tok/s 2626901.18 | norm 0.3817 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19819 | train 2.877554 | val 3.284945 | tok/s 2627458.15 | norm 0.3857 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19820 | train 2.901035 | val 3.284945 | tok/s 2627593.56 | norm 0.3940 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19821 | train 2.930952 | val 3.284945 | tok/s 2626635.47 | norm 0.3886 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19822 | train 2.916151 | val 3.284945 | tok/s 2628152.99 | norm 0.3792 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19823 | train 2.964120 | val 3.284945 | tok/s 2624743.30 | norm 0.3860 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19824 | train 2.960045 | val 3.284945 | tok/s 2626681.07 | norm 0.3905 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19825 | train 2.915443 | val 3.284945 | tok/s 2627347.80 | norm 0.3810 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19826 | train 2.834055 | val 3.284945 | tok/s 2626528.48 | norm 0.4054 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19827 | train 2.967303 | val 3.284945 | tok/s 2632783.48 | norm 0.4077 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19828 | train 2.929802 | val 3.284945 | tok/s 2623556.41 | norm 0.3746 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19829 | train 2.974195 | val 3.284945 | tok/s 2626677.42 | norm 0.3552 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19830 | train 2.874640 | val 3.284945 | tok/s 2627065.48 | norm 0.3698 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19831 | train 2.884193 | val 3.284945 | tok/s 2627426.39 | norm 0.3349 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19832 | train 3.006001 | val 3.284945 | tok/s 2629336.50 | norm 0.3344 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19833 | train 2.957158 | val 3.284945 | tok/s 2623449.14 | norm 0.3592 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19834 | train 2.844308 | val 3.284945 | tok/s 2625515.81 | norm 0.3421 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19835 | train 3.020282 | val 3.284945 | tok/s 2626980.17 | norm 0.3736 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19836 | train 2.935723 | val 3.284945 | tok/s 2624997.97 | norm 0.3976 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19837 | train 2.917245 | val 3.284945 | tok/s 2623901.63 | norm 0.3979 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19838 | train 2.961013 | val 3.284945 | tok/s 2522263.60 | norm 0.3703 | dt 207.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19839 | train 2.891868 | val 3.284945 | tok/s 2628837.87 | norm 0.3546 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19840 | train 3.033084 | val 3.284945 | tok/s 2587960.17 | norm 0.3807 | dt 202.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19841 | train 2.949218 | val 3.284945 | tok/s 2631030.30 | norm 0.3671 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19842 | train 2.978994 | val 3.284945 | tok/s 2627418.70 | norm 0.4694 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19843 | train 2.866150 | val 3.284945 | tok/s 2623837.39 | norm 0.3735 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19844 | train 2.921563 | val 3.284945 | tok/s 2628037.25 | norm 0.3309 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19845 | train 2.963352 | val 3.284945 | tok/s 2628281.89 | norm 0.3600 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19846 | train 2.997493 | val 3.284945 | tok/s 2626522.37 | norm 0.3833 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19847 | train 2.941739 | val 3.284945 | tok/s 2612292.70 | norm 0.3658 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19848 | train 2.853816 | val 3.284945 | tok/s 2621742.58 | norm 0.3899 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19849 | train 3.020906 | val 3.284945 | tok/s 2615731.06 | norm 0.3704 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19850 | train 2.947039 | val 3.284945 | tok/s 2624334.60 | norm 0.3406 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19851 | train 2.944696 | val 3.284945 | tok/s 2629376.48 | norm 0.3367 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19852 | train 2.964710 | val 3.284945 | tok/s 2630847.98 | norm 0.3775 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19853 | train 2.993046 | val 3.284945 | tok/s 2624118.04 | norm 0.3727 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19854 | train 2.904179 | val 3.284945 | tok/s 2631911.63 | norm 0.3703 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19855 | train 2.950950 | val 3.284945 | tok/s 2622041.81 | norm 0.4174 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19856 | train 2.877253 | val 3.284945 | tok/s 2622985.51 | norm 0.3751 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19857 | train 3.014539 | val 3.284945 | tok/s 2629067.16 | norm 0.3630 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19858 | train 2.905753 | val 3.284945 | tok/s 2624776.00 | norm 0.3644 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19859 | train 2.842163 | val 3.284945 | tok/s 2630811.71 | norm 0.3853 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19860 | train 2.970384 | val 3.284945 | tok/s 2627777.24 | norm 0.3872 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19861 | train 2.936936 | val 3.284945 | tok/s 2627597.25 | norm 0.3699 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19862 | train 2.994630 | val 3.284945 | tok/s 2625860.31 | norm 0.3570 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19863 | train 2.868424 | val 3.284945 | tok/s 2623789.10 | norm 0.3457 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19864 | train 2.866078 | val 3.284945 | tok/s 2632494.49 | norm 0.3542 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19865 | train 3.063141 | val 3.284945 | tok/s 2631749.52 | norm 0.3929 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19866 | train 3.050656 | val 3.284945 | tok/s 2628880.92 | norm 0.3734 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19867 | train 3.025357 | val 3.284945 | tok/s 2624860.18 | norm 0.3541 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19868 | train 2.997215 | val 3.284945 | tok/s 2626191.29 | norm 0.3588 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19869 | train 2.905575 | val 3.284945 | tok/s 2630912.22 | norm 0.3942 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19870 | train 3.179524 | val 3.284945 | tok/s 2495157.78 | norm 0.4860 | dt 210.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19871 | train 2.874478 | val 3.284945 | tok/s 2624506.13 | norm 0.4442 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19872 | train 2.973875 | val 3.284945 | tok/s 2490877.27 | norm 0.3692 | dt 210.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19873 | train 2.976505 | val 3.284945 | tok/s 2630082.77 | norm 0.3771 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19874 | train 2.905262 | val 3.284945 | tok/s 2629342.79 | norm 0.3854 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19875 | train 2.890292 | val 3.284945 | tok/s 2619932.06 | norm 0.3830 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19876 | train 2.986609 | val 3.284945 | tok/s 2629745.70 | norm 0.3691 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19877 | train 2.924114 | val 3.284945 | tok/s 2629207.03 | norm 0.3835 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19878 | train 2.904866 | val 3.284945 | tok/s 2626373.72 | norm 0.3643 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19879 | train 3.033288 | val 3.284945 | tok/s 2626619.29 | norm 0.4050 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19880 | train 2.922124 | val 3.284945 | tok/s 2602269.59 | norm 0.3817 | dt 201.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19881 | train 2.881550 | val 3.284945 | tok/s 2627211.72 | norm 0.3452 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19882 | train 2.891151 | val 3.284945 | tok/s 2628164.45 | norm 0.3517 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19883 | train 3.003102 | val 3.284945 | tok/s 2626380.23 | norm 0.3551 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19884 | train 2.840531 | val 3.284945 | tok/s 2629255.88 | norm 0.3752 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19885 | train 3.006341 | val 3.284945 | tok/s 2626341.66 | norm 0.3795 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19886 | train 2.974332 | val 3.284945 | tok/s 2625508.01 | norm 0.4057 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19887 | train 3.000510 | val 3.284945 | tok/s 2618736.10 | norm 0.3632 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19888 | train 2.868372 | val 3.284945 | tok/s 2622376.44 | norm 0.3486 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19889 | train 2.921640 | val 3.284945 | tok/s 2618457.38 | norm 0.3654 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19890 | train 2.921688 | val 3.284945 | tok/s 2622523.90 | norm 0.3662 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19891 | train 2.981869 | val 3.284945 | tok/s 2615696.19 | norm 0.3509 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19892 | train 2.812099 | val 3.284945 | tok/s 2627090.33 | norm 0.3745 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19893 | train 2.980731 | val 3.284945 | tok/s 2629212.99 | norm 0.3829 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19894 | train 2.944114 | val 3.284945 | tok/s 2623637.49 | norm 0.3506 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19895 | train 3.234910 | val 3.284945 | tok/s 2623799.68 | norm 0.4318 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19896 | train 2.919230 | val 3.284945 | tok/s 2548189.92 | norm 0.4315 | dt 205.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19897 | train 3.089681 | val 3.284945 | tok/s 2441632.27 | norm 0.3650 | dt 214.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19898 | train 3.011575 | val 3.284945 | tok/s 2624444.28 | norm 0.3742 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19899 | train 3.073797 | val 3.284945 | tok/s 2624940.48 | norm 0.3869 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19900 | train 2.923152 | val 3.284945 | tok/s 2627379.31 | norm 0.3884 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19901 | train 2.812864 | val 3.284945 | tok/s 2628721.51 | norm 0.3750 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19902 | train 2.931556 | val 3.284945 | tok/s 2622940.01 | norm 0.3687 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19903 | train 3.108936 | val 3.284945 | tok/s 2627953.55 | norm 0.3697 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19904 | train 2.972831 | val 3.284945 | tok/s 2628079.96 | norm 0.3930 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19905 | train 2.936767 | val 3.284945 | tok/s 2612238.22 | norm 0.3629 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19906 | train 2.886507 | val 3.284945 | tok/s 2614096.25 | norm 0.3832 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19907 | train 2.980482 | val 3.284945 | tok/s 2619956.92 | norm 0.3729 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19908 | train 2.907218 | val 3.284945 | tok/s 2626385.71 | norm 0.3736 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19909 | train 2.797820 | val 3.284945 | tok/s 2629849.51 | norm 0.3751 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19910 | train 2.930110 | val 3.284945 | tok/s 2629498.68 | norm 0.3663 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19911 | train 2.887851 | val 3.284945 | tok/s 2617990.04 | norm 0.3320 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19912 | train 2.924795 | val 3.284945 | tok/s 2617286.86 | norm 0.3564 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19913 | train 2.960030 | val 3.284945 | tok/s 2630624.96 | norm 0.3456 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19914 | train 2.864538 | val 3.284945 | tok/s 2630038.41 | norm 0.3470 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19915 | train 2.943723 | val 3.284945 | tok/s 2627803.57 | norm 0.3701 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19916 | train 2.790615 | val 3.284945 | tok/s 2624754.44 | norm 0.3719 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19917 | train 2.976233 | val 3.284945 | tok/s 2623476.83 | norm 0.3625 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19918 | train 2.881230 | val 3.284945 | tok/s 2629712.15 | norm 0.3584 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19919 | train 2.904795 | val 3.284945 | tok/s 2623936.31 | norm 0.3929 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19920 | train 2.861147 | val 3.284945 | tok/s 2627442.50 | norm 0.3482 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19921 | train 2.777121 | val 3.284945 | tok/s 2622973.21 | norm 0.3601 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19922 | train 2.905753 | val 3.284945 | tok/s 2623388.14 | norm 0.3874 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19923 | train 2.890689 | val 3.284945 | tok/s 2629971.93 | norm 0.3893 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19924 | train 2.818300 | val 3.284945 | tok/s 2629164.89 | norm 0.3802 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19925 | train 2.893676 | val 3.284945 | tok/s 2628352.99 | norm 0.3942 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19926 | train 2.931841 | val 3.284945 | tok/s 2626473.33 | norm 0.3475 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19927 | train 2.939727 | val 3.284945 | tok/s 2626207.63 | norm 0.3416 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19928 | train 2.960932 | val 3.284945 | tok/s 2630984.25 | norm 0.3321 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19929 | train 2.947703 | val 3.284945 | tok/s 2625820.56 | norm 0.3555 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19930 | train 2.926919 | val 3.284945 | tok/s 2627418.48 | norm 0.3749 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19931 | train 2.954420 | val 3.284945 | tok/s 2535775.03 | norm 0.4012 | dt 206.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19932 | train 2.840514 | val 3.284945 | tok/s 2457603.26 | norm 0.3761 | dt 213.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19933 | train 2.978662 | val 3.284945 | tok/s 2623323.95 | norm 0.3718 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19934 | train 2.979619 | val 3.284945 | tok/s 2612907.58 | norm 0.3942 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19935 | train 2.913739 | val 3.284945 | tok/s 2626675.24 | norm 0.3742 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19936 | train 2.975718 | val 3.284945 | tok/s 2624348.49 | norm 0.3700 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19937 | train 2.955183 | val 3.284945 | tok/s 2623094.35 | norm 0.3656 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19938 | train 2.868084 | val 3.284945 | tok/s 2617766.08 | norm 0.3613 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19939 | train 2.857337 | val 3.284945 | tok/s 2597457.96 | norm 0.3899 | dt 201.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19940 | train 2.830239 | val 3.284945 | tok/s 2626017.04 | norm 0.3864 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19941 | train 3.018940 | val 3.284945 | tok/s 2625451.51 | norm 0.3575 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19942 | train 2.989068 | val 3.284945 | tok/s 2625661.63 | norm 0.3751 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19943 | train 2.949855 | val 3.284945 | tok/s 2618166.42 | norm 0.3674 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19944 | train 2.876434 | val 3.284945 | tok/s 2621549.08 | norm 0.3474 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19945 | train 2.967622 | val 3.284945 | tok/s 2628657.76 | norm 0.3739 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19946 | train 2.904078 | val 3.284945 | tok/s 2629185.36 | norm 0.4210 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19947 | train 2.984456 | val 3.284945 | tok/s 2625561.67 | norm 0.3651 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19948 | train 2.881583 | val 3.284945 | tok/s 2610090.13 | norm 0.3576 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19949 | train 2.941173 | val 3.284945 | tok/s 2624457.39 | norm 0.3778 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19950 | train 3.045799 | val 3.284945 | tok/s 2631658.24 | norm 0.4823 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19951 | train 2.851412 | val 3.284945 | tok/s 2624405.16 | norm 0.3785 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19952 | train 3.074339 | val 3.284945 | tok/s 2626328.65 | norm 0.3919 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19953 | train 2.948039 | val 3.284945 | tok/s 2620333.30 | norm 0.3484 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19954 | train 2.766338 | val 3.284945 | tok/s 2623104.69 | norm 0.3375 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19955 | train 3.010122 | val 3.284945 | tok/s 2632361.42 | norm 0.4100 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19956 | train 2.874357 | val 3.284945 | tok/s 2625483.34 | norm 0.3869 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19957 | train 2.890091 | val 3.284945 | tok/s 2630570.34 | norm 0.4054 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19958 | train 2.877715 | val 3.284945 | tok/s 2626107.35 | norm 0.3828 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19959 | train 3.042987 | val 3.284945 | tok/s 2625082.58 | norm 0.3588 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19960 | train 3.009717 | val 3.284945 | tok/s 2632307.22 | norm 0.3970 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19961 | train 2.808378 | val 3.284945 | tok/s 2626811.37 | norm 0.3889 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19962 | train 2.959746 | val 3.284945 | tok/s 2630771.90 | norm 0.3970 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19963 | train 2.907072 | val 3.284945 | tok/s 2628766.91 | norm 0.3853 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19964 | train 2.835119 | val 3.284945 | tok/s 2625703.53 | norm 0.3753 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19965 | train 2.935977 | val 3.284945 | tok/s 2439162.76 | norm 0.4391 | dt 214.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19966 | train 3.040252 | val 3.284945 | tok/s 2614096.26 | norm 0.3498 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19967 | train 2.907765 | val 3.284945 | tok/s 2631834.15 | norm 0.3675 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19968 | train 3.073009 | val 3.284945 | tok/s 2631123.84 | norm 0.4154 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19969 | train 2.830076 | val 3.284945 | tok/s 2628738.03 | norm 0.3808 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19970 | train 2.868470 | val 3.284945 | tok/s 2626575.88 | norm 0.3354 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19971 | train 2.899627 | val 3.284945 | tok/s 2621552.06 | norm 0.3468 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19972 | train 2.942742 | val 3.284945 | tok/s 2551895.91 | norm 0.3480 | dt 205.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19973 | train 2.952573 | val 3.284945 | tok/s 2628033.20 | norm 0.3635 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19974 | train 2.959629 | val 3.284945 | tok/s 2627847.51 | norm 0.3911 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19975 | train 2.841570 | val 3.284945 | tok/s 2630696.37 | norm 0.3740 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19976 | train 2.941238 | val 3.284945 | tok/s 2622601.26 | norm 0.3627 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19977 | train 2.981025 | val 3.284945 | tok/s 2623809.72 | norm 0.3744 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19978 | train 2.696083 | val 3.284945 | tok/s 2628051.93 | norm 0.4985 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19979 | train 2.986900 | val 3.284945 | tok/s 2626665.92 | norm 0.3230 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19980 | train 2.865586 | val 3.284945 | tok/s 2618184.01 | norm 0.3408 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19981 | train 2.882065 | val 3.284945 | tok/s 2621544.14 | norm 0.3689 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19982 | train 3.111391 | val 3.284945 | tok/s 2629992.20 | norm 0.3935 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19983 | train 2.970130 | val 3.284945 | tok/s 2627913.02 | norm 0.3280 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19984 | train 2.873100 | val 3.284945 | tok/s 2627921.03 | norm 0.3769 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19985 | train 3.012438 | val 3.284945 | tok/s 2627804.91 | norm 0.3362 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19986 | train 2.940865 | val 3.284945 | tok/s 2625936.97 | norm 0.3618 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19987 | train 2.948059 | val 3.284945 | tok/s 2630857.42 | norm 0.3650 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19988 | train 2.831989 | val 3.284945 | tok/s 2625966.05 | norm 0.3514 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19989 | train 2.989441 | val 3.284945 | tok/s 2627259.91 | norm 0.3837 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19990 | train 2.926019 | val 3.284945 | tok/s 2626016.86 | norm 0.3881 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19991 | train 2.857678 | val 3.284945 | tok/s 2624793.67 | norm 0.3649 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19992 | train 2.942061 | val 3.284945 | tok/s 2622246.34 | norm 0.3407 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19993 | train 2.954084 | val 3.284945 | tok/s 2624123.85 | norm 0.3561 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19994 | train 2.888225 | val 3.284945 | tok/s 2630210.12 | norm 0.3894 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19995 | train 2.865453 | val 3.284945 | tok/s 2626045.51 | norm 0.3873 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19996 | train 2.983609 | val 3.284945 | tok/s 2622801.21 | norm 0.3652 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19997 | train 2.942782 | val 3.284945 | tok/s 2629345.40 | norm 0.3595 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19998 | train 2.912955 | val 3.284945 | tok/s 2629389.96 | norm 0.3729 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 19999 | train 2.912355 | val 3.284945 | tok/s 2627602.28 | norm 0.3688 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 20000 | val_loss 3.292327 step 20000 | train 2.978763 | val 3.292327 | tok/s 2605652.35 | norm 0.3594 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20001 | train 2.881557 | val 3.292327 | tok/s 2612350.13 | norm 0.3522 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20002 | train 3.040005 | val 3.292327 | tok/s 2625215.95 | norm 0.3531 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20003 | train 2.841701 | val 3.292327 | tok/s 2629111.68 | norm 0.3861 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20004 | train 2.910288 | val 3.292327 | tok/s 2627924.32 | norm 0.3895 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20005 | train 2.911560 | val 3.292327 | tok/s 2628121.66 | norm 0.3797 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20006 | train 2.873225 | val 3.292327 | tok/s 2623928.31 | norm 0.3886 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20007 | train 2.955483 | val 3.292327 | tok/s 2622671.02 | norm 0.3782 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20008 | train 2.988688 | val 3.292327 | tok/s 2630430.96 | norm 0.3734 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20009 | train 2.973115 | val 3.292327 | tok/s 2624899.09 | norm 0.3724 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20010 | train 2.981484 | val 3.292327 | tok/s 2624701.73 | norm 0.3918 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20011 | train 3.053862 | val 3.292327 | tok/s 2626848.67 | norm 0.3672 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20012 | train 2.945014 | val 3.292327 | tok/s 2625848.39 | norm 0.3511 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20013 | train 2.861069 | val 3.292327 | tok/s 2624534.73 | norm 0.3463 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20014 | train 2.953394 | val 3.292327 | tok/s 2621420.61 | norm 0.3485 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20015 | train 2.893577 | val 3.292327 | tok/s 2625975.23 | norm 0.3956 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20016 | train 2.925896 | val 3.292327 | tok/s 2627165.30 | norm 0.3969 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20017 | train 2.958142 | val 3.292327 | tok/s 2626999.76 | norm 0.3523 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20018 | train 3.009229 | val 3.292327 | tok/s 2626712.64 | norm 0.3618 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20019 | train 2.832658 | val 3.292327 | tok/s 2621221.64 | norm 0.3934 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20020 | train 2.968570 | val 3.292327 | tok/s 2629406.27 | norm 0.3682 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20021 | train 3.057663 | val 3.292327 | tok/s 2611923.39 | norm 0.3509 | dt 200.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20022 | train 3.087697 | val 3.292327 | tok/s 2629876.94 | norm 0.4000 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20023 | train 2.989768 | val 3.292327 | tok/s 2632426.95 | norm 0.4119 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20024 | train 2.899927 | val 3.292327 | tok/s 2520425.61 | norm 0.3662 | dt 208.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20025 | train 2.964610 | val 3.292327 | tok/s 2631392.53 | norm 0.3660 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20026 | train 2.910815 | val 3.292327 | tok/s 2582981.83 | norm 0.3535 | dt 202.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20027 | train 2.954740 | val 3.292327 | tok/s 2627844.16 | norm 0.3601 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20028 | train 2.882848 | val 3.292327 | tok/s 2629003.31 | norm 0.3640 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20029 | train 3.023939 | val 3.292327 | tok/s 2626431.51 | norm 0.3733 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20030 | train 2.916185 | val 3.292327 | tok/s 2631952.06 | norm 0.3616 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20031 | train 2.996814 | val 3.292327 | tok/s 2628522.09 | norm 0.3409 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20032 | train 2.876797 | val 3.292327 | tok/s 2625287.42 | norm 0.3888 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20033 | train 2.898287 | val 3.292327 | tok/s 2629441.38 | norm 0.3931 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20034 | train 2.952653 | val 3.292327 | tok/s 2625951.09 | norm 0.3809 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20035 | train 2.950050 | val 3.292327 | tok/s 2630612.50 | norm 0.3497 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20036 | train 2.833278 | val 3.292327 | tok/s 2627948.85 | norm 0.3436 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20037 | train 2.952749 | val 3.292327 | tok/s 2630172.69 | norm 0.3878 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20038 | train 2.835621 | val 3.292327 | tok/s 2628243.18 | norm 0.3987 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20039 | train 2.935409 | val 3.292327 | tok/s 2624681.34 | norm 0.3810 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20040 | train 2.910099 | val 3.292327 | tok/s 2602918.17 | norm 0.3894 | dt 201.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20041 | train 2.919700 | val 3.292327 | tok/s 2627566.37 | norm 0.3882 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20042 | train 2.906160 | val 3.292327 | tok/s 2620691.83 | norm 0.3648 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20043 | train 2.968627 | val 3.292327 | tok/s 2625031.62 | norm 0.3795 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20044 | train 2.878346 | val 3.292327 | tok/s 2632277.10 | norm 0.3655 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20045 | train 2.921782 | val 3.292327 | tok/s 2624408.21 | norm 0.3758 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20046 | train 3.004603 | val 3.292327 | tok/s 2628763.91 | norm 0.3554 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20047 | train 3.061528 | val 3.292327 | tok/s 2629092.08 | norm 0.3550 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20048 | train 3.027494 | val 3.292327 | tok/s 2623721.47 | norm 0.3860 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20049 | train 2.898289 | val 3.292327 | tok/s 2628544.26 | norm 0.3392 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20050 | train 2.892451 | val 3.292327 | tok/s 2630075.95 | norm 0.3645 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20051 | train 2.990255 | val 3.292327 | tok/s 2627699.89 | norm 0.4404 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20052 | train 3.125574 | val 3.292327 | tok/s 2625594.68 | norm 0.4304 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20053 | train 2.944301 | val 3.292327 | tok/s 2585199.54 | norm 0.3956 | dt 202.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20054 | train 3.053522 | val 3.292327 | tok/s 2630400.39 | norm 0.3909 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20055 | train 2.927194 | val 3.292327 | tok/s 2628951.30 | norm 0.4853 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20056 | train 2.992318 | val 3.292327 | tok/s 2626253.61 | norm 0.3523 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20057 | train 2.949342 | val 3.292327 | tok/s 2627832.40 | norm 0.3263 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20058 | train 3.255636 | val 3.292327 | tok/s 2626651.00 | norm 0.4462 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20059 | train 3.016585 | val 3.292327 | tok/s 2627230.50 | norm 0.4280 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20060 | train 2.898416 | val 3.292327 | tok/s 2627585.38 | norm 0.3632 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20061 | train 2.880020 | val 3.292327 | tok/s 2629603.92 | norm 0.3762 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20062 | train 2.980183 | val 3.292327 | tok/s 2624285.53 | norm 0.3969 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20063 | train 2.902062 | val 3.292327 | tok/s 2621855.25 | norm 0.4388 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20064 | train 2.965020 | val 3.292327 | tok/s 2627820.01 | norm 0.3457 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20065 | train 2.996948 | val 3.292327 | tok/s 2624086.33 | norm 0.3373 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20066 | train 2.920636 | val 3.292327 | tok/s 2625904.62 | norm 0.3710 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20067 | train 2.860429 | val 3.292327 | tok/s 2620953.87 | norm 0.3735 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20068 | train 2.986078 | val 3.292327 | tok/s 2621390.86 | norm 0.3969 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20069 | train 2.922048 | val 3.292327 | tok/s 2629522.07 | norm 0.3506 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20070 | train 3.009028 | val 3.292327 | tok/s 2626330.62 | norm 0.3525 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20071 | train 3.043625 | val 3.292327 | tok/s 2631577.47 | norm 0.3890 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20072 | train 3.075096 | val 3.292327 | tok/s 2623898.99 | norm 0.5337 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20073 | train 2.953336 | val 3.292327 | tok/s 2607674.06 | norm 0.3284 | dt 201.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20074 | train 2.952978 | val 3.292327 | tok/s 2626788.30 | norm 0.3191 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20075 | train 2.978080 | val 3.292327 | tok/s 2539691.69 | norm 0.3757 | dt 206.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20076 | train 2.808864 | val 3.292327 | tok/s 2629986.80 | norm 0.3441 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20077 | train 2.814780 | val 3.292327 | tok/s 2626627.68 | norm 0.3285 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20078 | train 2.905800 | val 3.292327 | tok/s 2627065.44 | norm 0.3239 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20079 | train 3.045743 | val 3.292327 | tok/s 2620692.02 | norm 0.4246 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20080 | train 2.918547 | val 3.292327 | tok/s 2616743.46 | norm 0.3938 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20081 | train 2.847831 | val 3.292327 | tok/s 2626840.56 | norm 0.3725 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20082 | train 2.814315 | val 3.292327 | tok/s 2624942.24 | norm 0.3886 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20083 | train 2.829492 | val 3.292327 | tok/s 2623397.47 | norm 0.3829 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20084 | train 2.992116 | val 3.292327 | tok/s 2619330.84 | norm 0.3711 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20085 | train 3.031156 | val 3.292327 | tok/s 2616731.07 | norm 0.3643 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20086 | train 2.880747 | val 3.292327 | tok/s 2627113.66 | norm 0.3454 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20087 | train 2.926746 | val 3.292327 | tok/s 2624979.69 | norm 0.3404 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20088 | train 3.005065 | val 3.292327 | tok/s 2604371.48 | norm 0.3606 | dt 201.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20089 | train 3.039268 | val 3.292327 | tok/s 2589725.20 | norm 0.3746 | dt 202.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20090 | train 2.956275 | val 3.292327 | tok/s 2623835.94 | norm 0.3591 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20091 | train 2.882455 | val 3.292327 | tok/s 2623696.77 | norm 0.3652 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20092 | train 2.859952 | val 3.292327 | tok/s 2625302.86 | norm 0.3353 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20093 | train 2.946846 | val 3.292327 | tok/s 2624413.05 | norm 0.3471 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20094 | train 2.961326 | val 3.292327 | tok/s 2623838.96 | norm 0.4118 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20095 | train 2.970559 | val 3.292327 | tok/s 2621038.97 | norm 0.3877 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20096 | train 3.022311 | val 3.292327 | tok/s 2626523.87 | norm 0.3490 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20097 | train 3.059474 | val 3.292327 | tok/s 2621930.45 | norm 0.3568 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20098 | train 2.982136 | val 3.292327 | tok/s 2630082.45 | norm 0.3576 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20099 | train 3.102200 | val 3.292327 | tok/s 2625312.73 | norm 0.4156 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20100 | train 2.855536 | val 3.292327 | tok/s 2623186.41 | norm 0.3772 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20101 | train 3.044344 | val 3.292327 | tok/s 2625257.64 | norm 0.3448 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20102 | train 2.846160 | val 3.292327 | tok/s 2624136.00 | norm 0.3673 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20103 | train 2.830642 | val 3.292327 | tok/s 2627495.67 | norm 0.3881 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20104 | train 2.944248 | val 3.292327 | tok/s 2624686.82 | norm 0.3765 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20105 | train 3.412421 | val 3.292327 | tok/s 2625533.43 | norm 2.2915 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20106 | train 2.874173 | val 3.292327 | tok/s 2627497.42 | norm 0.3919 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20107 | train 2.957445 | val 3.292327 | tok/s 2626995.21 | norm 0.4455 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20108 | train 2.952735 | val 3.292327 | tok/s 2619397.69 | norm 0.3273 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20109 | train 2.870632 | val 3.292327 | tok/s 2619613.98 | norm 0.3627 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20110 | train 2.909570 | val 3.292327 | tok/s 2626333.92 | norm 0.3743 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20111 | train 2.956149 | val 3.292327 | tok/s 2623578.05 | norm 0.3569 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20112 | train 3.062700 | val 3.292327 | tok/s 2623612.85 | norm 0.3694 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20113 | train 2.827063 | val 3.292327 | tok/s 2620969.72 | norm 0.3515 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20114 | train 3.210657 | val 3.292327 | tok/s 2559176.23 | norm 0.4618 | dt 204.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20115 | train 3.155321 | val 3.292327 | tok/s 2624247.46 | norm 0.3807 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20116 | train 3.027309 | val 3.292327 | tok/s 2625130.12 | norm 0.3306 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20117 | train 2.990536 | val 3.292327 | tok/s 2610241.49 | norm 0.3593 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20118 | train 2.889443 | val 3.292327 | tok/s 2626936.93 | norm 0.3858 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20119 | train 2.893215 | val 3.292327 | tok/s 2624670.88 | norm 0.3931 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20120 | train 2.942839 | val 3.292327 | tok/s 2624597.13 | norm 0.4189 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20121 | train 3.024009 | val 3.292327 | tok/s 2501681.99 | norm 0.3960 | dt 209.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20122 | train 2.970543 | val 3.292327 | tok/s 2619418.52 | norm 0.3691 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20123 | train 3.007842 | val 3.292327 | tok/s 2625179.29 | norm 0.3864 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20124 | train 2.854314 | val 3.292327 | tok/s 2625529.95 | norm 0.3571 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20125 | train 2.887369 | val 3.292327 | tok/s 2624546.51 | norm 0.3286 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20126 | train 2.971160 | val 3.292327 | tok/s 2623394.48 | norm 0.3815 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20127 | train 2.953172 | val 3.292327 | tok/s 2596263.24 | norm 0.3514 | dt 201.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20128 | train 2.951704 | val 3.292327 | tok/s 2625728.45 | norm 0.3399 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20129 | train 3.133413 | val 3.292327 | tok/s 2618362.93 | norm 0.4100 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20130 | train 2.987474 | val 3.292327 | tok/s 2624729.74 | norm 0.3473 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20131 | train 2.962014 | val 3.292327 | tok/s 2622302.53 | norm 0.3303 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20132 | train 3.037593 | val 3.292327 | tok/s 2621860.32 | norm 0.3643 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20133 | train 2.893764 | val 3.292327 | tok/s 2624905.87 | norm 0.3583 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20134 | train 2.968008 | val 3.292327 | tok/s 2623826.55 | norm 0.3795 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20135 | train 2.870511 | val 3.292327 | tok/s 2625422.73 | norm 0.3370 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20136 | train 2.957118 | val 3.292327 | tok/s 2626040.82 | norm 0.3604 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20137 | train 2.939457 | val 3.292327 | tok/s 2622267.22 | norm 0.3701 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20138 | train 2.911257 | val 3.292327 | tok/s 2626731.04 | norm 0.3630 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20139 | train 2.883443 | val 3.292327 | tok/s 2619181.07 | norm 0.3845 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20140 | train 2.759719 | val 3.292327 | tok/s 2629399.28 | norm 0.3564 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20141 | train 2.896709 | val 3.292327 | tok/s 2621821.08 | norm 0.3306 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20142 | train 2.911841 | val 3.292327 | tok/s 2623592.37 | norm 0.3456 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20143 | train 3.042228 | val 3.292327 | tok/s 2624599.61 | norm 0.3360 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20144 | train 2.893076 | val 3.292327 | tok/s 2625957.12 | norm 0.3445 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20145 | train 3.021871 | val 3.292327 | tok/s 2624973.92 | norm 0.3713 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20146 | train 3.014614 | val 3.292327 | tok/s 2624428.11 | norm 0.3620 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20147 | train 2.957290 | val 3.292327 | tok/s 2622705.13 | norm 0.3616 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20148 | train 2.976441 | val 3.292327 | tok/s 2624436.91 | norm 0.3623 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20149 | train 2.931194 | val 3.292327 | tok/s 2624743.32 | norm 0.3536 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20150 | train 3.010557 | val 3.292327 | tok/s 2626401.57 | norm 0.3670 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20151 | train 2.972303 | val 3.292327 | tok/s 2620828.52 | norm 0.4347 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20152 | train 2.957585 | val 3.292327 | tok/s 2620975.84 | norm 0.4007 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20153 | train 3.082071 | val 3.292327 | tok/s 2630981.79 | norm 0.4112 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20154 | train 3.121581 | val 3.292327 | tok/s 2620138.70 | norm 0.3773 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20155 | train 2.947615 | val 3.292327 | tok/s 2519469.43 | norm 0.4020 | dt 208.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20156 | train 2.920090 | val 3.292327 | tok/s 2615928.54 | norm 0.4218 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20157 | train 2.974439 | val 3.292327 | tok/s 2627170.62 | norm 0.3863 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20158 | train 2.972759 | val 3.292327 | tok/s 2624382.05 | norm 0.4090 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20159 | train 3.025449 | val 3.292327 | tok/s 2615455.50 | norm 0.4098 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20160 | train 2.965500 | val 3.292327 | tok/s 2629332.22 | norm 0.3320 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20161 | train 2.845236 | val 3.292327 | tok/s 2624877.11 | norm 0.3864 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20162 | train 2.946444 | val 3.292327 | tok/s 2625178.67 | norm 0.4099 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20163 | train 2.920508 | val 3.292327 | tok/s 2623557.11 | norm 0.3895 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20164 | train 2.850986 | val 3.292327 | tok/s 2621904.67 | norm 0.4139 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20165 | train 2.977618 | val 3.292327 | tok/s 2628258.62 | norm 0.3792 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20166 | train 2.994730 | val 3.292327 | tok/s 2628432.66 | norm 0.3474 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20167 | train 2.943355 | val 3.292327 | tok/s 2625592.02 | norm 0.3958 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20168 | train 2.918797 | val 3.292327 | tok/s 2626405.35 | norm 0.3753 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20169 | train 2.912978 | val 3.292327 | tok/s 2620046.58 | norm 0.3698 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20170 | train 2.867630 | val 3.292327 | tok/s 2627449.65 | norm 0.3865 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20171 | train 2.987490 | val 3.292327 | tok/s 2626146.38 | norm 0.4300 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20172 | train 2.945339 | val 3.292327 | tok/s 2625449.93 | norm 0.3929 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20173 | train 2.948683 | val 3.292327 | tok/s 2620550.51 | norm 0.3489 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20174 | train 2.833652 | val 3.292327 | tok/s 2621655.55 | norm 0.3318 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20175 | train 2.882301 | val 3.292327 | tok/s 2624140.15 | norm 0.3677 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20176 | train 2.976209 | val 3.292327 | tok/s 2616540.19 | norm 0.3791 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20177 | train 3.026556 | val 3.292327 | tok/s 2623077.13 | norm 0.3393 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20178 | train 2.962359 | val 3.292327 | tok/s 2626476.33 | norm 0.3166 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20179 | train 2.877275 | val 3.292327 | tok/s 2622144.36 | norm 0.3834 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20180 | train 3.041079 | val 3.292327 | tok/s 2630658.27 | norm 0.4716 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20181 | train 3.012075 | val 3.292327 | tok/s 2626083.52 | norm 0.3824 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20182 | train 2.927465 | val 3.292327 | tok/s 2627517.49 | norm 0.3398 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20183 | train 2.850709 | val 3.292327 | tok/s 2623616.95 | norm 0.3839 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20184 | train 2.967121 | val 3.292327 | tok/s 2618391.20 | norm 0.3777 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20185 | train 3.015777 | val 3.292327 | tok/s 2627136.03 | norm 0.3879 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20186 | train 2.981346 | val 3.292327 | tok/s 2619511.66 | norm 0.3689 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20187 | train 2.945083 | val 3.292327 | tok/s 2624660.81 | norm 0.3852 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20188 | train 3.009327 | val 3.292327 | tok/s 2629444.74 | norm 0.3683 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20189 | train 2.943183 | val 3.292327 | tok/s 2624693.08 | norm 0.3902 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20190 | train 2.982412 | val 3.292327 | tok/s 2627644.17 | norm 0.3981 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20191 | train 2.905982 | val 3.292327 | tok/s 2625152.43 | norm 0.3924 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20192 | train 3.137394 | val 3.292327 | tok/s 2631943.26 | norm 0.4558 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20193 | train 3.021663 | val 3.292327 | tok/s 2621890.18 | norm 0.4859 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20194 | train 2.929737 | val 3.292327 | tok/s 2623836.98 | norm 0.3414 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20195 | train 2.979567 | val 3.292327 | tok/s 2631571.03 | norm 0.3770 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20196 | train 2.980194 | val 3.292327 | tok/s 2615460.19 | norm 0.4093 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20197 | train 2.946672 | val 3.292327 | tok/s 2628719.21 | norm 0.3696 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20198 | train 2.864309 | val 3.292327 | tok/s 2626296.11 | norm 0.4669 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20199 | train 3.037767 | val 3.292327 | tok/s 2623322.82 | norm 0.4121 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20200 | train 2.946007 | val 3.292327 | tok/s 2625540.91 | norm 0.3597 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20201 | train 2.940484 | val 3.292327 | tok/s 2619427.43 | norm 0.3615 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20202 | train 2.906645 | val 3.292327 | tok/s 2629876.24 | norm 0.3862 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20203 | train 3.061194 | val 3.292327 | tok/s 2498919.35 | norm 0.3786 | dt 209.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20204 | train 2.968227 | val 3.292327 | tok/s 2544119.54 | norm 0.3651 | dt 206.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20205 | train 2.967964 | val 3.292327 | tok/s 2625853.48 | norm 0.3408 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20206 | train 3.027094 | val 3.292327 | tok/s 2618814.76 | norm 0.3584 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20207 | train 3.007017 | val 3.292327 | tok/s 2625958.89 | norm 0.3419 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20208 | train 2.848947 | val 3.292327 | tok/s 2622958.23 | norm 0.3617 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20209 | train 2.942602 | val 3.292327 | tok/s 2623366.57 | norm 0.3798 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20210 | train 2.875991 | val 3.292327 | tok/s 2623740.46 | norm 0.3245 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20211 | train 2.920132 | val 3.292327 | tok/s 2619325.34 | norm 0.3395 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20212 | train 2.913516 | val 3.292327 | tok/s 2628145.12 | norm 0.3613 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20213 | train 2.999064 | val 3.292327 | tok/s 2624817.06 | norm 0.3489 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20214 | train 2.893367 | val 3.292327 | tok/s 2611756.14 | norm 0.3465 | dt 200.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20215 | train 3.042844 | val 3.292327 | tok/s 2617167.83 | norm 0.3204 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20216 | train 2.827720 | val 3.292327 | tok/s 2622775.01 | norm 0.3724 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20217 | train 2.900847 | val 3.292327 | tok/s 2625183.10 | norm 0.3639 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20218 | train 2.898059 | val 3.292327 | tok/s 2624719.86 | norm 0.3615 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20219 | train 2.955311 | val 3.292327 | tok/s 2626871.61 | norm 0.3812 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20220 | train 2.952688 | val 3.292327 | tok/s 2621168.83 | norm 0.3868 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20221 | train 2.853221 | val 3.292327 | tok/s 2618485.88 | norm 0.3923 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20222 | train 2.970089 | val 3.292327 | tok/s 2624116.13 | norm 0.3852 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20223 | train 3.051047 | val 3.292327 | tok/s 2621286.56 | norm 0.3436 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20224 | train 2.951504 | val 3.292327 | tok/s 2626486.81 | norm 0.3631 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20225 | train 2.982927 | val 3.292327 | tok/s 2623701.91 | norm 0.3759 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20226 | train 2.902267 | val 3.292327 | tok/s 2623441.12 | norm 0.3690 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20227 | train 2.987963 | val 3.292327 | tok/s 2629171.13 | norm 0.3741 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20228 | train 3.031828 | val 3.292327 | tok/s 2625025.11 | norm 0.4078 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20229 | train 2.908533 | val 3.292327 | tok/s 2629191.28 | norm 0.4029 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20230 | train 3.131272 | val 3.292327 | tok/s 2626917.48 | norm 0.4117 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20231 | train 2.845051 | val 3.292327 | tok/s 2626587.24 | norm 0.3948 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20232 | train 2.848231 | val 3.292327 | tok/s 2626765.75 | norm 0.3570 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20233 | train 2.870332 | val 3.292327 | tok/s 2623927.77 | norm 0.3797 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20234 | train 2.872780 | val 3.292327 | tok/s 2627101.30 | norm 0.3915 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20235 | train 2.957331 | val 3.292327 | tok/s 2626520.74 | norm 0.3751 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20236 | train 2.882429 | val 3.292327 | tok/s 2624856.13 | norm 0.3863 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20237 | train 2.969745 | val 3.292327 | tok/s 2626191.49 | norm 0.3639 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20238 | train 3.005725 | val 3.292327 | tok/s 2617799.22 | norm 0.3820 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20239 | train 2.971744 | val 3.292327 | tok/s 2626837.40 | norm 0.3668 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20240 | train 2.858261 | val 3.292327 | tok/s 2621920.35 | norm 0.3255 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20241 | train 3.044990 | val 3.292327 | tok/s 2620834.89 | norm 0.3507 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20242 | train 3.020043 | val 3.292327 | tok/s 2632181.99 | norm 0.3817 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20243 | train 3.002070 | val 3.292327 | tok/s 2625952.61 | norm 0.3933 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20244 | train 2.918218 | val 3.292327 | tok/s 2625259.90 | norm 0.3543 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20245 | train 2.905317 | val 3.292327 | tok/s 2622437.36 | norm 0.3779 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20246 | train 2.995790 | val 3.292327 | tok/s 2623721.79 | norm 0.3926 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20247 | train 2.921362 | val 3.292327 | tok/s 2625549.93 | norm 0.3506 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20248 | train 2.868637 | val 3.292327 | tok/s 2618882.54 | norm 0.3324 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20249 | train 2.858687 | val 3.292327 | tok/s 2627567.72 | norm 0.3673 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 20250 | val_loss 3.297417 step 20250 | train 2.922544 | val 3.297417 | tok/s 2604383.48 | norm 0.3734 | dt 201.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20251 | train 2.930132 | val 3.297417 | tok/s 2624764.71 | norm 0.3690 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20252 | train 2.814491 | val 3.297417 | tok/s 2624599.30 | norm 0.3762 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20253 | train 2.940370 | val 3.297417 | tok/s 2618508.72 | norm 0.3693 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20254 | train 2.901318 | val 3.297417 | tok/s 2618421.58 | norm 0.3270 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20255 | train 2.827772 | val 3.297417 | tok/s 2622084.04 | norm 0.3548 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20256 | train 2.893158 | val 3.297417 | tok/s 2621989.03 | norm 0.3729 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20257 | train 2.937378 | val 3.297417 | tok/s 2626828.69 | norm 0.3659 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20258 | train 3.112230 | val 3.297417 | tok/s 2619162.08 | norm 0.3895 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20259 | train 2.990371 | val 3.297417 | tok/s 2615851.04 | norm 0.4130 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20260 | train 3.088129 | val 3.297417 | tok/s 2629382.42 | norm 0.3843 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20261 | train 2.917355 | val 3.297417 | tok/s 2624161.39 | norm 0.3769 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20262 | train 2.919464 | val 3.297417 | tok/s 2625727.46 | norm 0.4029 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20263 | train 3.138576 | val 3.297417 | tok/s 2628277.45 | norm 0.4819 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20264 | train 3.012034 | val 3.297417 | tok/s 2623435.24 | norm 0.4034 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20265 | train 2.967062 | val 3.297417 | tok/s 2629356.93 | norm 0.3059 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20266 | train 2.931195 | val 3.297417 | tok/s 2629506.30 | norm 0.3838 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20267 | train 2.971689 | val 3.297417 | tok/s 2627564.38 | norm 0.3856 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20268 | train 2.916548 | val 3.297417 | tok/s 2624407.52 | norm 0.3362 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20269 | train 2.903887 | val 3.297417 | tok/s 2623074.64 | norm 0.3349 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20270 | train 2.846462 | val 3.297417 | tok/s 2625656.67 | norm 0.3604 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20271 | train 2.974207 | val 3.297417 | tok/s 2627119.35 | norm 0.3474 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20272 | train 2.955308 | val 3.297417 | tok/s 2627875.73 | norm 0.3523 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20273 | train 3.074652 | val 3.297417 | tok/s 2625682.83 | norm 0.4043 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20274 | train 2.964078 | val 3.297417 | tok/s 2624551.35 | norm 0.3313 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20275 | train 2.965514 | val 3.297417 | tok/s 2628837.19 | norm 0.3423 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20276 | train 2.984838 | val 3.297417 | tok/s 2625833.60 | norm 0.3956 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20277 | train 2.816396 | val 3.297417 | tok/s 2624140.10 | norm 0.3697 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20278 | train 2.838008 | val 3.297417 | tok/s 2626619.55 | norm 0.3579 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20279 | train 2.988689 | val 3.297417 | tok/s 2622030.52 | norm 0.3801 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20280 | train 2.971501 | val 3.297417 | tok/s 2628686.72 | norm 0.3795 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20281 | train 3.061876 | val 3.297417 | tok/s 2625235.59 | norm 0.3631 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20282 | train 2.846010 | val 3.297417 | tok/s 2628892.87 | norm 0.3344 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20283 | train 2.946405 | val 3.297417 | tok/s 2626172.48 | norm 0.3613 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20284 | train 3.101136 | val 3.297417 | tok/s 2625528.25 | norm 0.3772 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20285 | train 2.991550 | val 3.297417 | tok/s 2627513.67 | norm 0.3636 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20286 | train 2.929955 | val 3.297417 | tok/s 2622851.08 | norm 0.3592 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20287 | train 3.022763 | val 3.297417 | tok/s 2625979.11 | norm 0.3696 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20288 | train 2.950222 | val 3.297417 | tok/s 2623748.65 | norm 0.3770 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20289 | train 3.009574 | val 3.297417 | tok/s 2627181.21 | norm 0.3588 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20290 | train 3.111113 | val 3.297417 | tok/s 2623893.10 | norm 0.3944 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20291 | train 2.896654 | val 3.297417 | tok/s 2618400.05 | norm 0.3801 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20292 | train 3.044648 | val 3.297417 | tok/s 2627182.68 | norm 0.3725 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20293 | train 3.079143 | val 3.297417 | tok/s 2626278.62 | norm 0.4623 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20294 | train 3.097068 | val 3.297417 | tok/s 2623011.79 | norm 0.3912 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20295 | train 2.903648 | val 3.297417 | tok/s 2626244.45 | norm 0.3424 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20296 | train 2.877246 | val 3.297417 | tok/s 2619687.68 | norm 0.3731 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20297 | train 2.977848 | val 3.297417 | tok/s 2627078.91 | norm 0.3665 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20298 | train 3.039696 | val 3.297417 | tok/s 2620142.02 | norm 0.3679 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20299 | train 2.878677 | val 3.297417 | tok/s 2622912.18 | norm 0.3680 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20300 | train 3.003136 | val 3.297417 | tok/s 2617799.95 | norm 0.3374 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20301 | train 2.905599 | val 3.297417 | tok/s 2618417.21 | norm 0.3557 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20302 | train 3.069411 | val 3.297417 | tok/s 2622250.16 | norm 0.4260 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20303 | train 2.857157 | val 3.297417 | tok/s 2621476.82 | norm 0.3831 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20304 | train 3.031304 | val 3.297417 | tok/s 2626561.72 | norm 0.3384 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20305 | train 2.927431 | val 3.297417 | tok/s 2623499.97 | norm 0.3749 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20306 | train 2.986364 | val 3.297417 | tok/s 2623458.19 | norm 0.4018 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20307 | train 3.110517 | val 3.297417 | tok/s 2625236.30 | norm 0.3759 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20308 | train 2.912478 | val 3.297417 | tok/s 2628332.90 | norm 0.3644 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20309 | train 2.984085 | val 3.297417 | tok/s 2624819.14 | norm 0.3793 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20310 | train 2.914551 | val 3.297417 | tok/s 2622233.93 | norm 0.3721 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20311 | train 2.959919 | val 3.297417 | tok/s 2620399.60 | norm 0.3632 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20312 | train 3.025187 | val 3.297417 | tok/s 2600745.57 | norm 0.3688 | dt 201.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20313 | train 2.873530 | val 3.297417 | tok/s 2626612.27 | norm 0.3671 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20314 | train 3.194470 | val 3.297417 | tok/s 2630051.62 | norm 0.4416 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20315 | train 2.928988 | val 3.297417 | tok/s 2630102.43 | norm 0.3463 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20316 | train 2.976351 | val 3.297417 | tok/s 2610241.31 | norm 0.3379 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20317 | train 2.852440 | val 3.297417 | tok/s 2630147.59 | norm 0.3758 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20318 | train 3.022858 | val 3.297417 | tok/s 2631439.21 | norm 0.3608 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20319 | train 2.913364 | val 3.297417 | tok/s 2631505.07 | norm 0.3608 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20320 | train 3.040237 | val 3.297417 | tok/s 2630286.94 | norm 0.3665 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20321 | train 2.904242 | val 3.297417 | tok/s 2623560.81 | norm 0.3520 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20322 | train 2.899714 | val 3.297417 | tok/s 2624829.76 | norm 0.3561 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20323 | train 2.863489 | val 3.297417 | tok/s 2628707.80 | norm 0.4111 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20324 | train 2.968824 | val 3.297417 | tok/s 2630069.39 | norm 0.3685 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20325 | train 3.037137 | val 3.297417 | tok/s 2625088.98 | norm 0.3682 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20326 | train 2.960427 | val 3.297417 | tok/s 2625753.90 | norm 0.3742 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20327 | train 2.858056 | val 3.297417 | tok/s 2628025.49 | norm 0.3935 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20328 | train 2.867433 | val 3.297417 | tok/s 2630047.90 | norm 0.3859 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20329 | train 3.001698 | val 3.297417 | tok/s 2629500.88 | norm 0.3790 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20330 | train 2.853135 | val 3.297417 | tok/s 2625496.27 | norm 0.3599 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20331 | train 2.921162 | val 3.297417 | tok/s 2625366.06 | norm 0.4253 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20332 | train 2.976797 | val 3.297417 | tok/s 2630491.81 | norm 0.4279 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20333 | train 2.896309 | val 3.297417 | tok/s 2627360.97 | norm 0.4085 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20334 | train 3.048528 | val 3.297417 | tok/s 2627418.77 | norm 0.4097 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20335 | train 2.918423 | val 3.297417 | tok/s 2629319.32 | norm 0.3718 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20336 | train 2.865175 | val 3.297417 | tok/s 2629070.21 | norm 0.3713 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20337 | train 2.938749 | val 3.297417 | tok/s 2631978.43 | norm 0.4028 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20338 | train 2.901927 | val 3.297417 | tok/s 2485362.12 | norm 0.3496 | dt 210.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20339 | train 2.974794 | val 3.297417 | tok/s 2619869.60 | norm 0.3483 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20340 | train 2.917494 | val 3.297417 | tok/s 2627485.98 | norm 0.3571 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20341 | train 2.875340 | val 3.297417 | tok/s 2629220.22 | norm 0.3494 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20342 | train 3.122113 | val 3.297417 | tok/s 2623081.26 | norm 0.3746 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20343 | train 2.883592 | val 3.297417 | tok/s 2618693.50 | norm 0.3536 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20344 | train 2.882521 | val 3.297417 | tok/s 2628008.89 | norm 0.3306 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20345 | train 3.016805 | val 3.297417 | tok/s 2556130.12 | norm 0.3810 | dt 205.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20346 | train 3.011796 | val 3.297417 | tok/s 2587498.64 | norm 0.3718 | dt 202.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20347 | train 3.023544 | val 3.297417 | tok/s 2619112.41 | norm 0.3452 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20348 | train 2.883792 | val 3.297417 | tok/s 2606168.57 | norm 0.3598 | dt 201.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20349 | train 2.977873 | val 3.297417 | tok/s 2627291.93 | norm 0.3811 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20350 | train 2.936370 | val 3.297417 | tok/s 2627469.39 | norm 0.3720 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20351 | train 2.870410 | val 3.297417 | tok/s 2436208.95 | norm 0.3388 | dt 215.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20352 | train 3.042478 | val 3.297417 | tok/s 2622874.07 | norm 0.3569 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20353 | train 2.984868 | val 3.297417 | tok/s 2626222.52 | norm 0.3787 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20354 | train 2.897345 | val 3.297417 | tok/s 2628432.38 | norm 0.3669 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20355 | train 2.803515 | val 3.297417 | tok/s 2623967.76 | norm 0.3772 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20356 | train 2.852472 | val 3.297417 | tok/s 2627912.18 | norm 0.3351 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20357 | train 2.972610 | val 3.297417 | tok/s 2629289.92 | norm 0.3611 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20358 | train 3.180741 | val 3.297417 | tok/s 2623655.89 | norm 0.7922 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20359 | train 2.829364 | val 3.297417 | tok/s 2592811.18 | norm 0.3449 | dt 202.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20360 | train 3.026245 | val 3.297417 | tok/s 2626561.51 | norm 0.3448 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20361 | train 3.159788 | val 3.297417 | tok/s 2629878.15 | norm 0.4145 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20362 | train 2.850847 | val 3.297417 | tok/s 2626309.51 | norm 0.3573 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20363 | train 2.874025 | val 3.297417 | tok/s 2624657.15 | norm 0.3575 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20364 | train 2.966089 | val 3.297417 | tok/s 2628557.36 | norm 0.3784 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20365 | train 2.934316 | val 3.297417 | tok/s 2626204.46 | norm 0.3681 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20366 | train 3.077000 | val 3.297417 | tok/s 2629242.00 | norm 0.3904 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20367 | train 3.008841 | val 3.297417 | tok/s 2626380.42 | norm 0.3754 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20368 | train 2.978178 | val 3.297417 | tok/s 2626389.72 | norm 0.3939 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20369 | train 3.014616 | val 3.297417 | tok/s 2628605.35 | norm 0.3773 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20370 | train 2.980589 | val 3.297417 | tok/s 2628291.72 | norm 0.3426 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20371 | train 2.896485 | val 3.297417 | tok/s 2629920.56 | norm 0.3590 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20372 | train 2.950375 | val 3.297417 | tok/s 2627023.88 | norm 0.3560 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20373 | train 2.950118 | val 3.297417 | tok/s 2626951.52 | norm 0.3215 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20374 | train 2.907528 | val 3.297417 | tok/s 2629276.22 | norm 0.3697 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20375 | train 2.904693 | val 3.297417 | tok/s 2630774.54 | norm 0.3812 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20376 | train 3.009552 | val 3.297417 | tok/s 2628421.32 | norm 0.3745 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20377 | train 2.870139 | val 3.297417 | tok/s 2623740.26 | norm 0.3794 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20378 | train 3.019500 | val 3.297417 | tok/s 2626427.48 | norm 0.3815 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20379 | train 2.991198 | val 3.297417 | tok/s 2628026.12 | norm 0.3545 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20380 | train 3.047230 | val 3.297417 | tok/s 2615863.69 | norm 0.3886 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20381 | train 2.942698 | val 3.297417 | tok/s 2631718.02 | norm 0.3929 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20382 | train 2.926756 | val 3.297417 | tok/s 2630745.45 | norm 0.3674 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20383 | train 2.848037 | val 3.297417 | tok/s 2630628.15 | norm 0.3659 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20384 | train 2.947738 | val 3.297417 | tok/s 2624974.20 | norm 0.3739 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20385 | train 2.978197 | val 3.297417 | tok/s 2620448.75 | norm 0.3577 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20386 | train 3.036938 | val 3.297417 | tok/s 2629095.36 | norm 0.3542 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20387 | train 2.957161 | val 3.297417 | tok/s 2627737.18 | norm 0.3773 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20388 | train 2.823617 | val 3.297417 | tok/s 2612397.03 | norm 0.3779 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20389 | train 3.225690 | val 3.297417 | tok/s 2552662.71 | norm 0.3603 | dt 205.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20390 | train 2.916517 | val 3.297417 | tok/s 2617919.36 | norm 0.3897 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20391 | train 2.916522 | val 3.297417 | tok/s 2628994.90 | norm 0.3599 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20392 | train 2.904489 | val 3.297417 | tok/s 2626712.81 | norm 0.3462 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20393 | train 2.889961 | val 3.297417 | tok/s 2624767.17 | norm 0.4060 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20394 | train 2.859618 | val 3.297417 | tok/s 2619903.14 | norm 0.4411 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20395 | train 2.972900 | val 3.297417 | tok/s 2622972.38 | norm 0.4040 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20396 | train 2.913892 | val 3.297417 | tok/s 2627545.51 | norm 0.3711 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20397 | train 3.003001 | val 3.297417 | tok/s 2621476.70 | norm 0.4110 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20398 | train 2.894005 | val 3.297417 | tok/s 2625269.28 | norm 0.3869 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20399 | train 2.905432 | val 3.297417 | tok/s 2616576.78 | norm 0.3481 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20400 | train 2.882392 | val 3.297417 | tok/s 2621073.94 | norm 0.3610 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20401 | train 3.083714 | val 3.297417 | tok/s 2628207.82 | norm 0.3534 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20402 | train 2.997443 | val 3.297417 | tok/s 2628304.46 | norm 0.4182 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20403 | train 2.909432 | val 3.297417 | tok/s 2627696.36 | norm 0.3853 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20404 | train 2.952185 | val 3.297417 | tok/s 2626272.50 | norm 0.3380 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20405 | train 3.181177 | val 3.297417 | tok/s 2623943.87 | norm 0.5650 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20406 | train 2.878706 | val 3.297417 | tok/s 2628878.71 | norm 0.3613 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20407 | train 3.028426 | val 3.297417 | tok/s 2626180.11 | norm 0.3668 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20408 | train 3.178900 | val 3.297417 | tok/s 2631704.69 | norm 0.4295 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20409 | train 3.065797 | val 3.297417 | tok/s 2623932.57 | norm 0.3435 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20410 | train 3.032878 | val 3.297417 | tok/s 2625826.14 | norm 0.3496 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20411 | train 2.933090 | val 3.297417 | tok/s 2630926.70 | norm 0.3820 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20412 | train 3.075689 | val 3.297417 | tok/s 2628764.13 | norm 0.3630 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20413 | train 2.879341 | val 3.297417 | tok/s 2631392.66 | norm 0.3591 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20414 | train 2.978029 | val 3.297417 | tok/s 2620517.88 | norm 0.3836 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20415 | train 2.950025 | val 3.297417 | tok/s 2626514.52 | norm 0.3859 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20416 | train 3.022938 | val 3.297417 | tok/s 2630398.20 | norm 0.3608 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20417 | train 2.953651 | val 3.297417 | tok/s 2626947.63 | norm 0.3530 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20418 | train 2.917090 | val 3.297417 | tok/s 2628236.38 | norm 0.3383 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20419 | train 2.941146 | val 3.297417 | tok/s 2624232.77 | norm 0.3389 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20420 | train 2.907360 | val 3.297417 | tok/s 2625306.76 | norm 0.3409 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20421 | train 2.820968 | val 3.297417 | tok/s 2633585.46 | norm 0.3678 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20422 | train 2.888813 | val 3.297417 | tok/s 2629930.89 | norm 0.3562 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20423 | train 3.022729 | val 3.297417 | tok/s 2630230.11 | norm 0.3531 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20424 | train 2.904557 | val 3.297417 | tok/s 2623955.74 | norm 0.3363 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20425 | train 2.919346 | val 3.297417 | tok/s 2626416.72 | norm 0.3572 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20426 | train 2.929827 | val 3.297417 | tok/s 2626885.13 | norm 0.3455 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20427 | train 2.946112 | val 3.297417 | tok/s 2622982.20 | norm 0.3223 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20428 | train 2.986985 | val 3.297417 | tok/s 2629592.90 | norm 0.3294 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20429 | train 2.972106 | val 3.297417 | tok/s 2625923.44 | norm 0.3284 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20430 | train 3.082182 | val 3.297417 | tok/s 2626850.59 | norm 0.3463 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20431 | train 2.941018 | val 3.297417 | tok/s 2621505.05 | norm 0.4084 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20432 | train 3.005967 | val 3.297417 | tok/s 2616311.11 | norm 0.4080 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20433 | train 2.889149 | val 3.297417 | tok/s 2633182.04 | norm 0.3369 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20434 | train 2.981579 | val 3.297417 | tok/s 2629304.01 | norm 0.3475 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20435 | train 2.853602 | val 3.297417 | tok/s 2624264.31 | norm 0.3834 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20436 | train 3.028920 | val 3.297417 | tok/s 2625234.29 | norm 0.4034 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20437 | train 2.951034 | val 3.297417 | tok/s 2618741.49 | norm 0.3935 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20438 | train 3.014690 | val 3.297417 | tok/s 2626164.17 | norm 0.3812 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20439 | train 2.935725 | val 3.297417 | tok/s 2627313.33 | norm 0.3416 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20440 | train 3.035870 | val 3.297417 | tok/s 2626560.24 | norm 0.3573 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20441 | train 2.962410 | val 3.297417 | tok/s 2630018.96 | norm 0.3814 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20442 | train 2.893798 | val 3.297417 | tok/s 2533860.96 | norm 0.3655 | dt 206.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20443 | train 2.948315 | val 3.297417 | tok/s 2625981.32 | norm 0.3491 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20444 | train 2.978704 | val 3.297417 | tok/s 2625350.85 | norm 0.3786 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20445 | train 2.884666 | val 3.297417 | tok/s 2629405.68 | norm 0.3606 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20446 | train 2.825170 | val 3.297417 | tok/s 2618875.06 | norm 0.3604 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20447 | train 2.991599 | val 3.297417 | tok/s 2620000.57 | norm 0.4024 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20448 | train 2.968949 | val 3.297417 | tok/s 2627543.94 | norm 0.4080 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20449 | train 2.896416 | val 3.297417 | tok/s 2626258.91 | norm 0.3950 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20450 | train 2.852416 | val 3.297417 | tok/s 2625495.84 | norm 0.3882 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20451 | train 2.869341 | val 3.297417 | tok/s 2619879.35 | norm 0.3774 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20452 | train 3.195478 | val 3.297417 | tok/s 2623023.97 | norm 0.4189 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20453 | train 3.047672 | val 3.297417 | tok/s 2629786.90 | norm 0.3697 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20454 | train 2.958853 | val 3.297417 | tok/s 2627593.67 | norm 0.3130 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20455 | train 2.901389 | val 3.297417 | tok/s 2629631.51 | norm 0.3441 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20456 | train 2.987693 | val 3.297417 | tok/s 2584862.78 | norm 0.3641 | dt 202.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20457 | train 2.944485 | val 3.297417 | tok/s 2626664.25 | norm 0.3613 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20458 | train 2.942615 | val 3.297417 | tok/s 2631387.86 | norm 0.3958 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20459 | train 3.176546 | val 3.297417 | tok/s 2624041.36 | norm 0.3553 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20460 | train 2.871738 | val 3.297417 | tok/s 2625891.28 | norm 0.3756 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20461 | train 3.034088 | val 3.297417 | tok/s 2630992.45 | norm 0.3996 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20462 | train 2.972871 | val 3.297417 | tok/s 2625538.73 | norm 0.3581 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20463 | train 2.959316 | val 3.297417 | tok/s 2630218.38 | norm 0.3459 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20464 | train 2.922391 | val 3.297417 | tok/s 2627027.85 | norm 0.3987 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20465 | train 2.951349 | val 3.297417 | tok/s 2630970.78 | norm 0.3658 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20466 | train 2.961978 | val 3.297417 | tok/s 2627981.72 | norm 0.3560 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20467 | train 3.050934 | val 3.297417 | tok/s 2627745.26 | norm 0.3816 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20468 | train 2.898395 | val 3.297417 | tok/s 2632683.67 | norm 0.3781 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20469 | train 2.882061 | val 3.297417 | tok/s 2625410.51 | norm 0.3617 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20470 | train 2.978486 | val 3.297417 | tok/s 2628011.43 | norm 0.3520 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20471 | train 2.968248 | val 3.297417 | tok/s 2618970.02 | norm 0.3482 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20472 | train 2.884264 | val 3.297417 | tok/s 2626383.98 | norm 0.3760 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20473 | train 2.923717 | val 3.297417 | tok/s 2632420.63 | norm 0.3992 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20474 | train 3.239473 | val 3.297417 | tok/s 2619540.85 | norm 0.4101 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20475 | train 2.903520 | val 3.297417 | tok/s 2623762.61 | norm 0.4240 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20476 | train 2.952550 | val 3.297417 | tok/s 2620377.24 | norm 0.3775 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20477 | train 2.912544 | val 3.297417 | tok/s 2625139.66 | norm 0.3582 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20478 | train 3.147913 | val 3.297417 | tok/s 2623991.45 | norm 0.3784 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20479 | train 2.940928 | val 3.297417 | tok/s 2622370.61 | norm 0.3765 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20480 | train 2.873324 | val 3.297417 | tok/s 2633785.99 | norm 0.3699 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20481 | train 2.983380 | val 3.297417 | tok/s 2627033.53 | norm 0.4072 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20482 | train 2.806631 | val 3.297417 | tok/s 2629557.48 | norm 0.4395 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20483 | train 2.988568 | val 3.297417 | tok/s 2628647.03 | norm 0.3172 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20484 | train 3.056831 | val 3.297417 | tok/s 2620379.36 | norm 0.3156 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20485 | train 2.925183 | val 3.297417 | tok/s 2627922.22 | norm 0.3659 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20486 | train 2.942619 | val 3.297417 | tok/s 2627638.18 | norm 0.3967 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20487 | train 2.987932 | val 3.297417 | tok/s 2629027.52 | norm 0.3668 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20488 | train 2.889176 | val 3.297417 | tok/s 2622868.43 | norm 0.3713 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20489 | train 2.938639 | val 3.297417 | tok/s 2623923.57 | norm 0.3811 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20490 | train 2.840446 | val 3.297417 | tok/s 2480686.09 | norm 0.3634 | dt 211.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20491 | train 2.932550 | val 3.297417 | tok/s 2625680.71 | norm 0.4126 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20492 | train 2.948320 | val 3.297417 | tok/s 2631857.71 | norm 0.3658 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20493 | train 2.980116 | val 3.297417 | tok/s 2626057.58 | norm 0.3640 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20494 | train 2.889177 | val 3.297417 | tok/s 2625100.87 | norm 0.4075 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20495 | train 2.872312 | val 3.297417 | tok/s 2633517.76 | norm 0.3854 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20496 | train 3.031621 | val 3.297417 | tok/s 2627611.84 | norm 0.4116 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20497 | train 2.972067 | val 3.297417 | tok/s 2627860.67 | norm 0.3755 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20498 | train 3.045619 | val 3.297417 | tok/s 2623700.73 | norm 0.3509 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20499 | train 2.964816 | val 3.297417 | tok/s 2621429.06 | norm 0.3762 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 20500 | val_loss 3.298592 step 20500 | train 2.795446 | val 3.298592 | tok/s 2599230.34 | norm 0.3846 | dt 201.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20501 | train 2.978981 | val 3.298592 | tok/s 2607748.36 | norm 0.3727 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20502 | train 2.935799 | val 3.298592 | tok/s 2627593.32 | norm 0.3695 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20503 | train 3.079120 | val 3.298592 | tok/s 2625484.09 | norm 0.3566 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20504 | train 2.986747 | val 3.298592 | tok/s 2627218.49 | norm 0.3610 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20505 | train 2.972868 | val 3.298592 | tok/s 2620983.84 | norm 0.3861 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20506 | train 2.933855 | val 3.298592 | tok/s 2632421.20 | norm 0.3724 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20507 | train 2.950020 | val 3.298592 | tok/s 2625543.45 | norm 0.3715 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20508 | train 2.986659 | val 3.298592 | tok/s 2625897.08 | norm 0.3640 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20509 | train 2.989910 | val 3.298592 | tok/s 2627456.99 | norm 0.3745 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20510 | train 3.109476 | val 3.298592 | tok/s 2628348.61 | norm 0.3809 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20511 | train 2.997317 | val 3.298592 | tok/s 2624626.09 | norm 0.3462 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20512 | train 2.939338 | val 3.298592 | tok/s 2624365.50 | norm 0.3774 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20513 | train 2.938245 | val 3.298592 | tok/s 2627680.60 | norm 0.3637 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20514 | train 2.988256 | val 3.298592 | tok/s 2630047.26 | norm 0.3386 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20515 | train 2.892064 | val 3.298592 | tok/s 2627126.55 | norm 0.3963 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20516 | train 3.054038 | val 3.298592 | tok/s 2633082.01 | norm 0.3952 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20517 | train 2.926708 | val 3.298592 | tok/s 2623831.89 | norm 0.3714 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20518 | train 2.935807 | val 3.298592 | tok/s 2634196.48 | norm 0.3768 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20519 | train 2.947889 | val 3.298592 | tok/s 2624160.44 | norm 0.4026 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20520 | train 2.846282 | val 3.298592 | tok/s 2624693.92 | norm 0.3961 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20521 | train 2.949569 | val 3.298592 | tok/s 2622427.54 | norm 0.3536 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20522 | train 3.034635 | val 3.298592 | tok/s 2613128.40 | norm 0.4173 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20523 | train 2.958488 | val 3.298592 | tok/s 2631999.60 | norm 0.3342 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20524 | train 3.178713 | val 3.298592 | tok/s 2628115.97 | norm 0.3874 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20525 | train 2.965517 | val 3.298592 | tok/s 2630689.53 | norm 0.3704 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20526 | train 2.892348 | val 3.298592 | tok/s 2625768.63 | norm 0.3388 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20527 | train 3.274868 | val 3.298592 | tok/s 2624590.10 | norm 0.3991 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20528 | train 2.949803 | val 3.298592 | tok/s 2629643.70 | norm 0.3431 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20529 | train 2.902303 | val 3.298592 | tok/s 2628626.81 | norm 0.3888 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20530 | train 3.015307 | val 3.298592 | tok/s 2627902.84 | norm 0.3948 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20531 | train 2.941859 | val 3.298592 | tok/s 2621734.24 | norm 0.3576 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20532 | train 2.897703 | val 3.298592 | tok/s 2627269.65 | norm 0.3768 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20533 | train 3.093467 | val 3.298592 | tok/s 2633158.75 | norm 0.4085 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20534 | train 2.893704 | val 3.298592 | tok/s 2623091.66 | norm 0.3784 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20535 | train 2.914088 | val 3.298592 | tok/s 2573854.60 | norm 0.3580 | dt 203.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20536 | train 2.912569 | val 3.298592 | tok/s 2445072.24 | norm 0.3611 | dt 214.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20537 | train 3.062784 | val 3.298592 | tok/s 2622896.78 | norm 0.3731 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20538 | train 3.016831 | val 3.298592 | tok/s 2628419.40 | norm 0.3417 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20539 | train 3.000165 | val 3.298592 | tok/s 2631288.86 | norm 0.3353 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20540 | train 3.010602 | val 3.298592 | tok/s 2632075.49 | norm 0.3655 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20541 | train 2.920198 | val 3.298592 | tok/s 2624478.56 | norm 0.3573 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20542 | train 2.918463 | val 3.298592 | tok/s 2621452.98 | norm 0.3636 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20543 | train 2.985925 | val 3.298592 | tok/s 2627542.05 | norm 0.3580 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20544 | train 3.023863 | val 3.298592 | tok/s 2628151.59 | norm 0.3807 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20545 | train 2.927584 | val 3.298592 | tok/s 2628851.31 | norm 0.4040 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20546 | train 2.918274 | val 3.298592 | tok/s 2630858.67 | norm 0.3808 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20547 | train 2.949243 | val 3.298592 | tok/s 2628030.61 | norm 0.3452 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20548 | train 2.889576 | val 3.298592 | tok/s 2629450.98 | norm 0.3814 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20549 | train 2.837600 | val 3.298592 | tok/s 2631078.51 | norm 0.4005 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20550 | train 2.999780 | val 3.298592 | tok/s 2630476.95 | norm 0.3843 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20551 | train 2.943684 | val 3.298592 | tok/s 2630234.44 | norm 0.3702 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20552 | train 2.974441 | val 3.298592 | tok/s 2626203.22 | norm 0.3515 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20553 | train 3.080787 | val 3.298592 | tok/s 2627060.60 | norm 0.3606 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20554 | train 2.850573 | val 3.298592 | tok/s 2624086.98 | norm 0.3506 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20555 | train 2.911351 | val 3.298592 | tok/s 2625501.86 | norm 0.3728 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20556 | train 3.050563 | val 3.298592 | tok/s 2627696.14 | norm 0.4053 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20557 | train 3.047504 | val 3.298592 | tok/s 2628723.56 | norm 0.3433 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20558 | train 2.994526 | val 3.298592 | tok/s 2633075.05 | norm 0.3463 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20559 | train 2.894486 | val 3.298592 | tok/s 2623267.22 | norm 0.3687 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20560 | train 3.099111 | val 3.298592 | tok/s 2629377.97 | norm 0.3908 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20561 | train 2.900894 | val 3.298592 | tok/s 2629795.50 | norm 0.4188 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20562 | train 2.903697 | val 3.298592 | tok/s 2622937.23 | norm 0.3886 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20563 | train 3.023693 | val 3.298592 | tok/s 2625571.89 | norm 0.3657 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20564 | train 2.901972 | val 3.298592 | tok/s 2623150.27 | norm 0.3466 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20565 | train 2.851659 | val 3.298592 | tok/s 2632006.90 | norm 0.3713 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20566 | train 2.980569 | val 3.298592 | tok/s 2626076.54 | norm 0.3966 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20567 | train 3.050076 | val 3.298592 | tok/s 2627843.06 | norm 0.3684 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20568 | train 2.940306 | val 3.298592 | tok/s 2630482.01 | norm 0.3433 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20569 | train 3.027115 | val 3.298592 | tok/s 2623869.19 | norm 0.3826 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20570 | train 2.946906 | val 3.298592 | tok/s 2631054.57 | norm 0.3745 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20571 | train 2.926701 | val 3.298592 | tok/s 2629027.37 | norm 0.3444 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20572 | train 2.912654 | val 3.298592 | tok/s 2626958.06 | norm 0.3767 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20573 | train 3.045270 | val 3.298592 | tok/s 2630275.04 | norm 0.3593 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20574 | train 3.057065 | val 3.298592 | tok/s 2619488.62 | norm 0.3665 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20575 | train 2.849077 | val 3.298592 | tok/s 2538466.11 | norm 0.3451 | dt 206.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20576 | train 3.007275 | val 3.298592 | tok/s 2626395.17 | norm 0.3257 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20577 | train 2.967224 | val 3.298592 | tok/s 2626113.37 | norm 0.3571 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20578 | train 2.824936 | val 3.298592 | tok/s 2620027.57 | norm 0.3948 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20579 | train 2.959147 | val 3.298592 | tok/s 2624199.00 | norm 0.3480 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20580 | train 2.937829 | val 3.298592 | tok/s 2627194.59 | norm 0.3577 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20581 | train 2.944702 | val 3.298592 | tok/s 2626135.27 | norm 0.4048 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20582 | train 2.988806 | val 3.298592 | tok/s 2632182.57 | norm 0.3819 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20583 | train 3.070219 | val 3.298592 | tok/s 2626092.17 | norm 0.4110 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20584 | train 2.921528 | val 3.298592 | tok/s 2625303.97 | norm 0.3827 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20585 | train 2.981588 | val 3.298592 | tok/s 2631156.70 | norm 0.3555 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20586 | train 2.805161 | val 3.298592 | tok/s 2623849.40 | norm 0.3859 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20587 | train 3.017444 | val 3.298592 | tok/s 2626901.00 | norm 0.3966 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20588 | train 2.827870 | val 3.298592 | tok/s 2627159.80 | norm 0.3433 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20589 | train 2.905689 | val 3.298592 | tok/s 2557465.82 | norm 0.3593 | dt 205.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20590 | train 2.916403 | val 3.298592 | tok/s 2631406.60 | norm 0.4023 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20591 | train 2.867573 | val 3.298592 | tok/s 2628912.66 | norm 0.3434 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20592 | train 2.914032 | val 3.298592 | tok/s 2629985.68 | norm 0.3582 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20593 | train 2.928652 | val 3.298592 | tok/s 2625963.05 | norm 0.4562 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20594 | train 2.979862 | val 3.298592 | tok/s 2625573.64 | norm 0.3496 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20595 | train 3.000311 | val 3.298592 | tok/s 2629008.38 | norm 0.3418 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20596 | train 2.945235 | val 3.298592 | tok/s 2625061.14 | norm 0.3682 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20597 | train 2.914845 | val 3.298592 | tok/s 2629924.80 | norm 0.3582 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20598 | train 2.953894 | val 3.298592 | tok/s 2629565.07 | norm 0.3560 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20599 | train 2.918696 | val 3.298592 | tok/s 2621082.09 | norm 0.3687 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20600 | train 2.937458 | val 3.298592 | tok/s 2631862.27 | norm 0.3477 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20601 | train 3.106502 | val 3.298592 | tok/s 2626735.83 | norm 0.3798 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20602 | train 3.049212 | val 3.298592 | tok/s 2627927.92 | norm 0.3816 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20603 | train 3.019561 | val 3.298592 | tok/s 2625050.01 | norm 0.3680 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20604 | train 2.954226 | val 3.298592 | tok/s 2631337.91 | norm 0.3609 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20605 | train 3.076564 | val 3.298592 | tok/s 2629447.71 | norm 0.5035 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20606 | train 3.001304 | val 3.298592 | tok/s 2582639.10 | norm 0.4255 | dt 203.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20607 | train 3.102890 | val 3.298592 | tok/s 2620155.12 | norm 0.3272 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20608 | train 2.962337 | val 3.298592 | tok/s 2624851.74 | norm 0.3422 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20609 | train 2.957939 | val 3.298592 | tok/s 2624852.60 | norm 0.3657 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20610 | train 2.904182 | val 3.298592 | tok/s 2625610.82 | norm 0.4110 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20611 | train 2.949533 | val 3.298592 | tok/s 2611415.75 | norm 0.4081 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20612 | train 2.884397 | val 3.298592 | tok/s 2626892.63 | norm 0.3683 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20613 | train 3.011545 | val 3.298592 | tok/s 2628005.23 | norm 0.3076 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20614 | train 2.890065 | val 3.298592 | tok/s 2546648.94 | norm 0.3266 | dt 205.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20615 | train 2.908086 | val 3.298592 | tok/s 2624844.76 | norm 0.3512 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20616 | train 2.973852 | val 3.298592 | tok/s 2623241.30 | norm 0.3762 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20617 | train 2.982301 | val 3.298592 | tok/s 2632052.67 | norm 0.3777 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20618 | train 2.871608 | val 3.298592 | tok/s 2630989.76 | norm 0.3538 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20619 | train 3.033689 | val 3.298592 | tok/s 2629369.88 | norm 0.3464 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20620 | train 2.917529 | val 3.298592 | tok/s 2624023.76 | norm 0.3725 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20621 | train 2.863345 | val 3.298592 | tok/s 2627334.78 | norm 0.3598 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20622 | train 2.922168 | val 3.298592 | tok/s 2629940.90 | norm 0.4067 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20623 | train 2.895012 | val 3.298592 | tok/s 2628571.63 | norm 0.3657 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20624 | train 2.921132 | val 3.298592 | tok/s 2630827.29 | norm 0.3651 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20625 | train 2.897724 | val 3.298592 | tok/s 2623731.21 | norm 0.3832 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20626 | train 2.998958 | val 3.298592 | tok/s 2628566.48 | norm 0.3375 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20627 | train 2.958131 | val 3.298592 | tok/s 2629557.34 | norm 0.3689 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20628 | train 3.037161 | val 3.298592 | tok/s 2630225.06 | norm 0.3937 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20629 | train 3.107460 | val 3.298592 | tok/s 2629849.09 | norm 0.3895 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20630 | train 3.010240 | val 3.298592 | tok/s 2627205.88 | norm 0.3675 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20631 | train 2.916928 | val 3.298592 | tok/s 2581949.78 | norm 0.3557 | dt 203.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20632 | train 2.901617 | val 3.298592 | tok/s 2630873.95 | norm 0.3694 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20633 | train 2.954621 | val 3.298592 | tok/s 2626248.95 | norm 0.3267 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20634 | train 2.981244 | val 3.298592 | tok/s 2627195.59 | norm 0.3480 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20635 | train 2.939334 | val 3.298592 | tok/s 2626621.77 | norm 0.4302 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20636 | train 3.009131 | val 3.298592 | tok/s 2627262.01 | norm 0.3621 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20637 | train 3.013937 | val 3.298592 | tok/s 2631429.27 | norm 0.3698 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20638 | train 2.982114 | val 3.298592 | tok/s 2630842.75 | norm 0.3742 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20639 | train 3.209018 | val 3.298592 | tok/s 2626701.69 | norm 0.4279 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20640 | train 2.996022 | val 3.298592 | tok/s 2628208.50 | norm 0.3680 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20641 | train 3.039106 | val 3.298592 | tok/s 2629410.74 | norm 0.3827 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20642 | train 2.836980 | val 3.298592 | tok/s 2629225.81 | norm 0.3465 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20643 | train 2.933757 | val 3.298592 | tok/s 2628443.24 | norm 0.3549 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20644 | train 3.121022 | val 3.298592 | tok/s 2630702.11 | norm 0.4013 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20645 | train 2.906983 | val 3.298592 | tok/s 2626509.21 | norm 0.3790 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20646 | train 2.860409 | val 3.298592 | tok/s 2627633.48 | norm 0.3496 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20647 | train 2.911768 | val 3.298592 | tok/s 2626715.48 | norm 0.3639 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20648 | train 2.987678 | val 3.298592 | tok/s 2622594.62 | norm 0.3491 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20649 | train 2.964994 | val 3.298592 | tok/s 2625197.55 | norm 0.3585 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20650 | train 3.033673 | val 3.298592 | tok/s 2624121.64 | norm 0.3993 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20651 | train 2.993236 | val 3.298592 | tok/s 2629644.82 | norm 0.3519 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20652 | train 2.976829 | val 3.298592 | tok/s 2626310.03 | norm 0.3285 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20653 | train 2.948827 | val 3.298592 | tok/s 2620409.62 | norm 0.3794 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20654 | train 2.973757 | val 3.298592 | tok/s 2631246.38 | norm 0.3752 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20655 | train 2.791405 | val 3.298592 | tok/s 2628316.00 | norm 0.3855 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20656 | train 2.988407 | val 3.298592 | tok/s 2629946.93 | norm 0.3594 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20657 | train 2.843035 | val 3.298592 | tok/s 2628377.97 | norm 0.3721 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20658 | train 2.858599 | val 3.298592 | tok/s 2622480.13 | norm 0.3453 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20659 | train 2.792247 | val 3.298592 | tok/s 2631620.19 | norm 0.3819 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20660 | train 2.960244 | val 3.298592 | tok/s 2628674.80 | norm 0.3793 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20661 | train 2.914921 | val 3.298592 | tok/s 2628711.81 | norm 0.3597 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20662 | train 2.921358 | val 3.298592 | tok/s 2629417.65 | norm 0.3549 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20663 | train 2.861274 | val 3.298592 | tok/s 2627853.74 | norm 0.3623 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20664 | train 2.921949 | val 3.298592 | tok/s 2634260.75 | norm 0.3807 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20665 | train 3.011362 | val 3.298592 | tok/s 2629047.89 | norm 0.3747 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20666 | train 2.846342 | val 3.298592 | tok/s 2628792.22 | norm 0.3635 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20667 | train 2.886654 | val 3.298592 | tok/s 2618428.47 | norm 0.4038 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20668 | train 2.930704 | val 3.298592 | tok/s 2623938.86 | norm 0.3881 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20669 | train 2.916101 | val 3.298592 | tok/s 2630045.71 | norm 0.3430 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20670 | train 2.961997 | val 3.298592 | tok/s 2627437.14 | norm 0.3579 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20671 | train 3.031936 | val 3.298592 | tok/s 2538058.89 | norm 0.3665 | dt 206.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20672 | train 3.002883 | val 3.298592 | tok/s 2507213.09 | norm 0.3787 | dt 209.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20673 | train 2.942655 | val 3.298592 | tok/s 2584614.63 | norm 0.3693 | dt 202.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20674 | train 2.884493 | val 3.298592 | tok/s 2631744.00 | norm 0.3855 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20675 | train 2.972222 | val 3.298592 | tok/s 2629719.94 | norm 0.3982 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20676 | train 2.924524 | val 3.298592 | tok/s 2631468.58 | norm 0.3494 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20677 | train 3.037438 | val 3.298592 | tok/s 2627360.80 | norm 0.3563 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20678 | train 2.750877 | val 3.298592 | tok/s 2626295.24 | norm 0.4041 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20679 | train 2.992849 | val 3.298592 | tok/s 2560025.95 | norm 0.8701 | dt 204.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20680 | train 2.930779 | val 3.298592 | tok/s 2632038.38 | norm 0.3814 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20681 | train 2.842561 | val 3.298592 | tok/s 2629890.72 | norm 0.3242 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20682 | train 3.005993 | val 3.298592 | tok/s 2629671.73 | norm 0.3638 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20683 | train 2.924457 | val 3.298592 | tok/s 2629885.84 | norm 0.3624 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20684 | train 2.959883 | val 3.298592 | tok/s 2630120.21 | norm 0.3604 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20685 | train 3.120087 | val 3.298592 | tok/s 2633110.82 | norm 0.4008 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20686 | train 3.034286 | val 3.298592 | tok/s 2623183.42 | norm 0.3731 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20687 | train 2.887043 | val 3.298592 | tok/s 2628829.26 | norm 0.3363 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20688 | train 2.885102 | val 3.298592 | tok/s 2631029.92 | norm 0.3716 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20689 | train 2.857300 | val 3.298592 | tok/s 2628400.48 | norm 0.3581 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20690 | train 2.835819 | val 3.298592 | tok/s 2628733.17 | norm 0.3400 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20691 | train 3.106068 | val 3.298592 | tok/s 2629872.33 | norm 0.3241 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20692 | train 3.059541 | val 3.298592 | tok/s 2621745.05 | norm 0.3346 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20693 | train 2.885615 | val 3.298592 | tok/s 2626013.68 | norm 0.3599 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20694 | train 2.954148 | val 3.298592 | tok/s 2632575.16 | norm 0.3589 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20695 | train 2.950736 | val 3.298592 | tok/s 2620251.86 | norm 0.3494 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20696 | train 2.945725 | val 3.298592 | tok/s 2630890.01 | norm 0.3358 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20697 | train 2.946029 | val 3.298592 | tok/s 2614399.87 | norm 0.3520 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20698 | train 3.026746 | val 3.298592 | tok/s 2629767.47 | norm 0.3400 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20699 | train 2.965446 | val 3.298592 | tok/s 2617103.02 | norm 0.3519 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20700 | train 2.892121 | val 3.298592 | tok/s 2617382.94 | norm 0.3845 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20701 | train 2.824393 | val 3.298592 | tok/s 2629211.00 | norm 0.4139 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20702 | train 2.977242 | val 3.298592 | tok/s 2628694.63 | norm 0.4417 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20703 | train 2.927238 | val 3.298592 | tok/s 2627976.54 | norm 0.4057 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20704 | train 2.996142 | val 3.298592 | tok/s 2629573.61 | norm 0.3779 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20705 | train 2.927684 | val 3.298592 | tok/s 2625664.62 | norm 0.4119 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20706 | train 2.871672 | val 3.298592 | tok/s 2630631.58 | norm 0.3676 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20707 | train 2.979766 | val 3.298592 | tok/s 2626977.75 | norm 0.3697 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20708 | train 3.053215 | val 3.298592 | tok/s 2630059.55 | norm 0.4243 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20709 | train 2.853316 | val 3.298592 | tok/s 2632786.06 | norm 0.3788 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20710 | train 2.985212 | val 3.298592 | tok/s 2621809.15 | norm 0.3329 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20711 | train 3.077760 | val 3.298592 | tok/s 2631320.44 | norm 0.4196 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20712 | train 2.887216 | val 3.298592 | tok/s 2625554.24 | norm 0.4033 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20713 | train 2.929980 | val 3.298592 | tok/s 2631555.95 | norm 0.3579 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20714 | train 3.008957 | val 3.298592 | tok/s 2618800.16 | norm 0.4151 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20715 | train 2.930380 | val 3.298592 | tok/s 2620252.40 | norm 0.4099 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20716 | train 2.892477 | val 3.298592 | tok/s 2634011.41 | norm 0.3482 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20717 | train 2.920255 | val 3.298592 | tok/s 2627902.35 | norm 0.3395 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20718 | train 2.889359 | val 3.298592 | tok/s 2628526.19 | norm 0.3517 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20719 | train 2.956013 | val 3.298592 | tok/s 2628453.70 | norm 0.3671 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20720 | train 2.923294 | val 3.298592 | tok/s 2542033.32 | norm 0.3727 | dt 206.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20721 | train 2.952351 | val 3.298592 | tok/s 2454635.49 | norm 0.3884 | dt 213.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20722 | train 2.908192 | val 3.298592 | tok/s 2628751.60 | norm 0.3875 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20723 | train 2.910691 | val 3.298592 | tok/s 2627273.34 | norm 0.3626 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20724 | train 3.015524 | val 3.298592 | tok/s 2626759.11 | norm 0.3552 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20725 | train 2.896255 | val 3.298592 | tok/s 2624785.49 | norm 0.3823 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20726 | train 2.890258 | val 3.298592 | tok/s 2629665.79 | norm 0.3661 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20727 | train 2.972148 | val 3.298592 | tok/s 2629184.59 | norm 0.3498 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20728 | train 3.026868 | val 3.298592 | tok/s 2580759.39 | norm 0.3396 | dt 203.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20729 | train 2.911723 | val 3.298592 | tok/s 2629653.45 | norm 0.3797 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20730 | train 2.922656 | val 3.298592 | tok/s 2626643.24 | norm 0.3702 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20731 | train 3.062726 | val 3.298592 | tok/s 2628388.29 | norm 0.3252 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20732 | train 2.907826 | val 3.298592 | tok/s 2627299.51 | norm 0.3667 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20733 | train 2.932369 | val 3.298592 | tok/s 2630449.40 | norm 0.3848 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20734 | train 2.913829 | val 3.298592 | tok/s 2624554.93 | norm 0.3924 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20735 | train 2.875956 | val 3.298592 | tok/s 2617771.69 | norm 0.3835 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20736 | train 2.970788 | val 3.298592 | tok/s 2628531.48 | norm 0.3892 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20737 | train 2.955672 | val 3.298592 | tok/s 2628946.29 | norm 0.3721 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20738 | train 2.959713 | val 3.298592 | tok/s 2628881.06 | norm 0.3695 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20739 | train 3.033146 | val 3.298592 | tok/s 2625523.11 | norm 0.3670 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20740 | train 3.070700 | val 3.298592 | tok/s 2620817.06 | norm 0.4096 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20741 | train 2.968017 | val 3.298592 | tok/s 2629203.91 | norm 0.3718 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20742 | train 2.928384 | val 3.298592 | tok/s 2620948.11 | norm 0.3463 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20743 | train 2.945822 | val 3.298592 | tok/s 2631450.72 | norm 0.3638 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20744 | train 2.940870 | val 3.298592 | tok/s 2630137.65 | norm 0.3624 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20745 | train 2.871330 | val 3.298592 | tok/s 2626308.01 | norm 0.3618 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20746 | train 2.920769 | val 3.298592 | tok/s 2625321.26 | norm 0.3609 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20747 | train 2.893902 | val 3.298592 | tok/s 2623268.88 | norm 0.3693 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20748 | train 3.105664 | val 3.298592 | tok/s 2628637.27 | norm 0.3706 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20749 | train 2.998230 | val 3.298592 | tok/s 2627634.90 | norm 0.3467 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 20750 | val_loss 3.296826 step 20750 | train 2.886449 | val 3.296826 | tok/s 2610688.79 | norm 0.3734 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20751 | train 2.962899 | val 3.296826 | tok/s 2626161.08 | norm 0.3711 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20752 | train 2.913359 | val 3.296826 | tok/s 2625628.26 | norm 0.3622 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20753 | train 3.150381 | val 3.296826 | tok/s 2622686.08 | norm 0.3851 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20754 | train 3.036485 | val 3.296826 | tok/s 2630917.71 | norm 0.3782 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20755 | train 2.884551 | val 3.296826 | tok/s 2627434.11 | norm 0.3528 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20756 | train 2.924660 | val 3.296826 | tok/s 2625690.89 | norm 0.3318 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20757 | train 2.975151 | val 3.296826 | tok/s 2630073.15 | norm 0.3456 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20758 | train 3.007301 | val 3.296826 | tok/s 2618828.23 | norm 0.3535 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20759 | train 2.988398 | val 3.296826 | tok/s 2626191.87 | norm 0.3490 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20760 | train 2.842454 | val 3.296826 | tok/s 2621424.31 | norm 0.3569 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20761 | train 2.942524 | val 3.296826 | tok/s 2623795.12 | norm 0.3313 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20762 | train 2.886145 | val 3.296826 | tok/s 2623262.37 | norm 0.3641 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20763 | train 2.899335 | val 3.296826 | tok/s 2621536.04 | norm 0.3564 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20764 | train 2.880759 | val 3.296826 | tok/s 2629791.34 | norm 0.3891 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20765 | train 2.899787 | val 3.296826 | tok/s 2626817.11 | norm 0.3922 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20766 | train 3.042633 | val 3.296826 | tok/s 2629243.07 | norm 0.3416 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20767 | train 2.906030 | val 3.296826 | tok/s 2627035.82 | norm 0.3036 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20768 | train 2.830345 | val 3.296826 | tok/s 2623368.87 | norm 0.3286 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20769 | train 3.004404 | val 3.296826 | tok/s 2626832.28 | norm 0.3772 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20770 | train 3.001072 | val 3.296826 | tok/s 2627287.45 | norm 0.4164 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20771 | train 3.049527 | val 3.296826 | tok/s 2628282.09 | norm 0.3583 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20772 | train 2.936048 | val 3.296826 | tok/s 2626777.73 | norm 0.3712 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20773 | train 2.944564 | val 3.296826 | tok/s 2627426.08 | norm 0.3541 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20774 | train 2.796019 | val 3.296826 | tok/s 2622796.02 | norm 0.3466 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20775 | train 2.963461 | val 3.296826 | tok/s 2616824.25 | norm 0.3943 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20776 | train 2.873733 | val 3.296826 | tok/s 2626818.78 | norm 0.4014 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20777 | train 2.831145 | val 3.296826 | tok/s 2628407.54 | norm 0.3975 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20778 | train 3.027635 | val 3.296826 | tok/s 2625478.25 | norm 0.4035 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20779 | train 2.952540 | val 3.296826 | tok/s 2629082.66 | norm 0.3791 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20780 | train 3.064252 | val 3.296826 | tok/s 2495315.24 | norm 0.3785 | dt 210.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20781 | train 2.945811 | val 3.296826 | tok/s 2631880.75 | norm 0.3747 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20782 | train 2.967957 | val 3.296826 | tok/s 2620316.92 | norm 0.3710 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20783 | train 3.187003 | val 3.296826 | tok/s 2603801.52 | norm 0.3896 | dt 201.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20784 | train 3.005048 | val 3.296826 | tok/s 2627543.72 | norm 0.3895 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20785 | train 2.913342 | val 3.296826 | tok/s 2629199.52 | norm 0.3659 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20786 | train 2.940001 | val 3.296826 | tok/s 2630525.45 | norm 0.4106 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20787 | train 2.931738 | val 3.296826 | tok/s 2625607.80 | norm 0.3853 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20788 | train 3.069647 | val 3.296826 | tok/s 2628597.88 | norm 0.3839 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20789 | train 3.012085 | val 3.296826 | tok/s 2633520.17 | norm 0.3797 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20790 | train 3.033113 | val 3.296826 | tok/s 2632315.68 | norm 0.3482 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20791 | train 2.948783 | val 3.296826 | tok/s 2629228.43 | norm 0.3497 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20792 | train 2.965958 | val 3.296826 | tok/s 2627443.62 | norm 0.3855 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20793 | train 2.960165 | val 3.296826 | tok/s 2622871.58 | norm 0.3893 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20794 | train 3.013302 | val 3.296826 | tok/s 2631533.68 | norm 0.3842 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20795 | train 3.073225 | val 3.296826 | tok/s 2628105.10 | norm 0.5211 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20796 | train 3.033026 | val 3.296826 | tok/s 2629753.00 | norm 0.3786 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20797 | train 2.940466 | val 3.296826 | tok/s 2625337.80 | norm 0.3286 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20798 | train 3.016425 | val 3.296826 | tok/s 2627555.77 | norm 0.3977 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20799 | train 2.881836 | val 3.296826 | tok/s 2626773.90 | norm 0.3427 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20800 | train 2.837600 | val 3.296826 | tok/s 2615873.10 | norm 0.3618 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20801 | train 3.052706 | val 3.296826 | tok/s 2631387.05 | norm 0.3756 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20802 | train 3.049700 | val 3.296826 | tok/s 2627156.38 | norm 0.3889 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20803 | train 3.187972 | val 3.296826 | tok/s 2629833.96 | norm 0.6472 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20804 | train 3.015069 | val 3.296826 | tok/s 2625556.37 | norm 0.3978 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20805 | train 2.991271 | val 3.296826 | tok/s 2620337.20 | norm 0.3163 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20806 | train 3.053188 | val 3.296826 | tok/s 2626663.80 | norm 0.4019 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20807 | train 3.023562 | val 3.296826 | tok/s 2631665.14 | norm 0.3777 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20808 | train 2.963391 | val 3.296826 | tok/s 2631223.11 | norm 0.3468 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20809 | train 2.962276 | val 3.296826 | tok/s 2626671.95 | norm 0.3484 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20810 | train 2.840973 | val 3.296826 | tok/s 2623587.80 | norm 0.3679 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20811 | train 2.990213 | val 3.296826 | tok/s 2630282.02 | norm 0.3747 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20812 | train 2.988238 | val 3.296826 | tok/s 2622266.97 | norm 0.3581 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20813 | train 2.925292 | val 3.296826 | tok/s 2628373.58 | norm 0.3391 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20814 | train 2.985805 | val 3.296826 | tok/s 2629409.24 | norm 0.3580 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20815 | train 2.815417 | val 3.296826 | tok/s 2622836.66 | norm 0.3679 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20816 | train 3.028824 | val 3.296826 | tok/s 2631765.39 | norm 0.3703 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20817 | train 3.060818 | val 3.296826 | tok/s 2629268.04 | norm 0.3628 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20818 | train 2.909469 | val 3.296826 | tok/s 2624385.23 | norm 0.3501 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20819 | train 2.833834 | val 3.296826 | tok/s 2618369.45 | norm 0.3532 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20820 | train 3.036363 | val 3.296826 | tok/s 2616656.03 | norm 0.3729 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20821 | train 2.962406 | val 3.296826 | tok/s 2629649.57 | norm 0.3546 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20822 | train 2.936274 | val 3.296826 | tok/s 2624571.00 | norm 0.4109 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20823 | train 2.879433 | val 3.296826 | tok/s 2547541.13 | norm 0.3875 | dt 205.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20824 | train 2.922585 | val 3.296826 | tok/s 2596784.07 | norm 0.3575 | dt 201.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20825 | train 2.970123 | val 3.296826 | tok/s 2626806.70 | norm 0.3770 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20826 | train 3.005629 | val 3.296826 | tok/s 2632459.58 | norm 0.3746 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20827 | train 2.939723 | val 3.296826 | tok/s 2629099.46 | norm 0.3752 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20828 | train 2.954701 | val 3.296826 | tok/s 2633554.87 | norm 0.3683 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20829 | train 2.983176 | val 3.296826 | tok/s 2627609.00 | norm 0.3758 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20830 | train 3.024613 | val 3.296826 | tok/s 2488728.22 | norm 0.3778 | dt 210.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20831 | train 2.905266 | val 3.296826 | tok/s 2575328.47 | norm 0.3495 | dt 203.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20832 | train 3.088259 | val 3.296826 | tok/s 2626650.57 | norm 0.3765 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20833 | train 2.928262 | val 3.296826 | tok/s 2628781.03 | norm 0.3867 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20834 | train 3.014777 | val 3.296826 | tok/s 2627295.05 | norm 0.4062 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20835 | train 2.893505 | val 3.296826 | tok/s 2625559.50 | norm 0.3786 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20836 | train 2.897192 | val 3.296826 | tok/s 2631036.15 | norm 0.3286 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20837 | train 2.942346 | val 3.296826 | tok/s 2629991.44 | norm 0.3553 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20838 | train 2.972320 | val 3.296826 | tok/s 2628815.50 | norm 0.3966 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20839 | train 3.035471 | val 3.296826 | tok/s 2623169.30 | norm 0.3727 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20840 | train 2.982468 | val 3.296826 | tok/s 2628713.66 | norm 0.3630 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20841 | train 2.909714 | val 3.296826 | tok/s 2631364.67 | norm 0.3284 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20842 | train 2.982793 | val 3.296826 | tok/s 2618357.15 | norm 0.3518 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20843 | train 2.867882 | val 3.296826 | tok/s 2631471.02 | norm 0.3811 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20844 | train 2.933879 | val 3.296826 | tok/s 2626536.60 | norm 0.3372 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20845 | train 2.980987 | val 3.296826 | tok/s 2626381.66 | norm 0.3615 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20846 | train 2.982012 | val 3.296826 | tok/s 2627631.11 | norm 0.3720 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20847 | train 2.993032 | val 3.296826 | tok/s 2616278.34 | norm 0.3917 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20848 | train 2.918046 | val 3.296826 | tok/s 2629500.67 | norm 0.3745 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20849 | train 2.926970 | val 3.296826 | tok/s 2627684.92 | norm 0.3645 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20850 | train 2.924482 | val 3.296826 | tok/s 2628997.29 | norm 0.4073 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20851 | train 3.047565 | val 3.296826 | tok/s 2625830.89 | norm 0.4370 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20852 | train 2.910043 | val 3.296826 | tok/s 2623437.09 | norm 0.3849 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20853 | train 2.910648 | val 3.296826 | tok/s 2624051.77 | norm 0.3452 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20854 | train 2.938658 | val 3.296826 | tok/s 2629850.93 | norm 0.3327 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20855 | train 2.921681 | val 3.296826 | tok/s 2626387.36 | norm 0.3456 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20856 | train 2.976895 | val 3.296826 | tok/s 2632347.59 | norm 0.3818 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20857 | train 2.986642 | val 3.296826 | tok/s 2622527.09 | norm 0.3494 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20858 | train 2.846950 | val 3.296826 | tok/s 2630347.76 | norm 0.3490 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20859 | train 3.024127 | val 3.296826 | tok/s 2625248.87 | norm 0.3587 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20860 | train 2.942665 | val 3.296826 | tok/s 2627360.57 | norm 0.3798 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20861 | train 3.001986 | val 3.296826 | tok/s 2623758.26 | norm 0.3664 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20862 | train 3.030939 | val 3.296826 | tok/s 2618176.10 | norm 0.3719 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20863 | train 3.103293 | val 3.296826 | tok/s 2621756.79 | norm 0.5833 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20864 | train 2.952017 | val 3.296826 | tok/s 2623248.48 | norm 0.3587 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20865 | train 3.014822 | val 3.296826 | tok/s 2630657.13 | norm 0.3482 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20866 | train 2.978109 | val 3.296826 | tok/s 2621130.59 | norm 0.3780 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20867 | train 2.867678 | val 3.296826 | tok/s 2625655.57 | norm 0.3732 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20868 | train 2.767474 | val 3.296826 | tok/s 2630613.77 | norm 0.3329 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20869 | train 3.035562 | val 3.296826 | tok/s 2631782.76 | norm 0.3900 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20870 | train 2.989360 | val 3.296826 | tok/s 2631944.34 | norm 0.3926 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20871 | train 2.917935 | val 3.296826 | tok/s 2631644.15 | norm 0.3583 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20872 | train 2.936435 | val 3.296826 | tok/s 2623583.77 | norm 0.3554 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20873 | train 2.928041 | val 3.296826 | tok/s 2630807.49 | norm 0.3683 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20874 | train 2.948995 | val 3.296826 | tok/s 2624606.97 | norm 0.3510 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20875 | train 2.872885 | val 3.296826 | tok/s 2628027.99 | norm 0.3407 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20876 | train 2.862992 | val 3.296826 | tok/s 2631775.19 | norm 0.3654 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20877 | train 2.900193 | val 3.296826 | tok/s 2628608.04 | norm 0.3870 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20878 | train 2.960549 | val 3.296826 | tok/s 2627821.98 | norm 0.3351 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20879 | train 3.089374 | val 3.296826 | tok/s 2625707.85 | norm 0.3629 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20880 | train 3.167500 | val 3.296826 | tok/s 2630102.69 | norm 0.4365 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20881 | train 2.949945 | val 3.296826 | tok/s 2616398.89 | norm 0.4168 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20882 | train 2.835610 | val 3.296826 | tok/s 2625934.81 | norm 0.3708 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20883 | train 2.945565 | val 3.296826 | tok/s 2631663.78 | norm 0.4037 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20884 | train 2.861902 | val 3.296826 | tok/s 2626221.29 | norm 0.3966 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20885 | train 3.022972 | val 3.296826 | tok/s 2630520.49 | norm 0.3710 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20886 | train 3.032817 | val 3.296826 | tok/s 2626084.99 | norm 0.3378 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20887 | train 3.061055 | val 3.296826 | tok/s 2560780.79 | norm 0.3509 | dt 204.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20888 | train 2.891098 | val 3.296826 | tok/s 2632009.78 | norm 0.3429 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20889 | train 2.784651 | val 3.296826 | tok/s 2631900.28 | norm 0.3253 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20890 | train 2.994385 | val 3.296826 | tok/s 2632633.48 | norm 0.3652 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20891 | train 2.818013 | val 3.296826 | tok/s 2627415.46 | norm 0.3190 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20892 | train 3.002969 | val 3.296826 | tok/s 2629625.25 | norm 0.3393 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20893 | train 2.949560 | val 3.296826 | tok/s 2630740.07 | norm 0.3643 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20894 | train 2.921144 | val 3.296826 | tok/s 2626361.67 | norm 0.3485 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20895 | train 3.060267 | val 3.296826 | tok/s 2628112.74 | norm 0.3723 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20896 | train 3.000517 | val 3.296826 | tok/s 2623682.30 | norm 0.3894 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20897 | train 2.893977 | val 3.296826 | tok/s 2627184.63 | norm 0.3617 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20898 | train 2.989691 | val 3.296826 | tok/s 2632877.47 | norm 0.3742 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20899 | train 2.972850 | val 3.296826 | tok/s 2620670.30 | norm 0.3537 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20900 | train 2.970062 | val 3.296826 | tok/s 2628804.73 | norm 0.3446 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20901 | train 2.985117 | val 3.296826 | tok/s 2630503.36 | norm 0.3407 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20902 | train 2.843535 | val 3.296826 | tok/s 2629329.10 | norm 0.3642 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20903 | train 2.873892 | val 3.296826 | tok/s 2625235.04 | norm 0.3736 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20904 | train 2.951631 | val 3.296826 | tok/s 2619044.78 | norm 0.3562 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20905 | train 2.954652 | val 3.296826 | tok/s 2629843.35 | norm 0.3658 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20906 | train 2.917402 | val 3.296826 | tok/s 2626293.31 | norm 0.3606 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20907 | train 2.876212 | val 3.296826 | tok/s 2629516.73 | norm 0.3631 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20908 | train 2.887076 | val 3.296826 | tok/s 2609911.49 | norm 0.4010 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20909 | train 2.835684 | val 3.296826 | tok/s 2599093.49 | norm 0.3979 | dt 201.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20910 | train 2.915553 | val 3.296826 | tok/s 2630197.24 | norm 0.3741 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20911 | train 2.854785 | val 3.296826 | tok/s 2626077.98 | norm 0.3742 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20912 | train 2.738203 | val 3.296826 | tok/s 2625157.96 | norm 0.3491 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20913 | train 2.908567 | val 3.296826 | tok/s 2625328.61 | norm 0.3431 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20914 | train 2.977793 | val 3.296826 | tok/s 2620920.65 | norm 0.3735 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20915 | train 2.943196 | val 3.296826 | tok/s 2630842.79 | norm 0.3925 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20916 | train 2.940336 | val 3.296826 | tok/s 2626678.90 | norm 0.3571 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20917 | train 2.927043 | val 3.296826 | tok/s 2629008.98 | norm 0.3600 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20918 | train 2.876613 | val 3.296826 | tok/s 2621820.91 | norm 0.3869 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20919 | train 2.925377 | val 3.296826 | tok/s 2623513.36 | norm 0.3824 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20920 | train 2.890041 | val 3.296826 | tok/s 2625679.90 | norm 0.3929 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20921 | train 2.901040 | val 3.296826 | tok/s 2631914.29 | norm 0.3953 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20922 | train 2.926856 | val 3.296826 | tok/s 2630146.02 | norm 0.3993 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20923 | train 2.909329 | val 3.296826 | tok/s 2630479.70 | norm 0.3795 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20924 | train 2.982340 | val 3.296826 | tok/s 2628562.97 | norm 0.3834 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20925 | train 2.922163 | val 3.296826 | tok/s 2630705.60 | norm 0.4114 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20926 | train 2.945742 | val 3.296826 | tok/s 2628577.36 | norm 0.4112 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20927 | train 2.782904 | val 3.296826 | tok/s 2520836.36 | norm 0.4231 | dt 207.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20928 | train 3.002015 | val 3.296826 | tok/s 2469215.99 | norm 0.3747 | dt 212.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20929 | train 2.901052 | val 3.296826 | tok/s 2621680.29 | norm 0.3468 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20930 | train 2.879220 | val 3.296826 | tok/s 2630550.52 | norm 0.3906 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20931 | train 2.984157 | val 3.296826 | tok/s 2627801.32 | norm 0.4069 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20932 | train 3.026819 | val 3.296826 | tok/s 2631840.30 | norm 0.3682 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20933 | train 2.878457 | val 3.296826 | tok/s 2625833.58 | norm 0.3644 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20934 | train 2.934850 | val 3.296826 | tok/s 2628971.33 | norm 0.3691 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20935 | train 2.924992 | val 3.296826 | tok/s 2627966.92 | norm 0.3704 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20936 | train 2.935361 | val 3.296826 | tok/s 2631042.30 | norm 0.3854 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20937 | train 3.048697 | val 3.296826 | tok/s 2629972.83 | norm 0.3807 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20938 | train 2.975628 | val 3.296826 | tok/s 2629657.21 | norm 0.3656 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20939 | train 2.939743 | val 3.296826 | tok/s 2625241.80 | norm 0.3646 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20940 | train 2.964873 | val 3.296826 | tok/s 2632828.32 | norm 0.3620 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20941 | train 2.926863 | val 3.296826 | tok/s 2623469.98 | norm 0.3492 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20942 | train 3.036167 | val 3.296826 | tok/s 2628211.05 | norm 0.4079 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20943 | train 2.921273 | val 3.296826 | tok/s 2625312.68 | norm 0.3832 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20944 | train 2.897826 | val 3.296826 | tok/s 2628709.83 | norm 0.3228 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20945 | train 3.036829 | val 3.296826 | tok/s 2621676.16 | norm 0.3628 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20946 | train 2.995647 | val 3.296826 | tok/s 2619929.82 | norm 0.3923 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20947 | train 2.976073 | val 3.296826 | tok/s 2625788.83 | norm 0.3830 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20948 | train 2.902473 | val 3.296826 | tok/s 2628000.20 | norm 0.3966 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20949 | train 3.001766 | val 3.296826 | tok/s 2630466.21 | norm 0.3893 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20950 | train 2.903013 | val 3.296826 | tok/s 2629870.18 | norm 0.3681 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20951 | train 3.045286 | val 3.296826 | tok/s 2626406.98 | norm 0.5012 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20952 | train 2.847032 | val 3.296826 | tok/s 2633338.42 | norm 0.3611 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20953 | train 3.057790 | val 3.296826 | tok/s 2627794.23 | norm 0.3918 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20954 | train 2.914260 | val 3.296826 | tok/s 2629544.39 | norm 0.4194 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20955 | train 2.860327 | val 3.296826 | tok/s 2626615.80 | norm 0.3771 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20956 | train 2.957260 | val 3.296826 | tok/s 2623120.53 | norm 0.3980 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20957 | train 2.901002 | val 3.296826 | tok/s 2630865.83 | norm 0.3310 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20958 | train 2.982344 | val 3.296826 | tok/s 2625204.24 | norm 0.3145 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20959 | train 3.028339 | val 3.296826 | tok/s 2628434.50 | norm 0.3278 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20960 | train 2.923419 | val 3.296826 | tok/s 2630361.90 | norm 0.3429 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20961 | train 2.834088 | val 3.296826 | tok/s 2626676.03 | norm 0.3667 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20962 | train 2.996062 | val 3.296826 | tok/s 2631615.44 | norm 0.3760 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20963 | train 2.948669 | val 3.296826 | tok/s 2628056.95 | norm 0.3583 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20964 | train 2.920656 | val 3.296826 | tok/s 2628471.94 | norm 0.3816 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20965 | train 2.988235 | val 3.296826 | tok/s 2619833.54 | norm 0.3995 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20966 | train 2.963799 | val 3.296826 | tok/s 2624381.01 | norm 0.3792 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20967 | train 2.972878 | val 3.296826 | tok/s 2630221.78 | norm 0.3833 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20968 | train 2.899300 | val 3.296826 | tok/s 2629327.40 | norm 0.3776 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20969 | train 2.961070 | val 3.296826 | tok/s 2526780.46 | norm 0.3935 | dt 207.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20970 | train 2.866395 | val 3.296826 | tok/s 2459066.65 | norm 0.4004 | dt 213.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20971 | train 2.891196 | val 3.296826 | tok/s 2624584.23 | norm 0.3694 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20972 | train 2.945650 | val 3.296826 | tok/s 2631460.34 | norm 0.3276 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20973 | train 2.885972 | val 3.296826 | tok/s 2628629.12 | norm 0.3608 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20974 | train 2.962073 | val 3.296826 | tok/s 2629052.16 | norm 0.3982 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20975 | train 2.863512 | val 3.296826 | tok/s 2627825.87 | norm 0.3578 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20976 | train 2.860048 | val 3.296826 | tok/s 2625587.57 | norm 0.3389 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20977 | train 2.870444 | val 3.296826 | tok/s 2573085.84 | norm 0.3860 | dt 203.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20978 | train 2.967305 | val 3.296826 | tok/s 2625588.93 | norm 0.3705 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20979 | train 2.943724 | val 3.296826 | tok/s 2607086.06 | norm 0.3824 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20980 | train 2.957289 | val 3.296826 | tok/s 2627314.87 | norm 0.3865 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20981 | train 2.899425 | val 3.296826 | tok/s 2621602.26 | norm 0.3722 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20982 | train 3.004070 | val 3.296826 | tok/s 2626745.11 | norm 0.3646 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20983 | train 3.060672 | val 3.296826 | tok/s 2619807.17 | norm 0.3714 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20984 | train 2.884894 | val 3.296826 | tok/s 2629123.71 | norm 0.3734 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20985 | train 2.933025 | val 3.296826 | tok/s 2625643.06 | norm 0.3980 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20986 | train 3.175720 | val 3.296826 | tok/s 2626624.22 | norm 0.3918 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20987 | train 3.051953 | val 3.296826 | tok/s 2276612.07 | norm 0.3481 | dt 230.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20988 | train 2.851349 | val 3.296826 | tok/s 2627125.93 | norm 0.3583 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20989 | train 2.917336 | val 3.296826 | tok/s 2629355.89 | norm 0.3889 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20990 | train 2.888412 | val 3.296826 | tok/s 2630392.77 | norm 0.3553 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20991 | train 2.968343 | val 3.296826 | tok/s 2628563.67 | norm 0.3663 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20992 | train 2.825096 | val 3.296826 | tok/s 2629378.47 | norm 0.4185 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20993 | train 2.959878 | val 3.296826 | tok/s 2626379.29 | norm 0.3694 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20994 | train 2.954567 | val 3.296826 | tok/s 2629999.68 | norm 0.3483 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20995 | train 2.843738 | val 3.296826 | tok/s 2182954.13 | norm 0.3905 | dt 240.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20996 | train 2.921874 | val 3.296826 | tok/s 2627699.85 | norm 0.3680 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20997 | train 2.890549 | val 3.296826 | tok/s 2628383.91 | norm 0.3858 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20998 | train 3.010695 | val 3.296826 | tok/s 2628523.33 | norm 0.4034 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 20999 | train 2.974063 | val 3.296826 | tok/s 2624011.61 | norm 0.3611 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 21000 | val_loss 3.295141 step 21000 | train 2.926488 | val 3.295141 | tok/s 2575900.01 | norm 0.3439 | dt 203.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21001 | train 2.943839 | val 3.295141 | tok/s 2603137.78 | norm 0.3816 | dt 201.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21002 | train 3.088626 | val 3.295141 | tok/s 2623438.23 | norm 0.3744 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21003 | train 2.898452 | val 3.295141 | tok/s 2628393.37 | norm 0.3407 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21004 | train 2.843718 | val 3.295141 | tok/s 2627617.53 | norm 0.3947 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21005 | train 2.989435 | val 3.295141 | tok/s 2628123.61 | norm 0.3540 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21006 | train 2.952367 | val 3.295141 | tok/s 2615842.86 | norm 0.3449 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21007 | train 2.931764 | val 3.295141 | tok/s 2630865.77 | norm 0.3461 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21008 | train 2.927502 | val 3.295141 | tok/s 2628312.68 | norm 0.3869 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21009 | train 2.892380 | val 3.295141 | tok/s 2626737.28 | norm 0.3670 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21010 | train 2.932757 | val 3.295141 | tok/s 2630579.40 | norm 0.3445 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21011 | train 2.850408 | val 3.295141 | tok/s 2613766.10 | norm 0.3652 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21012 | train 2.892332 | val 3.295141 | tok/s 2631179.79 | norm 0.3887 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21013 | train 3.015868 | val 3.295141 | tok/s 2621246.58 | norm 0.3893 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21014 | train 2.959952 | val 3.295141 | tok/s 2633473.31 | norm 0.3741 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21015 | train 3.000252 | val 3.295141 | tok/s 2627837.21 | norm 0.3807 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21016 | train 2.869590 | val 3.295141 | tok/s 2624923.56 | norm 0.3674 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21017 | train 2.943760 | val 3.295141 | tok/s 2630123.92 | norm 0.3506 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21018 | train 2.850652 | val 3.295141 | tok/s 2624842.94 | norm 0.3597 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21019 | train 2.964331 | val 3.295141 | tok/s 2628798.47 | norm 0.3637 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21020 | train 2.985274 | val 3.295141 | tok/s 2619735.97 | norm 0.3313 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21021 | train 2.873778 | val 3.295141 | tok/s 2622159.93 | norm 0.3480 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21022 | train 2.982136 | val 3.295141 | tok/s 2625951.37 | norm 0.3545 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21023 | train 2.963702 | val 3.295141 | tok/s 2628271.13 | norm 0.4050 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21024 | train 3.016686 | val 3.295141 | tok/s 2627897.28 | norm 0.3857 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21025 | train 2.955793 | val 3.295141 | tok/s 2624117.70 | norm 0.3420 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21026 | train 3.020858 | val 3.295141 | tok/s 2626068.43 | norm 0.3693 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21027 | train 3.003700 | val 3.295141 | tok/s 2631462.20 | norm 0.3545 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21028 | train 2.981273 | val 3.295141 | tok/s 2621946.68 | norm 0.3372 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21029 | train 2.997590 | val 3.295141 | tok/s 2629401.82 | norm 0.3395 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21030 | train 2.843029 | val 3.295141 | tok/s 2619354.48 | norm 0.3759 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21031 | train 2.974792 | val 3.295141 | tok/s 2626241.35 | norm 0.3743 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21032 | train 2.893881 | val 3.295141 | tok/s 2630582.54 | norm 0.3424 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21033 | train 3.006034 | val 3.295141 | tok/s 2627891.01 | norm 0.3300 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21034 | train 2.888649 | val 3.295141 | tok/s 2630160.86 | norm 0.3506 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21035 | train 2.838852 | val 3.295141 | tok/s 2629239.15 | norm 0.3496 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21036 | train 2.959034 | val 3.295141 | tok/s 2621435.78 | norm 0.3767 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21037 | train 2.997927 | val 3.295141 | tok/s 2631524.29 | norm 0.3259 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21038 | train 2.982190 | val 3.295141 | tok/s 2632251.08 | norm 0.3477 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21039 | train 2.819939 | val 3.295141 | tok/s 2628432.87 | norm 0.3585 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21040 | train 2.970628 | val 3.295141 | tok/s 2625026.92 | norm 0.3731 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21041 | train 2.957127 | val 3.295141 | tok/s 2626837.53 | norm 0.3615 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21042 | train 2.892822 | val 3.295141 | tok/s 2628205.70 | norm 0.3500 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21043 | train 2.830594 | val 3.295141 | tok/s 2629202.51 | norm 0.3571 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21044 | train 2.975503 | val 3.295141 | tok/s 2606714.36 | norm 0.3339 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21045 | train 3.015173 | val 3.295141 | tok/s 2624505.17 | norm 0.3294 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21046 | train 2.850839 | val 3.295141 | tok/s 2626485.93 | norm 0.3640 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21047 | train 2.958068 | val 3.295141 | tok/s 2632499.84 | norm 0.3815 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21048 | train 2.960233 | val 3.295141 | tok/s 2631525.56 | norm 0.3773 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21049 | train 2.997810 | val 3.295141 | tok/s 2526175.68 | norm 0.3449 | dt 207.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21050 | train 2.994718 | val 3.295141 | tok/s 2594566.85 | norm 0.3360 | dt 202.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21051 | train 2.972666 | val 3.295141 | tok/s 2629848.04 | norm 0.3654 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21052 | train 2.841378 | val 3.295141 | tok/s 2631632.14 | norm 0.3718 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21053 | train 2.977269 | val 3.295141 | tok/s 2627307.75 | norm 0.4276 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21054 | train 2.977646 | val 3.295141 | tok/s 2630686.76 | norm 0.3857 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21055 | train 2.961117 | val 3.295141 | tok/s 2625526.19 | norm 0.3826 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21056 | train 2.997146 | val 3.295141 | tok/s 2619887.49 | norm 0.3949 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21057 | train 2.873339 | val 3.295141 | tok/s 2625843.63 | norm 0.3681 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21058 | train 2.910506 | val 3.295141 | tok/s 2628823.18 | norm 0.3427 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21059 | train 2.970361 | val 3.295141 | tok/s 2633265.05 | norm 0.3955 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21060 | train 3.014974 | val 3.295141 | tok/s 2631270.39 | norm 0.3807 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21061 | train 3.004367 | val 3.295141 | tok/s 2628076.79 | norm 0.3849 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21062 | train 2.890061 | val 3.295141 | tok/s 2628114.48 | norm 0.4351 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21063 | train 2.956445 | val 3.295141 | tok/s 2621466.49 | norm 0.4062 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21064 | train 2.943272 | val 3.295141 | tok/s 2630291.73 | norm 0.3810 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21065 | train 3.037922 | val 3.295141 | tok/s 2626589.01 | norm 0.3743 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21066 | train 2.894311 | val 3.295141 | tok/s 2628532.25 | norm 0.3615 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21067 | train 2.897860 | val 3.295141 | tok/s 2625822.77 | norm 0.3448 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21068 | train 2.870649 | val 3.295141 | tok/s 2626584.40 | norm 0.3640 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21069 | train 2.978717 | val 3.295141 | tok/s 2630462.14 | norm 0.3959 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21070 | train 2.987594 | val 3.295141 | tok/s 2627584.19 | norm 0.3881 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21071 | train 2.934178 | val 3.295141 | tok/s 2629639.20 | norm 0.3544 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21072 | train 2.906518 | val 3.295141 | tok/s 2618474.39 | norm 0.3554 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21073 | train 2.960173 | val 3.295141 | tok/s 2621144.90 | norm 0.3757 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21074 | train 2.962084 | val 3.295141 | tok/s 2625516.54 | norm 0.3711 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21075 | train 2.873219 | val 3.295141 | tok/s 2620912.04 | norm 0.3447 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21076 | train 3.063299 | val 3.295141 | tok/s 2623112.38 | norm 0.3543 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21077 | train 2.898604 | val 3.295141 | tok/s 2624388.71 | norm 0.3775 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21078 | train 2.817710 | val 3.295141 | tok/s 2623019.18 | norm 0.3885 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21079 | train 2.901093 | val 3.295141 | tok/s 2625519.68 | norm 0.4310 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21080 | train 2.976764 | val 3.295141 | tok/s 2617209.53 | norm 0.3806 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21081 | train 2.788825 | val 3.295141 | tok/s 2629058.41 | norm 0.3762 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21082 | train 2.843443 | val 3.295141 | tok/s 2622279.16 | norm 0.3782 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21083 | train 3.037791 | val 3.295141 | tok/s 2623954.89 | norm 0.3639 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21084 | train 2.895280 | val 3.295141 | tok/s 2630654.78 | norm 0.3467 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21085 | train 2.894581 | val 3.295141 | tok/s 2629387.48 | norm 0.3672 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21086 | train 2.970121 | val 3.295141 | tok/s 2627142.50 | norm 0.3590 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21087 | train 2.965950 | val 3.295141 | tok/s 2630875.29 | norm 0.3896 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21088 | train 2.942842 | val 3.295141 | tok/s 2628732.80 | norm 0.3482 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21089 | train 2.905028 | val 3.295141 | tok/s 2631414.61 | norm 0.3503 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21090 | train 2.965136 | val 3.295141 | tok/s 2628657.18 | norm 0.3707 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21091 | train 2.970026 | val 3.295141 | tok/s 2626654.68 | norm 0.3688 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21092 | train 2.993284 | val 3.295141 | tok/s 2625723.43 | norm 0.3560 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21093 | train 2.990508 | val 3.295141 | tok/s 2628436.73 | norm 0.3908 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21094 | train 2.917846 | val 3.295141 | tok/s 2631043.22 | norm 0.3577 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21095 | train 2.855737 | val 3.295141 | tok/s 2622936.00 | norm 0.3438 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21096 | train 2.963305 | val 3.295141 | tok/s 2630853.99 | norm 0.3687 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21097 | train 2.953345 | val 3.295141 | tok/s 2475028.16 | norm 0.3682 | dt 211.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21098 | train 3.000666 | val 3.295141 | tok/s 2628052.06 | norm 0.3648 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21099 | train 2.965986 | val 3.295141 | tok/s 2572295.11 | norm 0.3642 | dt 203.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21100 | train 3.028074 | val 3.295141 | tok/s 2624027.79 | norm 0.4119 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21101 | train 3.064399 | val 3.295141 | tok/s 2626372.33 | norm 0.4147 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21102 | train 2.943883 | val 3.295141 | tok/s 2624366.68 | norm 0.3524 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21103 | train 2.969035 | val 3.295141 | tok/s 2629249.99 | norm 0.3475 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21104 | train 2.946337 | val 3.295141 | tok/s 2630551.10 | norm 0.3658 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21105 | train 2.921114 | val 3.295141 | tok/s 2618625.51 | norm 0.3715 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21106 | train 2.963922 | val 3.295141 | tok/s 2629029.56 | norm 0.3551 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21107 | train 2.923267 | val 3.295141 | tok/s 2628646.59 | norm 0.3492 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21108 | train 2.957387 | val 3.295141 | tok/s 2628748.91 | norm 0.3606 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21109 | train 3.076679 | val 3.295141 | tok/s 2626295.82 | norm 0.3419 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21110 | train 2.870851 | val 3.295141 | tok/s 2625375.82 | norm 0.3562 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21111 | train 2.856716 | val 3.295141 | tok/s 2631322.81 | norm 0.3686 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21112 | train 2.847827 | val 3.295141 | tok/s 2628770.54 | norm 0.3835 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21113 | train 2.950584 | val 3.295141 | tok/s 2629254.33 | norm 0.3560 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21114 | train 2.924069 | val 3.295141 | tok/s 2630680.78 | norm 0.3896 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21115 | train 2.902699 | val 3.295141 | tok/s 2613667.59 | norm 0.3595 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21116 | train 2.893201 | val 3.295141 | tok/s 2631223.63 | norm 0.3538 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21117 | train 3.166777 | val 3.295141 | tok/s 2620663.26 | norm 0.3932 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21118 | train 2.878504 | val 3.295141 | tok/s 2620343.99 | norm 0.4018 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21119 | train 3.017916 | val 3.295141 | tok/s 2626599.51 | norm 0.3768 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21120 | train 2.883935 | val 3.295141 | tok/s 2625580.05 | norm 0.3588 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21121 | train 2.975086 | val 3.295141 | tok/s 2628614.51 | norm 0.3566 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21122 | train 2.905622 | val 3.295141 | tok/s 2626252.06 | norm 0.3768 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21123 | train 2.879631 | val 3.295141 | tok/s 2627884.64 | norm 0.4109 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21124 | train 2.891796 | val 3.295141 | tok/s 2627036.03 | norm 0.3511 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21125 | train 2.775078 | val 3.295141 | tok/s 2625477.77 | norm 0.3576 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21126 | train 2.947043 | val 3.295141 | tok/s 2631365.04 | norm 0.3559 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21127 | train 2.883983 | val 3.295141 | tok/s 2624465.67 | norm 0.3574 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21128 | train 2.933411 | val 3.295141 | tok/s 2631429.58 | norm 0.3742 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21129 | train 3.015257 | val 3.295141 | tok/s 2624710.14 | norm 0.3508 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21130 | train 2.914906 | val 3.295141 | tok/s 2613232.26 | norm 0.3438 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21131 | train 2.921207 | val 3.295141 | tok/s 2632733.92 | norm 0.3793 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21132 | train 2.952040 | val 3.295141 | tok/s 2630041.97 | norm 0.3870 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21133 | train 2.963334 | val 3.295141 | tok/s 2628571.68 | norm 0.3496 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21134 | train 3.001665 | val 3.295141 | tok/s 2627958.05 | norm 0.3363 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21135 | train 3.027166 | val 3.295141 | tok/s 2627570.91 | norm 0.3734 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21136 | train 3.037803 | val 3.295141 | tok/s 2633837.13 | norm 0.3930 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21137 | train 2.883781 | val 3.295141 | tok/s 2632184.10 | norm 0.3898 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21138 | train 2.925107 | val 3.295141 | tok/s 2628274.97 | norm 0.4105 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21139 | train 2.911821 | val 3.295141 | tok/s 2624780.81 | norm 0.4066 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21140 | train 2.758343 | val 3.295141 | tok/s 2630105.73 | norm 0.4529 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21141 | train 2.955131 | val 3.295141 | tok/s 2630990.69 | norm 0.3692 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21142 | train 2.894870 | val 3.295141 | tok/s 2628999.66 | norm 0.3686 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21143 | train 2.829011 | val 3.295141 | tok/s 2626828.01 | norm 0.3857 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21144 | train 3.205436 | val 3.295141 | tok/s 2484193.15 | norm 0.4280 | dt 211.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21145 | train 2.891761 | val 3.295141 | tok/s 2629228.67 | norm 0.3465 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21146 | train 2.986310 | val 3.295141 | tok/s 2632918.66 | norm 0.3147 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21147 | train 2.953148 | val 3.295141 | tok/s 2632462.30 | norm 0.3906 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21148 | train 3.001845 | val 3.295141 | tok/s 2631832.47 | norm 0.3773 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21149 | train 2.974691 | val 3.295141 | tok/s 2630877.56 | norm 0.3740 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21150 | train 2.881704 | val 3.295141 | tok/s 2628263.76 | norm 0.3970 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21151 | train 2.978900 | val 3.295141 | tok/s 2634278.31 | norm 0.3805 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21152 | train 3.107270 | val 3.295141 | tok/s 2631062.60 | norm 0.3451 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21153 | train 3.008082 | val 3.295141 | tok/s 2613346.97 | norm 0.3476 | dt 200.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21154 | train 2.944085 | val 3.295141 | tok/s 2633097.63 | norm 0.3512 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21155 | train 2.892931 | val 3.295141 | tok/s 2625491.62 | norm 0.3489 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21156 | train 2.918653 | val 3.295141 | tok/s 2632951.44 | norm 0.3478 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21157 | train 2.985074 | val 3.295141 | tok/s 2620790.73 | norm 0.3537 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21158 | train 3.031338 | val 3.295141 | tok/s 2630922.99 | norm 0.3627 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21159 | train 2.885651 | val 3.295141 | tok/s 2627008.71 | norm 0.3605 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21160 | train 2.888326 | val 3.295141 | tok/s 2625103.00 | norm 0.3401 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21161 | train 3.234354 | val 3.295141 | tok/s 2627414.94 | norm 0.3906 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21162 | train 2.845147 | val 3.295141 | tok/s 2628104.28 | norm 0.3792 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21163 | train 3.348245 | val 3.295141 | tok/s 2630542.70 | norm 0.4451 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21164 | train 3.004608 | val 3.295141 | tok/s 2627682.82 | norm 0.4218 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21165 | train 2.909030 | val 3.295141 | tok/s 2627894.89 | norm 0.3860 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21166 | train 2.875834 | val 3.295141 | tok/s 2620138.79 | norm 0.4155 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21167 | train 2.857932 | val 3.295141 | tok/s 2621004.56 | norm 0.4048 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21168 | train 2.908439 | val 3.295141 | tok/s 2627465.01 | norm 0.3981 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21169 | train 2.946188 | val 3.295141 | tok/s 2630616.76 | norm 0.3857 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21170 | train 2.979617 | val 3.295141 | tok/s 2626575.69 | norm 0.3847 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21171 | train 2.920528 | val 3.295141 | tok/s 2628083.80 | norm 0.3739 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21172 | train 2.940397 | val 3.295141 | tok/s 2624130.97 | norm 0.3639 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21173 | train 2.948351 | val 3.295141 | tok/s 2628969.18 | norm 0.3642 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21174 | train 2.907565 | val 3.295141 | tok/s 2627202.77 | norm 0.3710 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21175 | train 2.888301 | val 3.295141 | tok/s 2626329.74 | norm 0.3648 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21176 | train 2.992301 | val 3.295141 | tok/s 2626676.44 | norm 0.3964 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21177 | train 2.932503 | val 3.295141 | tok/s 2627187.46 | norm 0.3968 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21178 | train 3.091506 | val 3.295141 | tok/s 2632429.30 | norm 0.3812 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21179 | train 2.947646 | val 3.295141 | tok/s 2630211.37 | norm 0.3480 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21180 | train 2.965899 | val 3.295141 | tok/s 2628374.99 | norm 0.3209 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21181 | train 2.918893 | val 3.295141 | tok/s 2622770.70 | norm 0.3329 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21182 | train 2.839269 | val 3.295141 | tok/s 2622716.95 | norm 0.3656 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21183 | train 3.127852 | val 3.295141 | tok/s 2627581.24 | norm 0.3734 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21184 | train 3.109188 | val 3.295141 | tok/s 2629546.93 | norm 0.3942 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21185 | train 2.977872 | val 3.295141 | tok/s 2512185.53 | norm 0.3837 | dt 208.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21186 | train 3.072033 | val 3.295141 | tok/s 2575785.54 | norm 0.3538 | dt 203.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21187 | train 2.929142 | val 3.295141 | tok/s 2625542.48 | norm 0.3473 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21188 | train 3.085486 | val 3.295141 | tok/s 2629177.76 | norm 0.3660 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21189 | train 3.083491 | val 3.295141 | tok/s 2628640.33 | norm 0.4232 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21190 | train 2.954865 | val 3.295141 | tok/s 2625859.90 | norm 0.3613 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21191 | train 2.829976 | val 3.295141 | tok/s 2627297.24 | norm 0.3708 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21192 | train 2.944004 | val 3.295141 | tok/s 2478138.56 | norm 0.4903 | dt 211.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21193 | train 2.942705 | val 3.295141 | tok/s 2626769.70 | norm 0.3313 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21194 | train 2.955655 | val 3.295141 | tok/s 2626221.11 | norm 0.3157 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21195 | train 2.926695 | val 3.295141 | tok/s 2627362.65 | norm 0.3661 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21196 | train 2.922668 | val 3.295141 | tok/s 2627659.72 | norm 0.3949 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21197 | train 2.915479 | val 3.295141 | tok/s 2627881.74 | norm 0.3610 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21198 | train 2.904081 | val 3.295141 | tok/s 2624257.77 | norm 0.3188 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21199 | train 3.039896 | val 3.295141 | tok/s 2614137.20 | norm 0.3313 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21200 | train 2.960623 | val 3.295141 | tok/s 2630371.76 | norm 0.3453 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21201 | train 3.149161 | val 3.295141 | tok/s 2630232.63 | norm 0.3379 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21202 | train 2.761409 | val 3.295141 | tok/s 2630721.22 | norm 0.3239 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21203 | train 2.938509 | val 3.295141 | tok/s 2626948.24 | norm 0.3670 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21204 | train 2.937106 | val 3.295141 | tok/s 2620737.18 | norm 0.3733 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21205 | train 2.983552 | val 3.295141 | tok/s 2627875.15 | norm 0.3533 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21206 | train 2.874987 | val 3.295141 | tok/s 2629863.07 | norm 0.3683 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21207 | train 2.893776 | val 3.295141 | tok/s 2633643.45 | norm 0.3852 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21208 | train 3.077263 | val 3.295141 | tok/s 2624119.72 | norm 0.3528 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21209 | train 2.899365 | val 3.295141 | tok/s 2624871.48 | norm 0.3393 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21210 | train 2.939551 | val 3.295141 | tok/s 2631002.85 | norm 0.3460 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21211 | train 2.895497 | val 3.295141 | tok/s 2625783.05 | norm 0.3594 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21212 | train 3.014539 | val 3.295141 | tok/s 2631172.14 | norm 0.3748 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21213 | train 2.943831 | val 3.295141 | tok/s 2627128.05 | norm 0.3756 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21214 | train 2.984051 | val 3.295141 | tok/s 2621822.79 | norm 0.3543 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21215 | train 2.861747 | val 3.295141 | tok/s 2628851.13 | norm 0.3542 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21216 | train 2.905948 | val 3.295141 | tok/s 2620144.06 | norm 0.3656 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21217 | train 2.845009 | val 3.295141 | tok/s 2621517.65 | norm 0.3759 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21218 | train 2.911058 | val 3.295141 | tok/s 2616402.42 | norm 0.4008 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21219 | train 3.116435 | val 3.295141 | tok/s 2624654.43 | norm 0.4453 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21220 | train 2.916791 | val 3.295141 | tok/s 2630336.05 | norm 0.3672 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21221 | train 2.918288 | val 3.295141 | tok/s 2631741.40 | norm 0.3603 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21222 | train 3.025703 | val 3.295141 | tok/s 2629281.69 | norm 0.3997 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21223 | train 2.940040 | val 3.295141 | tok/s 2626861.58 | norm 0.3638 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21224 | train 2.965533 | val 3.295141 | tok/s 2623215.46 | norm 0.3484 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21225 | train 2.886050 | val 3.295141 | tok/s 2631058.66 | norm 0.3988 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21226 | train 2.927704 | val 3.295141 | tok/s 2629742.43 | norm 0.4038 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21227 | train 2.991555 | val 3.295141 | tok/s 2633047.30 | norm 0.3671 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21228 | train 2.919483 | val 3.295141 | tok/s 2624245.91 | norm 0.3638 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21229 | train 2.957986 | val 3.295141 | tok/s 2624224.84 | norm 0.4004 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21230 | train 2.905768 | val 3.295141 | tok/s 2630096.91 | norm 0.3833 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21231 | train 2.939384 | val 3.295141 | tok/s 2627817.17 | norm 0.4117 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21232 | train 2.925848 | val 3.295141 | tok/s 2627653.89 | norm 0.3822 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21233 | train 2.813415 | val 3.295141 | tok/s 2622173.41 | norm 0.3848 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21234 | train 2.896863 | val 3.295141 | tok/s 2625160.76 | norm 0.4023 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21235 | train 2.897670 | val 3.295141 | tok/s 2629659.70 | norm 0.3641 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21236 | train 2.925163 | val 3.295141 | tok/s 2624776.04 | norm 0.3521 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21237 | train 2.946019 | val 3.295141 | tok/s 2631118.34 | norm 0.3194 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21238 | train 2.979982 | val 3.295141 | tok/s 2626885.26 | norm 0.3429 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21239 | train 3.012619 | val 3.295141 | tok/s 2627787.21 | norm 0.3716 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21240 | train 3.054548 | val 3.295141 | tok/s 2490104.26 | norm 0.3890 | dt 210.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21241 | train 2.991085 | val 3.295141 | tok/s 2626451.78 | norm 0.3744 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21242 | train 2.990562 | val 3.295141 | tok/s 2524524.78 | norm 0.3814 | dt 207.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21243 | train 2.972984 | val 3.295141 | tok/s 2629043.73 | norm 0.3576 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21244 | train 2.847406 | val 3.295141 | tok/s 2632152.85 | norm 0.3358 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21245 | train 3.150681 | val 3.295141 | tok/s 2628982.34 | norm 0.4177 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21246 | train 2.890344 | val 3.295141 | tok/s 2620363.58 | norm 0.3649 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21247 | train 2.887640 | val 3.295141 | tok/s 2632708.49 | norm 0.3455 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21248 | train 2.978449 | val 3.295141 | tok/s 2630371.25 | norm 0.3875 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21249 | train 3.067353 | val 3.295141 | tok/s 2626332.37 | norm 0.3548 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 21250 | val_loss 3.301631 step 21250 | train 2.891837 | val 3.301631 | tok/s 2608094.80 | norm 0.3677 | dt 201.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21251 | train 2.940362 | val 3.301631 | tok/s 2628993.85 | norm 0.4112 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21252 | train 2.850241 | val 3.301631 | tok/s 2620026.05 | norm 0.3640 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21253 | train 2.966442 | val 3.301631 | tok/s 2630496.75 | norm 0.3657 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21254 | train 2.871569 | val 3.301631 | tok/s 2625531.10 | norm 0.4359 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21255 | train 2.875299 | val 3.301631 | tok/s 2630825.52 | norm 0.3986 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21256 | train 2.849293 | val 3.301631 | tok/s 2628168.41 | norm 0.3860 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21257 | train 2.878940 | val 3.301631 | tok/s 2621962.03 | norm 0.4504 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21258 | train 2.991362 | val 3.301631 | tok/s 2628020.59 | norm 0.3571 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21259 | train 2.887961 | val 3.301631 | tok/s 2632582.29 | norm 0.3768 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21260 | train 2.853353 | val 3.301631 | tok/s 2626820.52 | norm 0.3831 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21261 | train 3.241667 | val 3.301631 | tok/s 2623987.29 | norm 0.3751 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21262 | train 3.072297 | val 3.301631 | tok/s 2622146.64 | norm 0.3603 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21263 | train 2.886780 | val 3.301631 | tok/s 2631329.10 | norm 0.3737 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21264 | train 3.154912 | val 3.301631 | tok/s 2628961.04 | norm 0.3673 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21265 | train 2.938425 | val 3.301631 | tok/s 2628060.39 | norm 0.3967 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21266 | train 2.838586 | val 3.301631 | tok/s 2627678.27 | norm 0.3899 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21267 | train 2.942798 | val 3.301631 | tok/s 2624066.95 | norm 0.3663 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21268 | train 2.963869 | val 3.301631 | tok/s 2630382.22 | norm 0.3449 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21269 | train 3.084377 | val 3.301631 | tok/s 2624508.67 | norm 0.3579 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21270 | train 2.906465 | val 3.301631 | tok/s 2629642.16 | norm 0.3388 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21271 | train 2.984231 | val 3.301631 | tok/s 2625450.74 | norm 0.3466 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21272 | train 2.994856 | val 3.301631 | tok/s 2625001.64 | norm 0.3515 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21273 | train 3.081320 | val 3.301631 | tok/s 2627832.48 | norm 0.3873 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21274 | train 2.901796 | val 3.301631 | tok/s 2622039.96 | norm 0.3901 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21275 | train 3.011605 | val 3.301631 | tok/s 2626753.42 | norm 0.3034 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21276 | train 2.919113 | val 3.301631 | tok/s 2623842.95 | norm 0.3033 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21277 | train 2.879072 | val 3.301631 | tok/s 2623425.49 | norm 0.3503 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21278 | train 2.960274 | val 3.301631 | tok/s 2629209.33 | norm 0.3916 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21279 | train 2.857609 | val 3.301631 | tok/s 2622813.95 | norm 0.3704 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21280 | train 2.954677 | val 3.301631 | tok/s 2629243.59 | norm 0.3565 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21281 | train 2.969641 | val 3.301631 | tok/s 2630680.98 | norm 0.3553 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21282 | train 2.926893 | val 3.301631 | tok/s 2625844.91 | norm 0.3615 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21283 | train 2.794711 | val 3.301631 | tok/s 2626917.18 | norm 0.3840 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21284 | train 2.854082 | val 3.301631 | tok/s 2629212.97 | norm 0.3576 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21285 | train 2.810701 | val 3.301631 | tok/s 2629342.41 | norm 0.3734 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21286 | train 2.929119 | val 3.301631 | tok/s 2625598.61 | norm 0.4085 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21287 | train 2.883607 | val 3.301631 | tok/s 2628951.11 | norm 0.3853 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21288 | train 3.029496 | val 3.301631 | tok/s 2630347.24 | norm 0.3962 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21289 | train 2.949791 | val 3.301631 | tok/s 2627141.62 | norm 0.4126 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21290 | train 2.999090 | val 3.301631 | tok/s 2630421.65 | norm 0.3855 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21291 | train 3.004195 | val 3.301631 | tok/s 2508585.28 | norm 0.3921 | dt 209.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21292 | train 2.887280 | val 3.301631 | tok/s 2626534.66 | norm 0.3783 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21293 | train 2.956430 | val 3.301631 | tok/s 2494100.53 | norm 0.3170 | dt 210.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21294 | train 2.920646 | val 3.301631 | tok/s 2622768.46 | norm 0.3730 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21295 | train 2.873036 | val 3.301631 | tok/s 2629218.12 | norm 0.3857 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21296 | train 2.949524 | val 3.301631 | tok/s 2626793.35 | norm 0.3651 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21297 | train 2.845409 | val 3.301631 | tok/s 2630500.48 | norm 0.3570 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21298 | train 2.846678 | val 3.301631 | tok/s 2634445.07 | norm 0.3902 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21299 | train 2.877393 | val 3.301631 | tok/s 2622091.24 | norm 0.3990 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21300 | train 2.983756 | val 3.301631 | tok/s 2633104.85 | norm 0.3634 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21301 | train 2.901930 | val 3.301631 | tok/s 2632868.45 | norm 0.3617 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21302 | train 2.894771 | val 3.301631 | tok/s 2626613.75 | norm 0.3826 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21303 | train 2.962572 | val 3.301631 | tok/s 2623406.65 | norm 0.3716 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21304 | train 2.937007 | val 3.301631 | tok/s 2619459.49 | norm 0.3596 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21305 | train 2.891306 | val 3.301631 | tok/s 2633307.94 | norm 0.3991 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21306 | train 2.928244 | val 3.301631 | tok/s 2630038.86 | norm 0.3991 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21307 | train 2.987821 | val 3.301631 | tok/s 2628866.68 | norm 0.3772 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21308 | train 2.976240 | val 3.301631 | tok/s 2621553.54 | norm 0.3807 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21309 | train 2.920910 | val 3.301631 | tok/s 2624985.07 | norm 0.3488 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21310 | train 3.158840 | val 3.301631 | tok/s 2633246.42 | norm 0.3821 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21311 | train 2.868201 | val 3.301631 | tok/s 2628034.65 | norm 0.3616 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21312 | train 2.923121 | val 3.301631 | tok/s 2630872.84 | norm 0.3581 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21313 | train 2.867548 | val 3.301631 | tok/s 2621496.72 | norm 0.3485 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21314 | train 2.915819 | val 3.301631 | tok/s 2625163.35 | norm 0.3631 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21315 | train 2.930248 | val 3.301631 | tok/s 2632842.41 | norm 0.3960 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21316 | train 3.056841 | val 3.301631 | tok/s 2621822.89 | norm 0.3780 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21317 | train 3.043553 | val 3.301631 | tok/s 2628901.73 | norm 0.3423 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21318 | train 3.006041 | val 3.301631 | tok/s 2620286.20 | norm 0.3556 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21319 | train 2.894720 | val 3.301631 | tok/s 2621485.80 | norm 0.3687 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21320 | train 2.866071 | val 3.301631 | tok/s 2632383.25 | norm 0.3716 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21321 | train 2.914704 | val 3.301631 | tok/s 2627725.36 | norm 0.3829 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21322 | train 2.983262 | val 3.301631 | tok/s 2631174.24 | norm 0.3623 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21323 | train 3.020296 | val 3.301631 | tok/s 2626016.04 | norm 0.3327 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21324 | train 2.860173 | val 3.301631 | tok/s 2627498.24 | norm 0.3380 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21325 | train 3.021031 | val 3.301631 | tok/s 2628566.77 | norm 0.3606 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21326 | train 2.939151 | val 3.301631 | tok/s 2628584.89 | norm 0.3251 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21327 | train 3.001063 | val 3.301631 | tok/s 2627717.55 | norm 0.3333 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21328 | train 3.046517 | val 3.301631 | tok/s 2623647.97 | norm 0.4037 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21329 | train 2.977044 | val 3.301631 | tok/s 2624826.98 | norm 0.3632 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21330 | train 2.903246 | val 3.301631 | tok/s 2631249.06 | norm 0.3606 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21331 | train 2.918524 | val 3.301631 | tok/s 2629336.55 | norm 0.3779 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21332 | train 2.936136 | val 3.301631 | tok/s 2631030.99 | norm 0.3716 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21333 | train 2.996983 | val 3.301631 | tok/s 2626749.13 | norm 0.3408 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21334 | train 2.928427 | val 3.301631 | tok/s 2542316.62 | norm 0.3604 | dt 206.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21335 | train 2.807961 | val 3.301631 | tok/s 2522332.73 | norm 0.3641 | dt 207.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21336 | train 3.018592 | val 3.301631 | tok/s 2623121.11 | norm 0.3516 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21337 | train 3.044184 | val 3.301631 | tok/s 2630351.62 | norm 0.3559 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21338 | train 2.896460 | val 3.301631 | tok/s 2625281.40 | norm 0.3622 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21339 | train 2.971440 | val 3.301631 | tok/s 2626910.13 | norm 0.3359 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21340 | train 3.010624 | val 3.301631 | tok/s 2630029.03 | norm 0.3708 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21341 | train 3.016104 | val 3.301631 | tok/s 2619148.44 | norm 0.4060 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21342 | train 2.988687 | val 3.301631 | tok/s 2631929.48 | norm 0.3757 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21343 | train 2.885035 | val 3.301631 | tok/s 2624051.29 | norm 0.3478 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21344 | train 2.837101 | val 3.301631 | tok/s 2629933.98 | norm 0.3594 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21345 | train 3.008344 | val 3.301631 | tok/s 2630195.14 | norm 0.3631 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21346 | train 2.956394 | val 3.301631 | tok/s 2624447.29 | norm 0.3943 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21347 | train 2.932572 | val 3.301631 | tok/s 2630538.02 | norm 0.3720 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21348 | train 2.923944 | val 3.301631 | tok/s 2629342.57 | norm 0.3933 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21349 | train 2.945011 | val 3.301631 | tok/s 2626932.10 | norm 0.4439 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21350 | train 2.985993 | val 3.301631 | tok/s 2635263.10 | norm 0.3238 | dt 198.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21351 | train 2.928156 | val 3.301631 | tok/s 2619226.45 | norm 0.3355 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21352 | train 2.914955 | val 3.301631 | tok/s 2630812.52 | norm 0.3762 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21353 | train 2.958225 | val 3.301631 | tok/s 2610630.29 | norm 0.3685 | dt 200.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21354 | train 2.934876 | val 3.301631 | tok/s 2627549.58 | norm 0.3520 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21355 | train 2.976337 | val 3.301631 | tok/s 2626654.34 | norm 0.3658 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21356 | train 2.954620 | val 3.301631 | tok/s 2624241.17 | norm 0.3477 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21357 | train 2.958582 | val 3.301631 | tok/s 2631131.65 | norm 0.3275 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21358 | train 2.927834 | val 3.301631 | tok/s 2629148.39 | norm 0.3562 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21359 | train 2.901992 | val 3.301631 | tok/s 2629985.89 | norm 0.3383 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21360 | train 3.004937 | val 3.301631 | tok/s 2619889.06 | norm 0.3659 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21361 | train 2.961953 | val 3.301631 | tok/s 2623833.72 | norm 0.3777 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21362 | train 2.946263 | val 3.301631 | tok/s 2625983.70 | norm 0.3760 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21363 | train 2.977132 | val 3.301631 | tok/s 2628159.02 | norm 0.3722 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21364 | train 3.074183 | val 3.301631 | tok/s 2630507.07 | norm 0.3609 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21365 | train 2.924615 | val 3.301631 | tok/s 2620945.15 | norm 0.4006 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21366 | train 2.924856 | val 3.301631 | tok/s 2622134.34 | norm 0.3876 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21367 | train 2.944995 | val 3.301631 | tok/s 2630356.13 | norm 0.3645 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21368 | train 2.935146 | val 3.301631 | tok/s 2622420.86 | norm 0.3719 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21369 | train 2.955926 | val 3.301631 | tok/s 2630378.46 | norm 0.3604 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21370 | train 2.930518 | val 3.301631 | tok/s 2626141.36 | norm 0.3344 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21371 | train 2.867558 | val 3.301631 | tok/s 2628077.22 | norm 0.3633 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21372 | train 2.977888 | val 3.301631 | tok/s 2630446.41 | norm 0.3452 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21373 | train 2.923077 | val 3.301631 | tok/s 2623562.02 | norm 0.3479 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21374 | train 2.940692 | val 3.301631 | tok/s 2631137.68 | norm 0.3720 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21375 | train 2.883860 | val 3.301631 | tok/s 2624171.47 | norm 0.3753 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21376 | train 3.169086 | val 3.301631 | tok/s 2605188.86 | norm 0.4088 | dt 201.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21377 | train 2.857113 | val 3.301631 | tok/s 2622880.33 | norm 0.3645 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21378 | train 2.846345 | val 3.301631 | tok/s 2624143.04 | norm 0.3400 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21379 | train 3.010554 | val 3.301631 | tok/s 2626490.01 | norm 0.3653 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21380 | train 2.954110 | val 3.301631 | tok/s 2614757.24 | norm 0.3566 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21381 | train 2.920348 | val 3.301631 | tok/s 2629090.63 | norm 0.3768 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21382 | train 2.976566 | val 3.301631 | tok/s 2631713.25 | norm 0.4063 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21383 | train 3.036796 | val 3.301631 | tok/s 2629275.36 | norm 0.4100 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21384 | train 2.855489 | val 3.301631 | tok/s 2624544.00 | norm 0.3900 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21385 | train 2.877791 | val 3.301631 | tok/s 2624635.78 | norm 0.4282 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21386 | train 2.978081 | val 3.301631 | tok/s 2625140.32 | norm 0.3656 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21387 | train 3.561622 | val 3.301631 | tok/s 2627658.72 | norm 0.4698 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21388 | train 3.049556 | val 3.301631 | tok/s 2627835.56 | norm 0.3950 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21389 | train 2.949875 | val 3.301631 | tok/s 2621048.91 | norm 0.3572 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21390 | train 2.913306 | val 3.301631 | tok/s 2620351.18 | norm 0.3526 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21391 | train 2.986270 | val 3.301631 | tok/s 2631012.93 | norm 0.3497 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21392 | train 3.034711 | val 3.301631 | tok/s 2629424.46 | norm 0.3741 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21393 | train 2.906473 | val 3.301631 | tok/s 2632893.56 | norm 0.3707 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21394 | train 2.983197 | val 3.301631 | tok/s 2624151.13 | norm 0.3557 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21395 | train 2.991199 | val 3.301631 | tok/s 2622573.76 | norm 0.4956 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21396 | train 3.052562 | val 3.301631 | tok/s 2630617.05 | norm 0.4291 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21397 | train 3.166034 | val 3.301631 | tok/s 2627323.22 | norm 0.4969 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21398 | train 2.915634 | val 3.301631 | tok/s 2626204.74 | norm 0.3833 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21399 | train 2.879305 | val 3.301631 | tok/s 2619403.68 | norm 0.3496 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21400 | train 2.925863 | val 3.301631 | tok/s 2625602.42 | norm 0.4331 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21401 | train 3.022864 | val 3.301631 | tok/s 2630667.14 | norm 0.4128 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21402 | train 2.991861 | val 3.301631 | tok/s 2625851.97 | norm 0.3534 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21403 | train 2.869380 | val 3.301631 | tok/s 2630250.66 | norm 0.3404 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21404 | train 3.029212 | val 3.301631 | tok/s 2600543.46 | norm 0.3495 | dt 201.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21405 | train 2.899703 | val 3.301631 | tok/s 2568254.78 | norm 0.3623 | dt 204.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21406 | train 2.983305 | val 3.301631 | tok/s 2499538.15 | norm 0.3750 | dt 209.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21407 | train 2.901418 | val 3.301631 | tok/s 2619438.67 | norm 0.3830 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21408 | train 2.920069 | val 3.301631 | tok/s 2623753.60 | norm 0.3683 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21409 | train 3.008149 | val 3.301631 | tok/s 2623821.20 | norm 0.3678 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21410 | train 2.914395 | val 3.301631 | tok/s 2629142.84 | norm 0.3768 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21411 | train 3.006566 | val 3.301631 | tok/s 2627318.81 | norm 0.4028 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21412 | train 2.964463 | val 3.301631 | tok/s 2619109.79 | norm 0.3885 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21413 | train 2.868394 | val 3.301631 | tok/s 2628965.97 | norm 0.3417 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21414 | train 3.070914 | val 3.301631 | tok/s 2630282.00 | norm 0.3507 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21415 | train 2.873753 | val 3.301631 | tok/s 2627656.02 | norm 0.3696 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21416 | train 2.956985 | val 3.301631 | tok/s 2592195.90 | norm 0.3729 | dt 202.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21417 | train 2.933430 | val 3.301631 | tok/s 2629839.69 | norm 0.3703 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21418 | train 2.902804 | val 3.301631 | tok/s 2631307.98 | norm 0.3714 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21419 | train 2.948687 | val 3.301631 | tok/s 2629641.02 | norm 0.4096 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21420 | train 2.919204 | val 3.301631 | tok/s 2629068.91 | norm 0.3878 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21421 | train 3.020870 | val 3.301631 | tok/s 2625490.42 | norm 0.3409 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21422 | train 3.088245 | val 3.301631 | tok/s 2620817.10 | norm 0.3859 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21423 | train 2.978947 | val 3.301631 | tok/s 2612260.54 | norm 0.3818 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21424 | train 2.874042 | val 3.301631 | tok/s 2621723.66 | norm 0.3676 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21425 | train 2.894324 | val 3.301631 | tok/s 2625170.30 | norm 0.3728 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21426 | train 2.902791 | val 3.301631 | tok/s 2624551.20 | norm 0.3776 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21427 | train 2.943531 | val 3.301631 | tok/s 2619049.75 | norm 0.3809 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21428 | train 2.986725 | val 3.301631 | tok/s 2621858.23 | norm 0.3664 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21429 | train 2.904722 | val 3.301631 | tok/s 2625716.49 | norm 0.3516 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21430 | train 3.015516 | val 3.301631 | tok/s 2625406.23 | norm 0.3378 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21431 | train 2.867633 | val 3.301631 | tok/s 2616887.13 | norm 0.3502 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21432 | train 2.984493 | val 3.301631 | tok/s 2621877.33 | norm 0.3526 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21433 | train 2.930277 | val 3.301631 | tok/s 2627074.73 | norm 0.3446 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21434 | train 2.895169 | val 3.301631 | tok/s 2622861.22 | norm 0.3712 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21435 | train 2.901358 | val 3.301631 | tok/s 2627247.15 | norm 0.3670 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21436 | train 2.916482 | val 3.301631 | tok/s 2617280.28 | norm 0.3281 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21437 | train 3.038843 | val 3.301631 | tok/s 2621877.49 | norm 0.3818 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21438 | train 3.002163 | val 3.301631 | tok/s 2628659.51 | norm 0.3730 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21439 | train 2.914510 | val 3.301631 | tok/s 2620622.75 | norm 0.3204 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21440 | train 2.955157 | val 3.301631 | tok/s 2627073.46 | norm 0.3239 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21441 | train 3.059286 | val 3.301631 | tok/s 2625087.45 | norm 0.3569 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21442 | train 2.883654 | val 3.301631 | tok/s 2619203.58 | norm 0.3771 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21443 | train 2.893593 | val 3.301631 | tok/s 2630678.23 | norm 0.3847 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21444 | train 2.944205 | val 3.301631 | tok/s 2620009.89 | norm 0.3870 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21445 | train 3.014200 | val 3.301631 | tok/s 2621553.16 | norm 0.3893 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21446 | train 2.988148 | val 3.301631 | tok/s 2625467.91 | norm 0.4038 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21447 | train 2.809077 | val 3.301631 | tok/s 2625967.08 | norm 0.3533 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21448 | train 2.933794 | val 3.301631 | tok/s 2629230.58 | norm 0.3451 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21449 | train 2.884013 | val 3.301631 | tok/s 2625692.14 | norm 0.3401 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21450 | train 3.041952 | val 3.301631 | tok/s 2628682.41 | norm 0.3798 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21451 | train 2.990633 | val 3.301631 | tok/s 2627564.83 | norm 0.3737 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21452 | train 2.976871 | val 3.301631 | tok/s 2624381.79 | norm 0.3621 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21453 | train 3.157296 | val 3.301631 | tok/s 2627005.80 | norm 0.3915 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21454 | train 2.884018 | val 3.301631 | tok/s 2621385.65 | norm 0.3789 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21455 | train 3.009312 | val 3.301631 | tok/s 2627760.38 | norm 0.3834 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21456 | train 2.881056 | val 3.301631 | tok/s 2625100.68 | norm 0.3429 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21457 | train 3.044689 | val 3.301631 | tok/s 2492965.03 | norm 0.3360 | dt 210.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21458 | train 2.959710 | val 3.301631 | tok/s 2625556.86 | norm 0.3911 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21459 | train 2.881401 | val 3.301631 | tok/s 2489297.46 | norm 0.3868 | dt 210.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21460 | train 2.977270 | val 3.301631 | tok/s 2625502.93 | norm 0.3055 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21461 | train 2.947299 | val 3.301631 | tok/s 2627676.26 | norm 0.3349 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21462 | train 2.993666 | val 3.301631 | tok/s 2627285.84 | norm 0.3794 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21463 | train 2.885334 | val 3.301631 | tok/s 2627478.51 | norm 0.3419 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21464 | train 2.917526 | val 3.301631 | tok/s 2618328.10 | norm 0.3501 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21465 | train 3.088136 | val 3.301631 | tok/s 2625317.75 | norm 0.3797 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21466 | train 2.985749 | val 3.301631 | tok/s 2629830.22 | norm 0.3506 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21467 | train 2.968418 | val 3.301631 | tok/s 2615212.59 | norm 0.3884 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21468 | train 2.961366 | val 3.301631 | tok/s 2628401.76 | norm 0.3848 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21469 | train 3.055591 | val 3.301631 | tok/s 2621305.76 | norm 0.3677 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21470 | train 2.937452 | val 3.301631 | tok/s 2623151.23 | norm 0.3902 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21471 | train 2.829771 | val 3.301631 | tok/s 2626220.85 | norm 0.3935 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21472 | train 2.970710 | val 3.301631 | tok/s 2620160.56 | norm 0.3994 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21473 | train 3.109500 | val 3.301631 | tok/s 2619022.63 | norm 0.3876 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21474 | train 3.041648 | val 3.301631 | tok/s 2620495.34 | norm 0.3566 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21475 | train 2.958207 | val 3.301631 | tok/s 2629403.79 | norm 0.3307 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21476 | train 2.850641 | val 3.301631 | tok/s 2623330.63 | norm 0.3465 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21477 | train 2.869885 | val 3.301631 | tok/s 2628890.37 | norm 0.3845 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21478 | train 2.933130 | val 3.301631 | tok/s 2622244.23 | norm 0.3735 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21479 | train 2.872360 | val 3.301631 | tok/s 2623074.25 | norm 0.3677 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21480 | train 3.049591 | val 3.301631 | tok/s 2629622.68 | norm 0.3915 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21481 | train 2.789104 | val 3.301631 | tok/s 2625029.58 | norm 0.3482 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21482 | train 2.986969 | val 3.301631 | tok/s 2626399.59 | norm 0.4020 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21483 | train 3.001689 | val 3.301631 | tok/s 2627525.37 | norm 0.3944 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21484 | train 2.953282 | val 3.301631 | tok/s 2625685.37 | norm 0.3568 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21485 | train 2.947815 | val 3.301631 | tok/s 2628844.25 | norm 0.3701 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21486 | train 3.049944 | val 3.301631 | tok/s 2624304.34 | norm 0.3998 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21487 | train 2.906906 | val 3.301631 | tok/s 2631196.08 | norm 0.3814 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21488 | train 2.878166 | val 3.301631 | tok/s 2628132.49 | norm 0.3682 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21489 | train 2.905367 | val 3.301631 | tok/s 2623562.90 | norm 0.3658 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21490 | train 2.885271 | val 3.301631 | tok/s 2586775.11 | norm 0.3464 | dt 202.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21491 | train 2.918381 | val 3.301631 | tok/s 2622766.62 | norm 0.3565 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21492 | train 2.946720 | val 3.301631 | tok/s 2548778.13 | norm 0.3651 | dt 205.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21493 | train 2.940641 | val 3.301631 | tok/s 2630036.95 | norm 0.3859 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21494 | train 2.976688 | val 3.301631 | tok/s 2626436.70 | norm 0.3881 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21495 | train 2.879383 | val 3.301631 | tok/s 2625823.10 | norm 0.3710 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21496 | train 2.879310 | val 3.301631 | tok/s 2621568.47 | norm 0.3797 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21497 | train 3.025039 | val 3.301631 | tok/s 2628540.00 | norm 0.3739 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21498 | train 2.944797 | val 3.301631 | tok/s 2627407.62 | norm 0.3605 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21499 | train 2.910228 | val 3.301631 | tok/s 2622988.98 | norm 0.3724 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 21500 | val_loss 3.304301 step 21500 | train 2.833159 | val 3.304301 | tok/s 2616616.07 | norm 0.3651 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21501 | train 3.020198 | val 3.304301 | tok/s 2626712.61 | norm 0.3630 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21502 | train 2.902593 | val 3.304301 | tok/s 2621983.37 | norm 0.3445 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21503 | train 2.865555 | val 3.304301 | tok/s 2626202.68 | norm 0.3297 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21504 | train 2.848679 | val 3.304301 | tok/s 2623725.23 | norm 0.3605 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21505 | train 2.954476 | val 3.304301 | tok/s 2625042.75 | norm 0.3694 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21506 | train 2.840773 | val 3.304301 | tok/s 2628962.82 | norm 0.3857 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21507 | train 2.928838 | val 3.304301 | tok/s 2621424.86 | norm 0.3417 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21508 | train 2.845531 | val 3.304301 | tok/s 2484471.28 | norm 0.3363 | dt 211.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21509 | train 2.903756 | val 3.304301 | tok/s 2631461.04 | norm 0.3654 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21510 | train 2.910600 | val 3.304301 | tok/s 2629187.90 | norm 0.3812 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21511 | train 2.936201 | val 3.304301 | tok/s 2626356.16 | norm 0.3694 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21512 | train 2.885947 | val 3.304301 | tok/s 2622946.69 | norm 0.3693 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21513 | train 2.956696 | val 3.304301 | tok/s 2630727.45 | norm 0.3769 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21514 | train 2.864452 | val 3.304301 | tok/s 2626885.48 | norm 0.3948 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21515 | train 2.919296 | val 3.304301 | tok/s 2626546.32 | norm 0.3941 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21516 | train 2.914977 | val 3.304301 | tok/s 2625525.74 | norm 0.3501 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21517 | train 2.969314 | val 3.304301 | tok/s 2618128.49 | norm 0.3811 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21518 | train 2.930297 | val 3.304301 | tok/s 2629137.35 | norm 0.3543 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21519 | train 2.931144 | val 3.304301 | tok/s 2629105.08 | norm 0.3422 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21520 | train 2.816355 | val 3.304301 | tok/s 2630397.89 | norm 0.3933 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21521 | train 2.826330 | val 3.304301 | tok/s 2617390.83 | norm 0.4065 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21522 | train 3.038737 | val 3.304301 | tok/s 2621651.09 | norm 0.3951 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21523 | train 2.933515 | val 3.304301 | tok/s 2627703.93 | norm 0.3637 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21524 | train 2.784177 | val 3.304301 | tok/s 2626536.68 | norm 0.3966 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21525 | train 2.945031 | val 3.304301 | tok/s 2626901.71 | norm 0.4482 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21526 | train 2.966272 | val 3.304301 | tok/s 2621721.18 | norm 0.3565 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21527 | train 2.896771 | val 3.304301 | tok/s 2625888.73 | norm 0.3674 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21528 | train 3.022953 | val 3.304301 | tok/s 2624306.60 | norm 0.3612 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21529 | train 2.918432 | val 3.304301 | tok/s 2628219.56 | norm 0.3493 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21530 | train 2.848049 | val 3.304301 | tok/s 2628202.73 | norm 0.3463 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21531 | train 2.866447 | val 3.304301 | tok/s 2623918.04 | norm 0.3778 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21532 | train 3.064439 | val 3.304301 | tok/s 2623315.83 | norm 0.4169 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21533 | train 2.838059 | val 3.304301 | tok/s 2623152.22 | norm 0.3967 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21534 | train 2.936135 | val 3.304301 | tok/s 2625032.40 | norm 0.3549 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21535 | train 3.006376 | val 3.304301 | tok/s 2623719.51 | norm 0.3566 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21536 | train 2.857957 | val 3.304301 | tok/s 2627710.36 | norm 0.3774 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21537 | train 3.062758 | val 3.304301 | tok/s 2626293.73 | norm 0.3853 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21538 | train 3.129419 | val 3.304301 | tok/s 2630943.65 | norm 0.4114 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21539 | train 2.904148 | val 3.304301 | tok/s 2626135.81 | norm 0.3842 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21540 | train 2.926694 | val 3.304301 | tok/s 2627597.84 | norm 0.3897 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21541 | train 2.961108 | val 3.304301 | tok/s 2620084.16 | norm 0.3916 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21542 | train 2.965582 | val 3.304301 | tok/s 2619996.97 | norm 0.4441 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21543 | train 2.987504 | val 3.304301 | tok/s 2625111.48 | norm 0.3546 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21544 | train 3.110992 | val 3.304301 | tok/s 2625301.73 | norm 0.4575 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21545 | train 2.937595 | val 3.304301 | tok/s 2627173.77 | norm 0.3262 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21546 | train 2.883880 | val 3.304301 | tok/s 2624290.71 | norm 0.3678 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21547 | train 2.946096 | val 3.304301 | tok/s 2626233.69 | norm 0.4167 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21548 | train 3.136631 | val 3.304301 | tok/s 2626804.26 | norm 0.4728 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21549 | train 2.845677 | val 3.304301 | tok/s 2628099.93 | norm 0.3785 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21550 | train 2.930360 | val 3.304301 | tok/s 2621622.14 | norm 0.3404 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21551 | train 2.960196 | val 3.304301 | tok/s 2624310.38 | norm 0.3622 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21552 | train 2.933465 | val 3.304301 | tok/s 2628430.05 | norm 0.3314 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21553 | train 2.825526 | val 3.304301 | tok/s 2628017.07 | norm 0.3312 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21554 | train 2.971835 | val 3.304301 | tok/s 2627034.28 | norm 0.3482 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21555 | train 2.867412 | val 3.304301 | tok/s 2491462.68 | norm 0.3509 | dt 210.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21556 | train 2.905278 | val 3.304301 | tok/s 2624585.28 | norm 0.3578 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21557 | train 3.110654 | val 3.304301 | tok/s 2624411.03 | norm 0.3926 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21558 | train 2.924253 | val 3.304301 | tok/s 2626467.29 | norm 0.3721 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21559 | train 2.876435 | val 3.304301 | tok/s 2622427.63 | norm 0.3512 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21560 | train 3.022647 | val 3.304301 | tok/s 2632129.21 | norm 0.4020 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21561 | train 3.015388 | val 3.304301 | tok/s 2629027.80 | norm 0.4065 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21562 | train 2.950504 | val 3.304301 | tok/s 2625564.37 | norm 0.3759 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21563 | train 2.925555 | val 3.304301 | tok/s 2607746.14 | norm 0.3812 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21564 | train 2.922007 | val 3.304301 | tok/s 2620321.28 | norm 0.3512 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21565 | train 2.864539 | val 3.304301 | tok/s 2628672.13 | norm 0.3348 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21566 | train 2.910926 | val 3.304301 | tok/s 2623351.75 | norm 0.3806 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21567 | train 2.948510 | val 3.304301 | tok/s 2625972.01 | norm 0.3671 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21568 | train 2.910453 | val 3.304301 | tok/s 2619841.88 | norm 0.3520 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21569 | train 2.955443 | val 3.304301 | tok/s 2612723.68 | norm 0.3637 | dt 200.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21570 | train 3.026136 | val 3.304301 | tok/s 2628291.56 | norm 0.3743 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21571 | train 3.043810 | val 3.304301 | tok/s 2629668.42 | norm 0.3831 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21572 | train 2.909697 | val 3.304301 | tok/s 2626832.86 | norm 0.3631 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21573 | train 2.984134 | val 3.304301 | tok/s 2620962.97 | norm 0.3234 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21574 | train 2.921342 | val 3.304301 | tok/s 2622600.24 | norm 0.3629 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21575 | train 3.109125 | val 3.304301 | tok/s 2627900.60 | norm 0.4514 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21576 | train 2.834687 | val 3.304301 | tok/s 2627672.21 | norm 0.3710 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21577 | train 2.940901 | val 3.304301 | tok/s 2630486.63 | norm 0.3587 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21578 | train 2.873845 | val 3.304301 | tok/s 2622025.71 | norm 0.3614 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21579 | train 2.890629 | val 3.304301 | tok/s 2622659.59 | norm 0.3743 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21580 | train 2.909347 | val 3.304301 | tok/s 2628026.42 | norm 0.4060 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21581 | train 3.026102 | val 3.304301 | tok/s 2625812.25 | norm 0.3728 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21582 | train 3.009931 | val 3.304301 | tok/s 2626339.90 | norm 0.3574 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21583 | train 2.960885 | val 3.304301 | tok/s 2621828.94 | norm 0.3755 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21584 | train 2.998893 | val 3.304301 | tok/s 2624832.66 | norm 0.3412 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21585 | train 3.115016 | val 3.304301 | tok/s 2626921.22 | norm 0.3620 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21586 | train 2.935192 | val 3.304301 | tok/s 2624353.91 | norm 0.3637 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21587 | train 2.914117 | val 3.304301 | tok/s 2622093.85 | norm 0.3466 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21588 | train 2.857833 | val 3.304301 | tok/s 2622515.72 | norm 0.3977 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21589 | train 3.080745 | val 3.304301 | tok/s 2625389.29 | norm 0.3587 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21590 | train 2.870425 | val 3.304301 | tok/s 2623496.89 | norm 0.3366 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21591 | train 3.067838 | val 3.304301 | tok/s 2624228.87 | norm 0.3864 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21592 | train 2.957968 | val 3.304301 | tok/s 2625534.95 | norm 0.3841 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21593 | train 2.911209 | val 3.304301 | tok/s 2616002.61 | norm 0.3464 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21594 | train 2.851214 | val 3.304301 | tok/s 2624581.87 | norm 0.3396 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21595 | train 2.965339 | val 3.304301 | tok/s 2530862.75 | norm 0.3554 | dt 207.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21596 | train 2.844033 | val 3.304301 | tok/s 2625034.17 | norm 0.3250 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21597 | train 3.062239 | val 3.304301 | tok/s 2629598.42 | norm 0.3256 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21598 | train 2.972894 | val 3.304301 | tok/s 2622782.93 | norm 0.3779 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21599 | train 2.933713 | val 3.304301 | tok/s 2622090.83 | norm 0.3701 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21600 | train 2.912674 | val 3.304301 | tok/s 2629604.70 | norm 0.3532 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21601 | train 2.864967 | val 3.304301 | tok/s 2620781.91 | norm 0.3760 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21602 | train 2.816588 | val 3.304301 | tok/s 2624146.07 | norm 0.3975 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21603 | train 2.954777 | val 3.304301 | tok/s 2597396.49 | norm 0.3818 | dt 201.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21604 | train 2.978301 | val 3.304301 | tok/s 2627012.48 | norm 0.3909 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21605 | train 2.869393 | val 3.304301 | tok/s 2628360.03 | norm 0.3750 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21606 | train 2.983181 | val 3.304301 | tok/s 2621657.73 | norm 0.3688 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21607 | train 3.001455 | val 3.304301 | tok/s 2627379.88 | norm 0.3464 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21608 | train 2.917649 | val 3.304301 | tok/s 2622676.49 | norm 0.3629 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21609 | train 2.978193 | val 3.304301 | tok/s 2627651.44 | norm 0.3716 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21610 | train 2.819119 | val 3.304301 | tok/s 2590522.61 | norm 0.3593 | dt 202.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21611 | train 2.940317 | val 3.304301 | tok/s 2625651.19 | norm 0.3533 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21612 | train 2.917178 | val 3.304301 | tok/s 2630543.07 | norm 0.3533 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21613 | train 3.047225 | val 3.304301 | tok/s 2624116.95 | norm 0.3527 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21614 | train 3.221598 | val 3.304301 | tok/s 2628287.43 | norm 0.5705 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21615 | train 2.808493 | val 3.304301 | tok/s 2624399.33 | norm 0.3147 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21616 | train 2.888434 | val 3.304301 | tok/s 2622203.93 | norm 0.3140 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21617 | train 2.918346 | val 3.304301 | tok/s 2622171.02 | norm 0.3493 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21618 | train 2.954163 | val 3.304301 | tok/s 2625584.95 | norm 0.3470 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21619 | train 2.947004 | val 3.304301 | tok/s 2628443.00 | norm 0.3363 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21620 | train 2.897344 | val 3.304301 | tok/s 2622691.50 | norm 0.3568 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21621 | train 2.867680 | val 3.304301 | tok/s 2624483.71 | norm 0.3634 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21622 | train 3.028749 | val 3.304301 | tok/s 2627566.08 | norm 0.3965 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21623 | train 2.829133 | val 3.304301 | tok/s 2628799.05 | norm 0.3588 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21624 | train 3.036534 | val 3.304301 | tok/s 2630901.23 | norm 0.6474 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21625 | train 2.794353 | val 3.304301 | tok/s 2629244.73 | norm 0.3784 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21626 | train 2.962948 | val 3.304301 | tok/s 2622750.67 | norm 0.3697 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21627 | train 3.026903 | val 3.304301 | tok/s 2630584.50 | norm 0.3805 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21628 | train 3.001151 | val 3.304301 | tok/s 2629920.81 | norm 0.3532 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21629 | train 2.801359 | val 3.304301 | tok/s 2631105.18 | norm 0.3573 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21630 | train 2.842577 | val 3.304301 | tok/s 2621120.14 | norm 0.3435 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21631 | train 2.943832 | val 3.304301 | tok/s 2624576.47 | norm 0.3159 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21632 | train 2.997866 | val 3.304301 | tok/s 2629134.70 | norm 0.3363 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21633 | train 2.919750 | val 3.304301 | tok/s 2624469.85 | norm 0.3612 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21634 | train 2.874815 | val 3.304301 | tok/s 2626848.47 | norm 0.3806 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21635 | train 3.005081 | val 3.304301 | tok/s 2625974.15 | norm 0.3819 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21636 | train 2.907490 | val 3.304301 | tok/s 2625604.14 | norm 0.3777 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21637 | train 2.948176 | val 3.304301 | tok/s 2631203.08 | norm 0.3483 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21638 | train 2.899366 | val 3.304301 | tok/s 2628414.18 | norm 0.3168 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21639 | train 2.912031 | val 3.304301 | tok/s 2629324.62 | norm 0.3518 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21640 | train 2.806887 | val 3.304301 | tok/s 2627676.24 | norm 0.4163 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21641 | train 2.881841 | val 3.304301 | tok/s 2627169.38 | norm 0.3918 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21642 | train 2.810341 | val 3.304301 | tok/s 2628534.49 | norm 0.3880 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21643 | train 2.835378 | val 3.304301 | tok/s 2626992.66 | norm 0.3919 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21644 | train 2.861353 | val 3.304301 | tok/s 2622985.76 | norm 0.3869 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21645 | train 2.903088 | val 3.304301 | tok/s 2628490.70 | norm 0.3598 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21646 | train 2.998906 | val 3.304301 | tok/s 2616579.95 | norm 0.3527 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21647 | train 2.927593 | val 3.304301 | tok/s 2626417.82 | norm 0.3605 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21648 | train 2.984084 | val 3.304301 | tok/s 2620742.79 | norm 0.3514 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21649 | train 2.917610 | val 3.304301 | tok/s 2632798.08 | norm 0.3655 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21650 | train 2.923117 | val 3.304301 | tok/s 2630922.05 | norm 0.3642 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21651 | train 2.999176 | val 3.304301 | tok/s 2633259.17 | norm 0.3681 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21652 | train 2.933377 | val 3.304301 | tok/s 2629492.58 | norm 0.3411 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21653 | train 2.925660 | val 3.304301 | tok/s 2574629.01 | norm 0.3357 | dt 203.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21654 | train 2.931278 | val 3.304301 | tok/s 2632266.84 | norm 0.3657 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21655 | train 2.955411 | val 3.304301 | tok/s 2629839.09 | norm 0.3692 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21656 | train 2.910020 | val 3.304301 | tok/s 2628083.73 | norm 0.3496 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21657 | train 2.879131 | val 3.304301 | tok/s 2623104.50 | norm 0.3508 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21658 | train 2.850893 | val 3.304301 | tok/s 2601002.76 | norm 0.3707 | dt 201.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21659 | train 2.937964 | val 3.304301 | tok/s 2631254.13 | norm 0.3509 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21660 | train 2.874455 | val 3.304301 | tok/s 2631945.45 | norm 0.3448 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21661 | train 2.860858 | val 3.304301 | tok/s 2623834.58 | norm 0.3342 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21662 | train 2.974200 | val 3.304301 | tok/s 2628457.46 | norm 0.3318 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21663 | train 2.946543 | val 3.304301 | tok/s 2625350.58 | norm 0.3490 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21664 | train 2.884234 | val 3.304301 | tok/s 2630276.55 | norm 0.3623 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21665 | train 2.953484 | val 3.304301 | tok/s 2626991.92 | norm 0.3495 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21666 | train 2.935739 | val 3.304301 | tok/s 2627986.96 | norm 0.3695 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21667 | train 2.963354 | val 3.304301 | tok/s 2629295.68 | norm 0.4003 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21668 | train 2.954364 | val 3.304301 | tok/s 2627271.07 | norm 0.3884 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21669 | train 2.802534 | val 3.304301 | tok/s 2634619.55 | norm 0.3712 | dt 199.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21670 | train 2.937855 | val 3.304301 | tok/s 2629910.56 | norm 0.3577 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21671 | train 3.120191 | val 3.304301 | tok/s 2624556.49 | norm 0.3753 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21672 | train 2.876840 | val 3.304301 | tok/s 2613907.18 | norm 0.3807 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21673 | train 3.006692 | val 3.304301 | tok/s 2618936.34 | norm 0.3284 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21674 | train 3.001685 | val 3.304301 | tok/s 2626111.96 | norm 0.3684 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21675 | train 2.838595 | val 3.304301 | tok/s 2626584.81 | norm 0.3895 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21676 | train 2.942817 | val 3.304301 | tok/s 2631649.63 | norm 0.3900 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21677 | train 2.822816 | val 3.304301 | tok/s 2621851.32 | norm 0.3982 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21678 | train 2.951946 | val 3.304301 | tok/s 2626539.89 | norm 0.3848 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21679 | train 2.954067 | val 3.304301 | tok/s 2630646.17 | norm 0.3491 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21680 | train 3.013110 | val 3.304301 | tok/s 2624922.42 | norm 0.4023 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21681 | train 2.860806 | val 3.304301 | tok/s 2629232.92 | norm 0.4186 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21682 | train 3.018599 | val 3.304301 | tok/s 2625830.22 | norm 0.3821 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21683 | train 2.963222 | val 3.304301 | tok/s 2624247.01 | norm 0.3913 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21684 | train 3.014838 | val 3.304301 | tok/s 2631576.58 | norm 0.3537 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21685 | train 3.165988 | val 3.304301 | tok/s 2624377.39 | norm 0.4174 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21686 | train 3.040625 | val 3.304301 | tok/s 2628942.28 | norm 0.4022 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21687 | train 2.827112 | val 3.304301 | tok/s 2627376.00 | norm 0.3510 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21688 | train 2.889971 | val 3.304301 | tok/s 2623313.39 | norm 0.3599 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21689 | train 2.894727 | val 3.304301 | tok/s 2424118.12 | norm 0.3509 | dt 216.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21690 | train 2.756643 | val 3.304301 | tok/s 2616428.92 | norm 0.3355 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21691 | train 2.901736 | val 3.304301 | tok/s 2628409.02 | norm 0.3580 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21692 | train 3.044824 | val 3.304301 | tok/s 2625189.12 | norm 0.4150 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21693 | train 2.973503 | val 3.304301 | tok/s 2628645.08 | norm 0.4013 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21694 | train 2.873325 | val 3.304301 | tok/s 2628119.26 | norm 0.3706 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21695 | train 2.911106 | val 3.304301 | tok/s 2624917.10 | norm 0.3763 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21696 | train 2.996051 | val 3.304301 | tok/s 2631806.70 | norm 0.3607 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21697 | train 3.001026 | val 3.304301 | tok/s 2627806.08 | norm 0.3528 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21698 | train 2.960619 | val 3.304301 | tok/s 2629420.73 | norm 0.3739 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21699 | train 3.025124 | val 3.304301 | tok/s 2615934.03 | norm 0.3630 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21700 | train 2.929862 | val 3.304301 | tok/s 2624724.67 | norm 0.3907 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21701 | train 2.915839 | val 3.304301 | tok/s 2629797.03 | norm 0.3118 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21702 | train 2.839880 | val 3.304301 | tok/s 2629585.59 | norm 0.3697 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21703 | train 2.925105 | val 3.304301 | tok/s 2608372.29 | norm 0.3501 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21704 | train 2.943689 | val 3.304301 | tok/s 2605940.69 | norm 0.3474 | dt 201.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21705 | train 2.826407 | val 3.304301 | tok/s 2621927.30 | norm 0.3562 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21706 | train 2.882499 | val 3.304301 | tok/s 2626580.94 | norm 0.3693 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21707 | train 3.010866 | val 3.304301 | tok/s 2624993.13 | norm 0.3894 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21708 | train 2.937391 | val 3.304301 | tok/s 2627848.10 | norm 0.3319 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21709 | train 2.915583 | val 3.304301 | tok/s 2617077.11 | norm 0.3465 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21710 | train 3.024798 | val 3.304301 | tok/s 2623119.94 | norm 0.3643 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21711 | train 3.021667 | val 3.304301 | tok/s 2629850.38 | norm 0.3484 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21712 | train 3.026925 | val 3.304301 | tok/s 2628453.40 | norm 0.3321 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21713 | train 3.047996 | val 3.304301 | tok/s 2625106.70 | norm 0.3428 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21714 | train 2.998405 | val 3.304301 | tok/s 2617342.69 | norm 0.3738 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21715 | train 2.900111 | val 3.304301 | tok/s 2622048.80 | norm 0.3744 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21716 | train 3.005666 | val 3.304301 | tok/s 2626927.28 | norm 0.3954 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21717 | train 3.156244 | val 3.304301 | tok/s 2624557.96 | norm 0.4674 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21718 | train 2.908428 | val 3.304301 | tok/s 2628167.11 | norm 0.3490 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21719 | train 2.777052 | val 3.304301 | tok/s 2621841.08 | norm 0.3446 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21720 | train 2.953654 | val 3.304301 | tok/s 2624460.43 | norm 0.3784 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21721 | train 2.905961 | val 3.304301 | tok/s 2629544.74 | norm 0.3386 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21722 | train 2.965543 | val 3.304301 | tok/s 2626123.39 | norm 0.3668 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21723 | train 2.923686 | val 3.304301 | tok/s 2629591.60 | norm 0.3947 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21724 | train 2.970391 | val 3.304301 | tok/s 2620987.85 | norm 0.3636 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21725 | train 2.936175 | val 3.304301 | tok/s 2623970.33 | norm 0.3733 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21726 | train 2.869041 | val 3.304301 | tok/s 2629839.00 | norm 0.3850 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21727 | train 2.977767 | val 3.304301 | tok/s 2626162.51 | norm 0.3854 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21728 | train 2.940473 | val 3.304301 | tok/s 2626835.69 | norm 0.3741 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21729 | train 2.889894 | val 3.304301 | tok/s 2625688.68 | norm 0.3613 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21730 | train 2.977061 | val 3.304301 | tok/s 2624638.66 | norm 0.3496 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21731 | train 3.033917 | val 3.304301 | tok/s 2633411.52 | norm 0.3848 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21732 | train 2.813729 | val 3.304301 | tok/s 2630117.98 | norm 0.3249 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21733 | train 2.933893 | val 3.304301 | tok/s 2624447.75 | norm 0.3187 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21734 | train 2.908918 | val 3.304301 | tok/s 2616822.30 | norm 0.3816 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21735 | train 2.963753 | val 3.304301 | tok/s 2622592.45 | norm 0.3764 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21736 | train 2.991601 | val 3.304301 | tok/s 2625612.72 | norm 0.3841 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21737 | train 2.903935 | val 3.304301 | tok/s 2625860.72 | norm 0.3584 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21738 | train 3.115889 | val 3.304301 | tok/s 2625161.76 | norm 0.3648 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21739 | train 2.850309 | val 3.304301 | tok/s 2626373.79 | norm 0.3530 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21740 | train 3.036636 | val 3.304301 | tok/s 2624611.20 | norm 0.3788 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21741 | train 3.038589 | val 3.304301 | tok/s 2626405.59 | norm 0.3898 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21742 | train 3.019785 | val 3.304301 | tok/s 2620996.96 | norm 0.3720 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21743 | train 2.904910 | val 3.304301 | tok/s 2622043.73 | norm 0.3391 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21744 | train 2.974704 | val 3.304301 | tok/s 2619563.40 | norm 0.3634 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21745 | train 2.937106 | val 3.304301 | tok/s 2628063.98 | norm 0.3745 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21746 | train 2.890064 | val 3.304301 | tok/s 2618457.58 | norm 0.3315 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21747 | train 2.987187 | val 3.304301 | tok/s 2621696.39 | norm 0.3470 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21748 | train 3.063018 | val 3.304301 | tok/s 2622462.85 | norm 0.3872 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21749 | train 2.857433 | val 3.304301 | tok/s 2618866.96 | norm 0.3575 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 21750 | val_loss 3.306115 step 21750 | train 3.041424 | val 3.306115 | tok/s 2612486.45 | norm 0.3611 | dt 200.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21751 | train 2.970472 | val 3.306115 | tok/s 2623945.51 | norm 0.3882 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21752 | train 2.907070 | val 3.306115 | tok/s 2625644.21 | norm 0.3896 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21753 | train 2.876228 | val 3.306115 | tok/s 2617038.93 | norm 0.3772 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21754 | train 3.131433 | val 3.306115 | tok/s 2627256.04 | norm 0.4129 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21755 | train 2.961070 | val 3.306115 | tok/s 2625321.80 | norm 0.3708 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21756 | train 2.874326 | val 3.306115 | tok/s 2625981.48 | norm 0.3434 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21757 | train 2.962441 | val 3.306115 | tok/s 2626841.48 | norm 0.3805 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21758 | train 2.902065 | val 3.306115 | tok/s 2622161.13 | norm 0.3614 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21759 | train 2.990510 | val 3.306115 | tok/s 2627863.63 | norm 0.4043 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21760 | train 3.017270 | val 3.306115 | tok/s 2477721.82 | norm 0.3515 | dt 211.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21761 | train 2.846793 | val 3.306115 | tok/s 2627508.39 | norm 0.3513 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21762 | train 2.912541 | val 3.306115 | tok/s 2623920.21 | norm 0.3780 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21763 | train 2.988886 | val 3.306115 | tok/s 2615236.87 | norm 0.3540 | dt 200.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21764 | train 2.843942 | val 3.306115 | tok/s 2623082.46 | norm 0.4077 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21765 | train 2.940778 | val 3.306115 | tok/s 2624381.18 | norm 0.4215 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21766 | train 2.897294 | val 3.306115 | tok/s 2625372.74 | norm 0.3413 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21767 | train 3.020451 | val 3.306115 | tok/s 2607802.29 | norm 0.3161 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21768 | train 3.001675 | val 3.306115 | tok/s 2621325.29 | norm 0.3514 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21769 | train 3.005106 | val 3.306115 | tok/s 2626513.62 | norm 0.3478 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21770 | train 2.931022 | val 3.306115 | tok/s 2626730.39 | norm 0.3888 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21771 | train 2.968393 | val 3.306115 | tok/s 2625787.00 | norm 0.3956 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21772 | train 2.963275 | val 3.306115 | tok/s 2624292.54 | norm 0.3476 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21773 | train 2.989778 | val 3.306115 | tok/s 2619740.73 | norm 0.3350 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21774 | train 2.910517 | val 3.306115 | tok/s 2628032.66 | norm 0.3451 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21775 | train 2.881251 | val 3.306115 | tok/s 2627297.19 | norm 0.3741 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21776 | train 2.920124 | val 3.306115 | tok/s 2627827.04 | norm 0.3661 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21777 | train 2.867764 | val 3.306115 | tok/s 2620004.92 | norm 0.3488 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21778 | train 2.896973 | val 3.306115 | tok/s 2625126.68 | norm 0.3728 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21779 | train 2.979666 | val 3.306115 | tok/s 2624282.24 | norm 0.4206 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21780 | train 3.118075 | val 3.306115 | tok/s 2627342.68 | norm 0.4464 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21781 | train 2.971914 | val 3.306115 | tok/s 2629079.01 | norm 0.3482 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21782 | train 2.940034 | val 3.306115 | tok/s 2626919.12 | norm 0.3428 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21783 | train 2.920850 | val 3.306115 | tok/s 2624005.46 | norm 0.3667 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21784 | train 2.913553 | val 3.306115 | tok/s 2627498.24 | norm 0.3613 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21785 | train 2.848832 | val 3.306115 | tok/s 2622599.92 | norm 0.3862 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21786 | train 2.977342 | val 3.306115 | tok/s 2627984.15 | norm 0.3812 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21787 | train 2.903710 | val 3.306115 | tok/s 2622377.37 | norm 0.3608 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21788 | train 2.892665 | val 3.306115 | tok/s 2623687.37 | norm 0.3843 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21789 | train 2.993013 | val 3.306115 | tok/s 2628194.67 | norm 0.4158 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21790 | train 2.856494 | val 3.306115 | tok/s 2628010.88 | norm 0.3543 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21791 | train 3.033612 | val 3.306115 | tok/s 2627650.70 | norm 0.3644 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21792 | train 2.947885 | val 3.306115 | tok/s 2626206.64 | norm 0.4040 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21793 | train 2.798174 | val 3.306115 | tok/s 2630550.52 | norm 0.4076 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21794 | train 3.162242 | val 3.306115 | tok/s 2627092.52 | norm 0.3975 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21795 | train 2.968527 | val 3.306115 | tok/s 2525151.03 | norm 0.3500 | dt 207.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21796 | train 2.906705 | val 3.306115 | tok/s 2630267.63 | norm 0.3393 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21797 | train 2.998044 | val 3.306115 | tok/s 2622717.22 | norm 0.3578 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21798 | train 2.893935 | val 3.306115 | tok/s 2503804.95 | norm 0.3376 | dt 209.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21799 | train 2.893501 | val 3.306115 | tok/s 2623147.45 | norm 0.3702 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21800 | train 2.940483 | val 3.306115 | tok/s 2475204.09 | norm 0.3924 | dt 211.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21801 | train 2.921371 | val 3.306115 | tok/s 2632287.93 | norm 0.3740 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21802 | train 2.960185 | val 3.306115 | tok/s 2629187.23 | norm 0.3899 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21803 | train 2.904679 | val 3.306115 | tok/s 2626374.35 | norm 0.3829 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21804 | train 2.989706 | val 3.306115 | tok/s 2616460.50 | norm 0.4111 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21805 | train 2.944464 | val 3.306115 | tok/s 2586202.36 | norm 0.4152 | dt 202.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21806 | train 3.072232 | val 3.306115 | tok/s 2626536.23 | norm 0.3824 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21807 | train 2.835481 | val 3.306115 | tok/s 2619662.14 | norm 0.3324 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21808 | train 3.209674 | val 3.306115 | tok/s 2624562.28 | norm 0.4344 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21809 | train 2.862278 | val 3.306115 | tok/s 2616066.57 | norm 0.3871 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21810 | train 2.885802 | val 3.306115 | tok/s 2619194.76 | norm 0.3576 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21811 | train 2.904263 | val 3.306115 | tok/s 2628803.84 | norm 0.3893 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21812 | train 2.901043 | val 3.306115 | tok/s 2625681.96 | norm 0.3798 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21813 | train 2.935991 | val 3.306115 | tok/s 2626489.23 | norm 0.3558 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21814 | train 2.907645 | val 3.306115 | tok/s 2619852.74 | norm 0.3781 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21815 | train 2.856277 | val 3.306115 | tok/s 2625620.85 | norm 0.3962 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21816 | train 3.188093 | val 3.306115 | tok/s 2628069.49 | norm 0.4430 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21817 | train 2.918810 | val 3.306115 | tok/s 2623220.63 | norm 0.3321 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21818 | train 3.020348 | val 3.306115 | tok/s 2625705.64 | norm 0.3500 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21819 | train 3.061667 | val 3.306115 | tok/s 2624698.98 | norm 0.4847 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21820 | train 2.880700 | val 3.306115 | tok/s 2622830.44 | norm 0.3689 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21821 | train 2.881022 | val 3.306115 | tok/s 2629203.37 | norm 0.3207 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21822 | train 3.071256 | val 3.306115 | tok/s 2625489.27 | norm 0.3580 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21823 | train 2.998375 | val 3.306115 | tok/s 2588657.15 | norm 0.3656 | dt 202.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21824 | train 2.929086 | val 3.306115 | tok/s 2620081.38 | norm 0.3623 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21825 | train 3.009860 | val 3.306115 | tok/s 2623330.83 | norm 0.3568 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21826 | train 2.988701 | val 3.306115 | tok/s 2629420.33 | norm 0.3696 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21827 | train 2.912956 | val 3.306115 | tok/s 2625166.92 | norm 0.3869 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21828 | train 2.879889 | val 3.306115 | tok/s 2627901.37 | norm 0.3868 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21829 | train 2.937303 | val 3.306115 | tok/s 2624690.27 | norm 0.4151 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21830 | train 2.961729 | val 3.306115 | tok/s 2624814.87 | norm 0.3903 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21831 | train 2.914085 | val 3.306115 | tok/s 2629570.88 | norm 0.3709 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21832 | train 2.934878 | val 3.306115 | tok/s 2624477.03 | norm 0.3658 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21833 | train 2.952604 | val 3.306115 | tok/s 2593139.62 | norm 0.3256 | dt 202.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21834 | train 3.002000 | val 3.306115 | tok/s 2626591.98 | norm 0.3571 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21835 | train 2.924214 | val 3.306115 | tok/s 2622376.75 | norm 0.3629 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21836 | train 2.897789 | val 3.306115 | tok/s 2624388.24 | norm 0.3238 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21837 | train 2.910756 | val 3.306115 | tok/s 2625349.57 | norm 0.3240 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21838 | train 2.962038 | val 3.306115 | tok/s 2628791.28 | norm 0.3160 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21839 | train 3.135612 | val 3.306115 | tok/s 2625021.95 | norm 0.6568 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21840 | train 2.936095 | val 3.306115 | tok/s 2625648.11 | norm 0.3319 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21841 | train 2.952780 | val 3.306115 | tok/s 2623455.79 | norm 0.3424 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21842 | train 2.946421 | val 3.306115 | tok/s 2618101.72 | norm 0.3498 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21843 | train 2.804840 | val 3.306115 | tok/s 2630715.83 | norm 0.3733 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21844 | train 2.899298 | val 3.306115 | tok/s 2623877.94 | norm 0.4062 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21845 | train 2.957500 | val 3.306115 | tok/s 2626709.72 | norm 0.3740 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21846 | train 2.978627 | val 3.306115 | tok/s 2618843.14 | norm 0.3427 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21847 | train 3.001006 | val 3.306115 | tok/s 2622073.87 | norm 0.3538 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21848 | train 2.866071 | val 3.306115 | tok/s 2629363.93 | norm 0.3517 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21849 | train 2.964116 | val 3.306115 | tok/s 2504530.65 | norm 0.3442 | dt 209.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21850 | train 3.005928 | val 3.306115 | tok/s 2584739.04 | norm 0.3670 | dt 202.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21851 | train 2.865982 | val 3.306115 | tok/s 2621007.15 | norm 0.3831 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21852 | train 2.923876 | val 3.306115 | tok/s 2623772.13 | norm 0.4058 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21853 | train 2.884179 | val 3.306115 | tok/s 2628292.02 | norm 0.3627 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21854 | train 2.999501 | val 3.306115 | tok/s 2621797.43 | norm 0.3244 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21855 | train 2.906562 | val 3.306115 | tok/s 2629044.85 | norm 0.3636 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21856 | train 2.901161 | val 3.306115 | tok/s 2616928.66 | norm 0.3939 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21857 | train 2.910192 | val 3.306115 | tok/s 2620679.49 | norm 0.3767 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21858 | train 2.850426 | val 3.306115 | tok/s 2630571.89 | norm 0.3466 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21859 | train 3.009792 | val 3.306115 | tok/s 2622661.88 | norm 0.3963 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21860 | train 2.893945 | val 3.306115 | tok/s 2626282.34 | norm 0.3938 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21861 | train 2.851469 | val 3.306115 | tok/s 2622142.09 | norm 0.3614 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21862 | train 2.829215 | val 3.306115 | tok/s 2625013.16 | norm 0.3411 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21863 | train 2.933274 | val 3.306115 | tok/s 2627386.97 | norm 0.3579 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21864 | train 3.020308 | val 3.306115 | tok/s 2629543.35 | norm 0.4440 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21865 | train 2.967214 | val 3.306115 | tok/s 2629272.81 | norm 0.3419 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21866 | train 2.870196 | val 3.306115 | tok/s 2622404.82 | norm 0.3275 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21867 | train 2.864297 | val 3.306115 | tok/s 2624795.41 | norm 0.3455 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21868 | train 2.928597 | val 3.306115 | tok/s 2632175.86 | norm 0.3295 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21869 | train 2.979448 | val 3.306115 | tok/s 2622359.74 | norm 0.3519 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21870 | train 2.896761 | val 3.306115 | tok/s 2626814.70 | norm 0.3960 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21871 | train 3.011141 | val 3.306115 | tok/s 2620510.36 | norm 0.3803 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21872 | train 2.954449 | val 3.306115 | tok/s 2626778.41 | norm 0.3850 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21873 | train 2.981105 | val 3.306115 | tok/s 2628074.01 | norm 0.3648 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21874 | train 2.958787 | val 3.306115 | tok/s 2628277.02 | norm 0.3754 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21875 | train 2.937382 | val 3.306115 | tok/s 2630081.70 | norm 0.4297 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21876 | train 2.951273 | val 3.306115 | tok/s 2628066.56 | norm 0.3901 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21877 | train 2.975304 | val 3.306115 | tok/s 2627595.27 | norm 0.3470 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21878 | train 3.060076 | val 3.306115 | tok/s 2628918.08 | norm 0.3928 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21879 | train 3.067794 | val 3.306115 | tok/s 2627078.74 | norm 0.3812 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21880 | train 2.940763 | val 3.306115 | tok/s 2630462.58 | norm 0.3384 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21881 | train 2.849385 | val 3.306115 | tok/s 2621430.73 | norm 0.3283 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21882 | train 2.933145 | val 3.306115 | tok/s 2629208.09 | norm 0.3721 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21883 | train 2.924578 | val 3.306115 | tok/s 2631933.39 | norm 0.3709 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21884 | train 2.869025 | val 3.306115 | tok/s 2628427.65 | norm 0.3708 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21885 | train 3.115714 | val 3.306115 | tok/s 2580790.92 | norm 0.4014 | dt 203.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21886 | train 2.943801 | val 3.306115 | tok/s 2623209.09 | norm 0.3842 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21887 | train 2.820622 | val 3.306115 | tok/s 2624095.14 | norm 0.3346 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21888 | train 2.824859 | val 3.306115 | tok/s 2629495.70 | norm 0.3470 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21889 | train 2.920809 | val 3.306115 | tok/s 2620947.87 | norm 0.3596 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21890 | train 2.994605 | val 3.306115 | tok/s 2631267.53 | norm 0.3624 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21891 | train 3.015778 | val 3.306115 | tok/s 2528728.75 | norm 0.3545 | dt 207.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21892 | train 2.962299 | val 3.306115 | tok/s 2476008.29 | norm 0.3389 | dt 211.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21893 | train 2.935791 | val 3.306115 | tok/s 2617756.54 | norm 0.3677 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21894 | train 2.933818 | val 3.306115 | tok/s 2613247.31 | norm 0.3881 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21895 | train 2.802895 | val 3.306115 | tok/s 2630306.75 | norm 0.4309 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21896 | train 3.073637 | val 3.306115 | tok/s 2626398.64 | norm 0.3556 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21897 | train 2.988325 | val 3.306115 | tok/s 2632136.17 | norm 0.3459 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21898 | train 2.975875 | val 3.306115 | tok/s 2624353.31 | norm 0.3657 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21899 | train 2.987907 | val 3.306115 | tok/s 2621352.88 | norm 0.3803 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21900 | train 2.961902 | val 3.306115 | tok/s 2633576.30 | norm 0.3600 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21901 | train 3.028103 | val 3.306115 | tok/s 2630919.12 | norm 0.3891 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21902 | train 2.889213 | val 3.306115 | tok/s 2628104.69 | norm 0.3770 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21903 | train 3.135814 | val 3.306115 | tok/s 2623497.74 | norm 0.3761 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21904 | train 3.007518 | val 3.306115 | tok/s 2623352.44 | norm 0.3862 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21905 | train 2.835793 | val 3.306115 | tok/s 2631560.71 | norm 0.3738 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21906 | train 3.051436 | val 3.306115 | tok/s 2625752.15 | norm 0.3677 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21907 | train 2.986856 | val 3.306115 | tok/s 2628684.96 | norm 0.3648 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21908 | train 3.150559 | val 3.306115 | tok/s 2623699.73 | norm 0.4285 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21909 | train 2.880397 | val 3.306115 | tok/s 2625664.69 | norm 0.3904 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21910 | train 2.919035 | val 3.306115 | tok/s 2627687.34 | norm 0.3816 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21911 | train 2.941587 | val 3.306115 | tok/s 2590784.02 | norm 0.3932 | dt 202.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21912 | train 3.015442 | val 3.306115 | tok/s 2627329.61 | norm 0.3746 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21913 | train 2.892282 | val 3.306115 | tok/s 2623426.17 | norm 0.3583 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21914 | train 2.926235 | val 3.306115 | tok/s 2625001.88 | norm 0.3506 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21915 | train 2.881833 | val 3.306115 | tok/s 2630024.33 | norm 0.3716 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21916 | train 2.815241 | val 3.306115 | tok/s 2628847.82 | norm 0.3947 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21917 | train 2.920755 | val 3.306115 | tok/s 2631416.29 | norm 0.3566 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21918 | train 2.956599 | val 3.306115 | tok/s 2626057.30 | norm 0.3121 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21919 | train 3.004228 | val 3.306115 | tok/s 2626500.70 | norm 0.3285 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21920 | train 2.889148 | val 3.306115 | tok/s 2631663.80 | norm 0.3499 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21921 | train 3.027103 | val 3.306115 | tok/s 2623387.59 | norm 0.3555 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21922 | train 3.018064 | val 3.306115 | tok/s 2630038.50 | norm 0.3315 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21923 | train 2.954502 | val 3.306115 | tok/s 2627990.95 | norm 0.3629 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21924 | train 3.005454 | val 3.306115 | tok/s 2629797.02 | norm 0.3721 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21925 | train 2.889464 | val 3.306115 | tok/s 2631074.67 | norm 0.3214 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21926 | train 2.918695 | val 3.306115 | tok/s 2538338.66 | norm 0.3385 | dt 206.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21927 | train 2.951718 | val 3.306115 | tok/s 2631943.76 | norm 0.3695 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21928 | train 2.903219 | val 3.306115 | tok/s 2629824.91 | norm 0.3465 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21929 | train 3.125103 | val 3.306115 | tok/s 2631274.26 | norm 0.3592 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21930 | train 2.971686 | val 3.306115 | tok/s 2627005.71 | norm 0.3892 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21931 | train 2.997840 | val 3.306115 | tok/s 2621796.34 | norm 0.3621 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21932 | train 2.979044 | val 3.306115 | tok/s 2630899.27 | norm 0.3525 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21933 | train 2.962507 | val 3.306115 | tok/s 2624022.29 | norm 0.4046 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21934 | train 2.983864 | val 3.306115 | tok/s 2627918.61 | norm 0.3861 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21935 | train 2.866665 | val 3.306115 | tok/s 2629000.99 | norm 0.3416 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21936 | train 2.904644 | val 3.306115 | tok/s 2620369.16 | norm 0.3739 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21937 | train 2.874105 | val 3.306115 | tok/s 2629926.56 | norm 0.3720 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21938 | train 2.858898 | val 3.306115 | tok/s 2621365.53 | norm 0.3727 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21939 | train 2.909476 | val 3.306115 | tok/s 2623348.00 | norm 0.3576 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21940 | train 2.928505 | val 3.306115 | tok/s 2625803.42 | norm 0.3292 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21941 | train 2.930856 | val 3.306115 | tok/s 2622797.18 | norm 0.3498 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21942 | train 2.950061 | val 3.306115 | tok/s 2603844.65 | norm 0.3670 | dt 201.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21943 | train 2.883627 | val 3.306115 | tok/s 2601586.72 | norm 0.3336 | dt 201.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21944 | train 2.936264 | val 3.306115 | tok/s 2627183.62 | norm 0.3314 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21945 | train 2.892471 | val 3.306115 | tok/s 2625358.95 | norm 0.3731 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21946 | train 2.837624 | val 3.306115 | tok/s 2625580.24 | norm 0.3561 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21947 | train 2.954530 | val 3.306115 | tok/s 2630989.51 | norm 0.3730 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21948 | train 2.968263 | val 3.306115 | tok/s 2629930.44 | norm 0.3551 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21949 | train 2.961053 | val 3.306115 | tok/s 2627613.04 | norm 0.3246 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21950 | train 3.043107 | val 3.306115 | tok/s 2624668.51 | norm 0.3457 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21951 | train 2.850243 | val 3.306115 | tok/s 2623589.70 | norm 0.3387 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21952 | train 2.948111 | val 3.306115 | tok/s 2632772.80 | norm 0.3441 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21953 | train 2.980301 | val 3.306115 | tok/s 2629177.95 | norm 0.4452 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21954 | train 3.218174 | val 3.306115 | tok/s 2632884.33 | norm 0.3995 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21955 | train 3.001223 | val 3.306115 | tok/s 2626525.00 | norm 0.3840 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21956 | train 2.935460 | val 3.306115 | tok/s 2625347.71 | norm 0.3443 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21957 | train 2.862040 | val 3.306115 | tok/s 2629128.77 | norm 0.3568 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21958 | train 2.808324 | val 3.306115 | tok/s 2629913.50 | norm 0.3682 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21959 | train 2.877238 | val 3.306115 | tok/s 2629035.73 | norm 0.3768 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21960 | train 2.893707 | val 3.306115 | tok/s 2625623.31 | norm 0.3678 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21961 | train 3.007123 | val 3.306115 | tok/s 2625744.30 | norm 0.3604 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21962 | train 3.099201 | val 3.306115 | tok/s 2634480.10 | norm 0.4082 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21963 | train 2.843543 | val 3.306115 | tok/s 2625617.12 | norm 0.3715 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21964 | train 3.048949 | val 3.306115 | tok/s 2632449.67 | norm 0.3506 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21965 | train 2.914222 | val 3.306115 | tok/s 2626568.69 | norm 0.4386 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21966 | train 2.849981 | val 3.306115 | tok/s 2628680.53 | norm 0.4030 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21967 | train 2.991723 | val 3.306115 | tok/s 2625566.01 | norm 0.3814 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21968 | train 2.960575 | val 3.306115 | tok/s 2621063.12 | norm 0.3965 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21969 | train 2.899461 | val 3.306115 | tok/s 2626678.83 | norm 0.3961 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21970 | train 2.945441 | val 3.306115 | tok/s 2622865.68 | norm 0.3920 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21971 | train 2.973618 | val 3.306115 | tok/s 2626459.19 | norm 0.3551 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21972 | train 2.969643 | val 3.306115 | tok/s 2628840.83 | norm 0.3651 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21973 | train 2.939188 | val 3.306115 | tok/s 2623737.83 | norm 0.3409 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21974 | train 2.865136 | val 3.306115 | tok/s 2632590.05 | norm 0.3408 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21975 | train 3.025246 | val 3.306115 | tok/s 2623503.63 | norm 0.3326 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21976 | train 2.993861 | val 3.306115 | tok/s 2460151.29 | norm 0.3511 | dt 213.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21977 | train 2.885725 | val 3.306115 | tok/s 2617704.42 | norm 0.3570 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21978 | train 2.842069 | val 3.306115 | tok/s 2627088.54 | norm 0.3454 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21979 | train 2.777817 | val 3.306115 | tok/s 2630317.45 | norm 0.3614 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21980 | train 2.941456 | val 3.306115 | tok/s 2626141.73 | norm 0.3791 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21981 | train 2.949293 | val 3.306115 | tok/s 2629603.95 | norm 0.3543 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21982 | train 2.909320 | val 3.306115 | tok/s 2620525.07 | norm 0.3734 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21983 | train 2.833077 | val 3.306115 | tok/s 2546961.73 | norm 0.3844 | dt 205.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21984 | train 2.901902 | val 3.306115 | tok/s 2605066.70 | norm 0.3705 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21985 | train 2.871472 | val 3.306115 | tok/s 2628952.57 | norm 0.3245 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21986 | train 2.941629 | val 3.306115 | tok/s 2626921.10 | norm 0.3314 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21987 | train 2.872800 | val 3.306115 | tok/s 2625999.98 | norm 0.3578 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21988 | train 2.976259 | val 3.306115 | tok/s 2626632.35 | norm 0.3553 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21989 | train 2.875925 | val 3.306115 | tok/s 2628456.49 | norm 0.3366 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21990 | train 2.974347 | val 3.306115 | tok/s 2627093.74 | norm 0.3607 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21991 | train 2.878102 | val 3.306115 | tok/s 2629374.31 | norm 0.3547 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21992 | train 2.963874 | val 3.306115 | tok/s 2621153.60 | norm 0.3471 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21993 | train 2.759908 | val 3.306115 | tok/s 2622598.23 | norm 0.3466 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21994 | train 2.988236 | val 3.306115 | tok/s 2628293.66 | norm 0.3938 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21995 | train 2.927421 | val 3.306115 | tok/s 2630416.24 | norm 0.3646 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21996 | train 2.974291 | val 3.306115 | tok/s 2626061.08 | norm 0.3559 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21997 | train 2.896719 | val 3.306115 | tok/s 2621310.69 | norm 0.3707 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21998 | train 3.032650 | val 3.306115 | tok/s 2625394.07 | norm 0.3772 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 21999 | train 3.203514 | val 3.306115 | tok/s 2629314.47 | norm 0.3907 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 22000 | val_loss 3.306181 step 22000 | train 2.944909 | val 3.306181 | tok/s 2605429.85 | norm 0.3394 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22001 | train 3.122802 | val 3.306181 | tok/s 2626640.99 | norm 0.3622 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22002 | train 3.075388 | val 3.306181 | tok/s 2630421.52 | norm 0.3442 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22003 | train 3.128573 | val 3.306181 | tok/s 2624075.86 | norm 0.3380 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22004 | train 2.943654 | val 3.306181 | tok/s 2625291.43 | norm 0.4007 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22005 | train 2.986778 | val 3.306181 | tok/s 2630421.59 | norm 0.3598 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22006 | train 2.832811 | val 3.306181 | tok/s 2627971.49 | norm 0.3495 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22007 | train 2.886210 | val 3.306181 | tok/s 2628873.10 | norm 0.3753 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22008 | train 2.935744 | val 3.306181 | tok/s 2621698.71 | norm 0.3930 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22009 | train 2.953733 | val 3.306181 | tok/s 2625449.40 | norm 0.3609 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22010 | train 3.175563 | val 3.306181 | tok/s 2632530.01 | norm 0.5484 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22011 | train 2.988636 | val 3.306181 | tok/s 2627613.76 | norm 0.3977 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22012 | train 2.807353 | val 3.306181 | tok/s 2630946.37 | norm 0.2988 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22013 | train 2.960278 | val 3.306181 | tok/s 2625060.89 | norm 0.4000 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22014 | train 2.944949 | val 3.306181 | tok/s 2626083.72 | norm 0.3656 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22015 | train 3.069703 | val 3.306181 | tok/s 2629753.72 | norm 0.3517 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22016 | train 2.924103 | val 3.306181 | tok/s 2625529.41 | norm 0.3366 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22017 | train 2.920185 | val 3.306181 | tok/s 2629714.03 | norm 0.3506 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22018 | train 2.944371 | val 3.306181 | tok/s 2628082.28 | norm 0.3491 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22019 | train 2.998321 | val 3.306181 | tok/s 2626330.44 | norm 0.3587 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22020 | train 2.986097 | val 3.306181 | tok/s 2625747.51 | norm 0.3663 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22021 | train 3.028666 | val 3.306181 | tok/s 2627872.95 | norm 0.3645 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22022 | train 2.995566 | val 3.306181 | tok/s 2627387.46 | norm 0.3696 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22023 | train 3.064853 | val 3.306181 | tok/s 2624173.52 | norm 0.3824 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22024 | train 2.935040 | val 3.306181 | tok/s 2626930.78 | norm 0.3910 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22025 | train 2.941341 | val 3.306181 | tok/s 2626324.50 | norm 0.3587 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22026 | train 3.045985 | val 3.306181 | tok/s 2427688.07 | norm 0.3274 | dt 215.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22027 | train 2.912329 | val 3.306181 | tok/s 2622968.59 | norm 0.3470 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22028 | train 2.937123 | val 3.306181 | tok/s 2628582.82 | norm 0.3477 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22029 | train 2.904658 | val 3.306181 | tok/s 2628169.15 | norm 0.3558 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22030 | train 2.863309 | val 3.306181 | tok/s 2628413.38 | norm 0.3534 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22031 | train 2.923635 | val 3.306181 | tok/s 2625057.63 | norm 0.3635 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22032 | train 2.920912 | val 3.306181 | tok/s 2629694.63 | norm 0.3772 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22033 | train 2.938456 | val 3.306181 | tok/s 2630106.32 | norm 0.3683 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22034 | train 2.931008 | val 3.306181 | tok/s 2525918.32 | norm 0.3768 | dt 207.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22035 | train 2.836531 | val 3.306181 | tok/s 2623183.25 | norm 0.3907 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22036 | train 2.874933 | val 3.306181 | tok/s 2621465.91 | norm 0.3526 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22037 | train 2.948394 | val 3.306181 | tok/s 2627834.67 | norm 0.3367 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22038 | train 2.922805 | val 3.306181 | tok/s 2622434.89 | norm 0.3842 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22039 | train 2.894050 | val 3.306181 | tok/s 2626599.46 | norm 0.3623 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22040 | train 2.879565 | val 3.306181 | tok/s 2615060.60 | norm 0.3490 | dt 200.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22041 | train 2.918839 | val 3.306181 | tok/s 2607968.10 | norm 0.3458 | dt 201.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22042 | train 2.939885 | val 3.306181 | tok/s 2631913.24 | norm 0.3427 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22043 | train 2.905810 | val 3.306181 | tok/s 2625648.47 | norm 0.3603 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22044 | train 2.831176 | val 3.306181 | tok/s 2629276.99 | norm 0.3742 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22045 | train 2.897350 | val 3.306181 | tok/s 2626922.33 | norm 0.3459 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22046 | train 2.959237 | val 3.306181 | tok/s 2628821.20 | norm 0.3324 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22047 | train 2.917218 | val 3.306181 | tok/s 2632453.36 | norm 0.3402 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22048 | train 2.815534 | val 3.306181 | tok/s 2628407.58 | norm 0.3454 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22049 | train 2.848588 | val 3.306181 | tok/s 2628468.71 | norm 0.3589 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22050 | train 2.888936 | val 3.306181 | tok/s 2626244.64 | norm 0.3446 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22051 | train 2.943014 | val 3.306181 | tok/s 2627734.91 | norm 0.3505 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22052 | train 2.951405 | val 3.306181 | tok/s 2631822.70 | norm 0.3052 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22053 | train 2.904321 | val 3.306181 | tok/s 2628395.02 | norm 0.3397 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22054 | train 3.135877 | val 3.306181 | tok/s 2634070.37 | norm 0.3688 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22055 | train 2.859930 | val 3.306181 | tok/s 2629536.65 | norm 0.3688 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22056 | train 2.829426 | val 3.306181 | tok/s 2629755.37 | norm 0.3794 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22057 | train 2.826099 | val 3.306181 | tok/s 2631500.67 | norm 0.3671 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22058 | train 2.888706 | val 3.306181 | tok/s 2625578.49 | norm 0.3306 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22059 | train 3.038099 | val 3.306181 | tok/s 2630961.48 | norm 0.3184 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22060 | train 2.888284 | val 3.306181 | tok/s 2626251.13 | norm 0.3006 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22061 | train 2.913856 | val 3.306181 | tok/s 2627859.29 | norm 0.3244 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22062 | train 2.983481 | val 3.306181 | tok/s 2626556.15 | norm 0.3343 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22063 | train 3.039423 | val 3.306181 | tok/s 2625158.69 | norm 0.3484 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22064 | train 2.926491 | val 3.306181 | tok/s 2626507.35 | norm 0.3309 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22065 | train 3.018389 | val 3.306181 | tok/s 2626335.46 | norm 0.3651 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22066 | train 2.975813 | val 3.306181 | tok/s 2616105.83 | norm 0.3476 | dt 200.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22067 | train 2.970043 | val 3.306181 | tok/s 2630787.60 | norm 0.3297 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22068 | train 3.020648 | val 3.306181 | tok/s 2627716.88 | norm 0.3950 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22069 | train 2.894234 | val 3.306181 | tok/s 2630105.86 | norm 0.3760 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22070 | train 2.923889 | val 3.306181 | tok/s 2627538.94 | norm 0.3478 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22071 | train 2.934972 | val 3.306181 | tok/s 2602595.35 | norm 0.3718 | dt 201.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22072 | train 2.993625 | val 3.306181 | tok/s 2631034.42 | norm 0.3739 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22073 | train 2.914493 | val 3.306181 | tok/s 2597974.66 | norm 0.3714 | dt 201.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22074 | train 2.969029 | val 3.306181 | tok/s 2617163.66 | norm 0.3559 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22075 | train 2.953064 | val 3.306181 | tok/s 2628997.59 | norm 0.3486 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22076 | train 2.924748 | val 3.306181 | tok/s 2628357.42 | norm 0.3705 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22077 | train 2.855471 | val 3.306181 | tok/s 2626258.58 | norm 0.3292 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22078 | train 2.872216 | val 3.306181 | tok/s 2624865.71 | norm 0.3197 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22079 | train 2.964466 | val 3.306181 | tok/s 2627709.15 | norm 0.3858 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22080 | train 2.940054 | val 3.306181 | tok/s 2620892.08 | norm 0.4143 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22081 | train 2.916816 | val 3.306181 | tok/s 2627318.89 | norm 0.3338 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22082 | train 2.911337 | val 3.306181 | tok/s 2623754.89 | norm 0.3258 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22083 | train 2.958960 | val 3.306181 | tok/s 2618203.93 | norm 0.3852 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22084 | train 2.844273 | val 3.306181 | tok/s 2625463.25 | norm 0.4101 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22085 | train 2.845502 | val 3.306181 | tok/s 2615083.00 | norm 0.3613 | dt 200.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22086 | train 3.011505 | val 3.306181 | tok/s 2627255.05 | norm 0.3588 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22087 | train 2.959852 | val 3.306181 | tok/s 2610149.95 | norm 0.3668 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22088 | train 2.985646 | val 3.306181 | tok/s 2626074.05 | norm 0.3262 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22089 | train 2.900578 | val 3.306181 | tok/s 2628899.32 | norm 0.3545 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22090 | train 2.946658 | val 3.306181 | tok/s 2625089.56 | norm 0.3912 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22091 | train 2.968288 | val 3.306181 | tok/s 2626509.27 | norm 0.3648 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22092 | train 2.961350 | val 3.306181 | tok/s 2625696.58 | norm 0.3469 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22093 | train 2.963863 | val 3.306181 | tok/s 2627354.04 | norm 0.3626 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22094 | train 2.973188 | val 3.306181 | tok/s 2630140.61 | norm 0.3710 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22095 | train 2.939933 | val 3.306181 | tok/s 2627971.63 | norm 0.3939 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22096 | train 2.864076 | val 3.306181 | tok/s 2627290.24 | norm 0.4097 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22097 | train 2.900398 | val 3.306181 | tok/s 2622621.42 | norm 0.4082 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22098 | train 2.807663 | val 3.306181 | tok/s 2623092.67 | norm 0.3824 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22099 | train 2.848333 | val 3.306181 | tok/s 2625834.91 | norm 0.3676 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22100 | train 2.886360 | val 3.306181 | tok/s 2627937.76 | norm 0.3670 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22101 | train 2.911699 | val 3.306181 | tok/s 2631421.10 | norm 0.3110 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22102 | train 2.937699 | val 3.306181 | tok/s 2623219.83 | norm 0.3642 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22103 | train 2.869253 | val 3.306181 | tok/s 2627532.99 | norm 0.3986 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22104 | train 2.955603 | val 3.306181 | tok/s 2630005.23 | norm 0.3589 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22105 | train 3.034309 | val 3.306181 | tok/s 2626233.28 | norm 0.3959 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22106 | train 2.959195 | val 3.306181 | tok/s 2626556.43 | norm 0.3813 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22107 | train 2.911515 | val 3.306181 | tok/s 2619456.38 | norm 0.3664 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22108 | train 3.066223 | val 3.306181 | tok/s 2620989.01 | norm 0.3845 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22109 | train 2.901783 | val 3.306181 | tok/s 2631791.61 | norm 0.3700 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22110 | train 2.839057 | val 3.306181 | tok/s 2622787.57 | norm 0.3819 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22111 | train 2.964646 | val 3.306181 | tok/s 2626769.83 | norm 0.3660 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22112 | train 2.946410 | val 3.306181 | tok/s 2626428.42 | norm 0.3502 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22113 | train 2.948061 | val 3.306181 | tok/s 2616578.96 | norm 0.3392 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22114 | train 2.889412 | val 3.306181 | tok/s 2627728.27 | norm 0.3339 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22115 | train 2.940804 | val 3.306181 | tok/s 2618549.51 | norm 0.3824 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22116 | train 2.955888 | val 3.306181 | tok/s 2625577.08 | norm 0.3712 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22117 | train 2.975520 | val 3.306181 | tok/s 2542767.63 | norm 0.4046 | dt 206.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22118 | train 2.999058 | val 3.306181 | tok/s 2515836.30 | norm 0.4025 | dt 208.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22119 | train 2.955031 | val 3.306181 | tok/s 2613127.88 | norm 0.3805 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22120 | train 3.053972 | val 3.306181 | tok/s 2623034.03 | norm 0.3841 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22121 | train 2.867897 | val 3.306181 | tok/s 2628211.69 | norm 0.3632 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22122 | train 2.939011 | val 3.306181 | tok/s 2623224.30 | norm 0.3156 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22123 | train 2.901457 | val 3.306181 | tok/s 2627200.18 | norm 0.3049 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22124 | train 2.802432 | val 3.306181 | tok/s 2617066.21 | norm 0.3339 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22125 | train 2.957812 | val 3.306181 | tok/s 2622442.49 | norm 0.3833 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22126 | train 2.855922 | val 3.306181 | tok/s 2628675.85 | norm 0.3904 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22127 | train 2.863849 | val 3.306181 | tok/s 2620743.49 | norm 0.3662 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22128 | train 2.960998 | val 3.306181 | tok/s 2627374.19 | norm 0.3744 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22129 | train 2.956489 | val 3.306181 | tok/s 2613814.90 | norm 0.4114 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22130 | train 3.032435 | val 3.306181 | tok/s 2627557.01 | norm 0.3834 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22131 | train 2.929995 | val 3.306181 | tok/s 2631925.57 | norm 0.3488 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22132 | train 2.962827 | val 3.306181 | tok/s 2630120.77 | norm 0.4551 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22133 | train 2.920294 | val 3.306181 | tok/s 2627178.10 | norm 0.3842 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22134 | train 2.965356 | val 3.306181 | tok/s 2622196.77 | norm 0.3256 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22135 | train 2.941800 | val 3.306181 | tok/s 2616706.69 | norm 0.3541 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22136 | train 2.955365 | val 3.306181 | tok/s 2609423.53 | norm 0.3735 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22137 | train 3.067478 | val 3.306181 | tok/s 2624989.37 | norm 0.3642 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22138 | train 2.939328 | val 3.306181 | tok/s 2627935.52 | norm 0.3411 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22139 | train 2.938413 | val 3.306181 | tok/s 2622175.93 | norm 0.3222 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22140 | train 2.879632 | val 3.306181 | tok/s 2622698.69 | norm 0.3554 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22141 | train 2.787293 | val 3.306181 | tok/s 2629902.88 | norm 0.3479 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22142 | train 2.931781 | val 3.306181 | tok/s 2582327.39 | norm 0.3670 | dt 203.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22143 | train 2.918022 | val 3.306181 | tok/s 2631166.48 | norm 0.3924 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22144 | train 3.016657 | val 3.306181 | tok/s 2621700.99 | norm 0.4003 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22145 | train 2.969908 | val 3.306181 | tok/s 2624668.33 | norm 0.3806 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22146 | train 2.896119 | val 3.306181 | tok/s 2631816.46 | norm 0.3629 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22147 | train 2.859211 | val 3.306181 | tok/s 2627556.66 | norm 0.3959 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22148 | train 2.991692 | val 3.306181 | tok/s 2628360.60 | norm 0.4047 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22149 | train 2.968711 | val 3.306181 | tok/s 2625041.31 | norm 0.3951 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22150 | train 2.883942 | val 3.306181 | tok/s 2625579.34 | norm 0.4058 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22151 | train 3.110020 | val 3.306181 | tok/s 2626564.14 | norm 0.4099 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22152 | train 2.911990 | val 3.306181 | tok/s 2625683.10 | norm 0.3471 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22153 | train 2.911176 | val 3.306181 | tok/s 2628179.11 | norm 0.3382 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22154 | train 2.859799 | val 3.306181 | tok/s 2622218.12 | norm 0.3888 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22155 | train 2.993487 | val 3.306181 | tok/s 2625620.68 | norm 0.3899 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22156 | train 2.838879 | val 3.306181 | tok/s 2629059.20 | norm 0.3570 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22157 | train 2.888655 | val 3.306181 | tok/s 2628937.89 | norm 0.3324 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22158 | train 2.829881 | val 3.306181 | tok/s 2630420.73 | norm 0.3607 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22159 | train 2.964081 | val 3.306181 | tok/s 2626472.10 | norm 0.3803 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22160 | train 3.109330 | val 3.306181 | tok/s 2623674.11 | norm 0.3669 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22161 | train 2.989167 | val 3.306181 | tok/s 2625318.25 | norm 0.3524 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22162 | train 2.920043 | val 3.306181 | tok/s 2620064.17 | norm 0.3712 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22163 | train 3.027687 | val 3.306181 | tok/s 2633791.84 | norm 0.3721 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22164 | train 2.901020 | val 3.306181 | tok/s 2634265.26 | norm 0.3930 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22165 | train 2.860691 | val 3.306181 | tok/s 2629316.42 | norm 0.3766 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22166 | train 3.047360 | val 3.306181 | tok/s 2543764.50 | norm 0.3867 | dt 206.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22167 | train 2.997395 | val 3.306181 | tok/s 2616543.44 | norm 0.3917 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22168 | train 3.000241 | val 3.306181 | tok/s 2628520.99 | norm 0.3855 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22169 | train 3.036955 | val 3.306181 | tok/s 2628545.77 | norm 0.4432 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22170 | train 2.993395 | val 3.306181 | tok/s 2628822.35 | norm 0.3890 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22171 | train 2.951921 | val 3.306181 | tok/s 2621973.22 | norm 0.3610 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22172 | train 3.001087 | val 3.306181 | tok/s 2623015.29 | norm 0.3927 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22173 | train 2.948692 | val 3.306181 | tok/s 2630721.40 | norm 0.3535 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22174 | train 2.899687 | val 3.306181 | tok/s 2625640.87 | norm 0.3877 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22175 | train 2.893743 | val 3.306181 | tok/s 2622684.27 | norm 0.3895 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22176 | train 3.155900 | val 3.306181 | tok/s 2616341.10 | norm 0.3988 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22177 | train 3.054947 | val 3.306181 | tok/s 2624145.57 | norm 0.3704 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22178 | train 2.820277 | val 3.306181 | tok/s 2629611.86 | norm 0.3694 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22179 | train 2.972891 | val 3.306181 | tok/s 2627508.22 | norm 0.3912 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22180 | train 2.983518 | val 3.306181 | tok/s 2627719.24 | norm 0.3500 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22181 | train 2.772315 | val 3.306181 | tok/s 2624039.77 | norm 0.3589 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22182 | train 2.934885 | val 3.306181 | tok/s 2622972.87 | norm 0.3771 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22183 | train 2.947853 | val 3.306181 | tok/s 2630682.18 | norm 0.3990 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22184 | train 2.850537 | val 3.306181 | tok/s 2628382.54 | norm 0.4181 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22185 | train 2.784252 | val 3.306181 | tok/s 2630214.50 | norm 0.4070 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22186 | train 3.031806 | val 3.306181 | tok/s 2622062.26 | norm 0.4726 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22187 | train 2.996649 | val 3.306181 | tok/s 2593628.18 | norm 0.3991 | dt 202.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22188 | train 2.953665 | val 3.306181 | tok/s 2626338.59 | norm 0.3026 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22189 | train 2.913464 | val 3.306181 | tok/s 2625005.82 | norm 0.3351 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22190 | train 2.881486 | val 3.306181 | tok/s 2610127.56 | norm 0.3620 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22191 | train 2.971102 | val 3.306181 | tok/s 2623612.09 | norm 0.3740 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22192 | train 3.013233 | val 3.306181 | tok/s 2624125.83 | norm 0.3602 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22193 | train 2.835106 | val 3.306181 | tok/s 2625952.78 | norm 0.3527 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22194 | train 2.833654 | val 3.306181 | tok/s 2548296.43 | norm 0.3871 | dt 205.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22195 | train 2.878111 | val 3.306181 | tok/s 2631450.41 | norm 0.3791 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22196 | train 2.886233 | val 3.306181 | tok/s 2628646.35 | norm 0.3850 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22197 | train 2.934432 | val 3.306181 | tok/s 2627086.86 | norm 0.4206 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22198 | train 2.888165 | val 3.306181 | tok/s 2620355.41 | norm 0.3885 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22199 | train 2.827870 | val 3.306181 | tok/s 2520539.56 | norm 0.3622 | dt 208.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22200 | train 2.941745 | val 3.306181 | tok/s 2611300.10 | norm 0.3746 | dt 200.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22201 | train 2.779017 | val 3.306181 | tok/s 2480254.88 | norm 0.3562 | dt 211.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22202 | train 3.013038 | val 3.306181 | tok/s 2628290.27 | norm 0.3541 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22203 | train 2.781455 | val 3.306181 | tok/s 2435630.92 | norm 0.3529 | dt 215.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22204 | train 3.027894 | val 3.306181 | tok/s 2625017.96 | norm 0.3713 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22205 | train 3.088665 | val 3.306181 | tok/s 2602911.14 | norm 0.3983 | dt 201.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22206 | train 2.870706 | val 3.306181 | tok/s 2626594.19 | norm 0.3490 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22207 | train 2.907930 | val 3.306181 | tok/s 2624391.40 | norm 0.3299 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22208 | train 3.000205 | val 3.306181 | tok/s 2624849.76 | norm 0.4006 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22209 | train 2.893950 | val 3.306181 | tok/s 2624112.03 | norm 0.3524 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22210 | train 2.948460 | val 3.306181 | tok/s 2629683.29 | norm 0.3190 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22211 | train 3.041465 | val 3.306181 | tok/s 2627783.94 | norm 0.3322 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22212 | train 2.841192 | val 3.306181 | tok/s 2631198.89 | norm 0.3181 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22213 | train 3.056261 | val 3.306181 | tok/s 2624552.36 | norm 0.3710 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22214 | train 2.961416 | val 3.306181 | tok/s 2626485.28 | norm 0.3576 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22215 | train 2.886608 | val 3.306181 | tok/s 2632071.93 | norm 0.3443 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22216 | train 2.988279 | val 3.306181 | tok/s 2628180.27 | norm 0.4093 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22217 | train 2.954766 | val 3.306181 | tok/s 2624989.34 | norm 0.3914 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22218 | train 3.006097 | val 3.306181 | tok/s 2626772.57 | norm 0.3556 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22219 | train 2.930991 | val 3.306181 | tok/s 2627153.60 | norm 0.3759 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22220 | train 2.943674 | val 3.306181 | tok/s 2628493.97 | norm 0.3573 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22221 | train 2.918033 | val 3.306181 | tok/s 2626302.15 | norm 0.3538 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22222 | train 3.002003 | val 3.306181 | tok/s 2629743.90 | norm 0.3649 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22223 | train 2.970575 | val 3.306181 | tok/s 2626235.90 | norm 0.3734 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22224 | train 2.961014 | val 3.306181 | tok/s 2630494.99 | norm 0.3630 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22225 | train 2.924137 | val 3.306181 | tok/s 2630175.79 | norm 0.3673 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22226 | train 3.008239 | val 3.306181 | tok/s 2622521.75 | norm 0.3833 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22227 | train 2.860949 | val 3.306181 | tok/s 2626483.99 | norm 0.3839 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22228 | train 2.860122 | val 3.306181 | tok/s 2626941.62 | norm 0.3891 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22229 | train 2.877174 | val 3.306181 | tok/s 2626696.41 | norm 0.3882 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22230 | train 2.879863 | val 3.306181 | tok/s 2599704.49 | norm 0.4004 | dt 201.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22231 | train 2.965549 | val 3.306181 | tok/s 2627693.49 | norm 0.3827 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22232 | train 2.910991 | val 3.306181 | tok/s 2630359.00 | norm 0.3680 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22233 | train 2.890068 | val 3.306181 | tok/s 2630743.97 | norm 0.3749 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22234 | train 2.824154 | val 3.306181 | tok/s 2627030.50 | norm 0.3527 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22235 | train 2.801701 | val 3.306181 | tok/s 2621805.18 | norm 0.3402 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22236 | train 3.157592 | val 3.306181 | tok/s 2620280.00 | norm 0.3889 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22237 | train 2.893825 | val 3.306181 | tok/s 2630099.64 | norm 0.3944 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22238 | train 3.027509 | val 3.306181 | tok/s 2628684.35 | norm 0.3743 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22239 | train 2.862022 | val 3.306181 | tok/s 2629558.86 | norm 0.4063 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22240 | train 2.910739 | val 3.306181 | tok/s 2622354.98 | norm 0.3560 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22241 | train 2.950674 | val 3.306181 | tok/s 2626115.94 | norm 0.3338 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22242 | train 2.825342 | val 3.306181 | tok/s 2628457.63 | norm 0.3394 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22243 | train 3.104218 | val 3.306181 | tok/s 2625795.74 | norm 0.3722 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22244 | train 2.941964 | val 3.306181 | tok/s 2629491.62 | norm 0.4038 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22245 | train 2.966866 | val 3.306181 | tok/s 2622332.62 | norm 0.3891 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22246 | train 3.010317 | val 3.306181 | tok/s 2625169.51 | norm 0.4007 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22247 | train 3.012861 | val 3.306181 | tok/s 2628042.37 | norm 0.3635 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22248 | train 2.954754 | val 3.306181 | tok/s 2625695.75 | norm 0.3357 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22249 | train 2.935641 | val 3.306181 | tok/s 2630359.46 | norm 0.3579 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 22250 | val_loss 3.305745 step 22250 | train 2.947101 | val 3.305745 | tok/s 2624196.99 | norm 0.4110 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22251 | train 2.956062 | val 3.305745 | tok/s 2620885.96 | norm 0.3929 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22252 | train 2.860859 | val 3.305745 | tok/s 2621894.60 | norm 0.3960 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22253 | train 2.944481 | val 3.305745 | tok/s 2630430.93 | norm 0.4098 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22254 | train 2.852985 | val 3.305745 | tok/s 2628484.42 | norm 0.3532 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22255 | train 3.020322 | val 3.305745 | tok/s 2628717.21 | norm 0.3586 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22256 | train 3.148976 | val 3.305745 | tok/s 2620940.16 | norm 0.4018 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22257 | train 2.965796 | val 3.305745 | tok/s 2624592.50 | norm 0.3632 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22258 | train 3.007353 | val 3.305745 | tok/s 2628877.25 | norm 0.3494 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22259 | train 2.922445 | val 3.305745 | tok/s 2625695.81 | norm 0.3753 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22260 | train 3.007480 | val 3.305745 | tok/s 2630317.68 | norm 0.3512 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22261 | train 2.936499 | val 3.305745 | tok/s 2613631.76 | norm 0.3647 | dt 200.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22262 | train 2.921335 | val 3.305745 | tok/s 2619446.99 | norm 0.3398 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22263 | train 3.020308 | val 3.305745 | tok/s 2620351.60 | norm 0.3352 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22264 | train 2.769688 | val 3.305745 | tok/s 2618519.29 | norm 0.3192 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22265 | train 2.968764 | val 3.305745 | tok/s 2630121.22 | norm 0.3318 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22266 | train 2.946718 | val 3.305745 | tok/s 2626397.88 | norm 0.4688 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22267 | train 2.899853 | val 3.305745 | tok/s 2625503.75 | norm 0.3233 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22268 | train 2.938705 | val 3.305745 | tok/s 2617515.45 | norm 0.3542 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22269 | train 2.974026 | val 3.305745 | tok/s 2620495.68 | norm 0.3492 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22270 | train 3.025818 | val 3.305745 | tok/s 2625912.26 | norm 0.3769 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22271 | train 2.912309 | val 3.305745 | tok/s 2625537.28 | norm 0.3730 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22272 | train 2.976776 | val 3.305745 | tok/s 2625983.06 | norm 0.3211 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22273 | train 2.981852 | val 3.305745 | tok/s 2630683.70 | norm 0.3502 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22274 | train 2.870384 | val 3.305745 | tok/s 2631707.71 | norm 0.3463 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22275 | train 3.019002 | val 3.305745 | tok/s 2631763.14 | norm 0.3629 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22276 | train 2.945039 | val 3.305745 | tok/s 2627517.93 | norm 0.3733 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22277 | train 2.864582 | val 3.305745 | tok/s 2629362.27 | norm 0.3915 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22278 | train 2.973604 | val 3.305745 | tok/s 2628903.77 | norm 0.3783 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22279 | train 2.945171 | val 3.305745 | tok/s 2628448.52 | norm 0.3440 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22280 | train 3.132166 | val 3.305745 | tok/s 2632087.12 | norm 0.5296 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22281 | train 3.175606 | val 3.305745 | tok/s 2628779.93 | norm 0.3711 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22282 | train 2.915686 | val 3.305745 | tok/s 2629771.99 | norm 0.3558 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22283 | train 2.840473 | val 3.305745 | tok/s 2623048.69 | norm 0.3340 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22284 | train 2.877707 | val 3.305745 | tok/s 2620789.11 | norm 0.3596 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22285 | train 2.913915 | val 3.305745 | tok/s 2630179.68 | norm 0.3602 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22286 | train 3.027296 | val 3.305745 | tok/s 2628669.97 | norm 0.3658 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22287 | train 2.903113 | val 3.305745 | tok/s 2627797.76 | norm 0.3709 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22288 | train 2.845212 | val 3.305745 | tok/s 2612824.82 | norm 0.3543 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22289 | train 2.946329 | val 3.305745 | tok/s 2624678.54 | norm 0.3442 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22290 | train 3.003856 | val 3.305745 | tok/s 2630988.55 | norm 0.3389 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22291 | train 2.993316 | val 3.305745 | tok/s 2630009.72 | norm 0.3568 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22292 | train 2.889665 | val 3.305745 | tok/s 2629487.68 | norm 0.3393 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22293 | train 2.925510 | val 3.305745 | tok/s 2621062.87 | norm 0.3661 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22294 | train 2.983028 | val 3.305745 | tok/s 2557598.92 | norm 0.3760 | dt 204.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22295 | train 2.839428 | val 3.305745 | tok/s 2461956.11 | norm 0.3877 | dt 212.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22296 | train 2.918587 | val 3.305745 | tok/s 2628485.80 | norm 0.4073 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22297 | train 2.987283 | val 3.305745 | tok/s 2631898.92 | norm 0.3752 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22298 | train 2.896329 | val 3.305745 | tok/s 2623527.49 | norm 0.3533 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22299 | train 3.058881 | val 3.305745 | tok/s 2624923.06 | norm 0.5837 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22300 | train 2.906701 | val 3.305745 | tok/s 2625743.80 | norm 0.3490 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22301 | train 2.921253 | val 3.305745 | tok/s 2626603.34 | norm 0.3164 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22302 | train 2.819921 | val 3.305745 | tok/s 2631731.36 | norm 0.3445 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22303 | train 2.926367 | val 3.305745 | tok/s 2615673.28 | norm 0.3393 | dt 200.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22304 | train 2.904790 | val 3.305745 | tok/s 2624424.77 | norm 0.3665 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22305 | train 2.950257 | val 3.305745 | tok/s 2629803.85 | norm 0.3680 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22306 | train 2.981184 | val 3.305745 | tok/s 2620829.12 | norm 0.3693 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22307 | train 3.021722 | val 3.305745 | tok/s 2627469.77 | norm 0.3804 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22308 | train 2.861104 | val 3.305745 | tok/s 2627159.13 | norm 0.3765 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22309 | train 2.989451 | val 3.305745 | tok/s 2623529.21 | norm 0.3571 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22310 | train 3.086101 | val 3.305745 | tok/s 2628027.80 | norm 0.3544 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22311 | train 2.929780 | val 3.305745 | tok/s 2629403.52 | norm 0.3578 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22312 | train 2.789420 | val 3.305745 | tok/s 2628448.81 | norm 0.3470 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22313 | train 2.894981 | val 3.305745 | tok/s 2621016.51 | norm 0.3413 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22314 | train 2.916109 | val 3.305745 | tok/s 2625367.65 | norm 0.3145 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22315 | train 2.920437 | val 3.305745 | tok/s 2630949.73 | norm 0.3655 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22316 | train 2.888873 | val 3.305745 | tok/s 2626911.65 | norm 0.3775 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22317 | train 2.924596 | val 3.305745 | tok/s 2630446.47 | norm 0.3933 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22318 | train 2.919519 | val 3.305745 | tok/s 2623487.40 | norm 0.3721 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22319 | train 2.965605 | val 3.305745 | tok/s 2626434.65 | norm 0.3942 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22320 | train 2.864491 | val 3.305745 | tok/s 2631900.63 | norm 0.3637 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22321 | train 2.929374 | val 3.305745 | tok/s 2620178.19 | norm 0.3437 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22322 | train 2.958048 | val 3.305745 | tok/s 2632384.59 | norm 0.3658 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22323 | train 2.856132 | val 3.305745 | tok/s 2624015.93 | norm 0.3610 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22324 | train 3.032401 | val 3.305745 | tok/s 2627816.49 | norm 0.3523 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22325 | train 2.985956 | val 3.305745 | tok/s 2625936.88 | norm 0.3590 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22326 | train 2.955272 | val 3.305745 | tok/s 2621990.87 | norm 0.3664 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22327 | train 2.889886 | val 3.305745 | tok/s 2628994.61 | norm 0.3711 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22328 | train 2.955961 | val 3.305745 | tok/s 2625002.43 | norm 0.3513 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22329 | train 2.909233 | val 3.305745 | tok/s 2627086.31 | norm 0.3714 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22330 | train 2.880662 | val 3.305745 | tok/s 2618896.19 | norm 0.3858 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22331 | train 3.014784 | val 3.305745 | tok/s 2622282.79 | norm 0.3747 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22332 | train 2.907697 | val 3.305745 | tok/s 2631814.07 | norm 0.3828 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22333 | train 2.909426 | val 3.305745 | tok/s 2625129.45 | norm 0.3482 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22334 | train 2.979352 | val 3.305745 | tok/s 2506666.78 | norm 0.3205 | dt 209.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22335 | train 2.889155 | val 3.305745 | tok/s 2608360.34 | norm 0.3514 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22336 | train 2.934138 | val 3.305745 | tok/s 2618008.93 | norm 0.3569 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22337 | train 2.934055 | val 3.305745 | tok/s 2630556.14 | norm 0.3679 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22338 | train 3.007774 | val 3.305745 | tok/s 2625921.78 | norm 0.3706 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22339 | train 2.866487 | val 3.305745 | tok/s 2625796.50 | norm 0.3714 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22340 | train 2.969307 | val 3.305745 | tok/s 2624243.15 | norm 0.3793 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22341 | train 3.016412 | val 3.305745 | tok/s 2625742.04 | norm 0.3880 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22342 | train 3.025655 | val 3.305745 | tok/s 2629230.46 | norm 0.3592 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22343 | train 2.973451 | val 3.305745 | tok/s 2625789.42 | norm 0.3543 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22344 | train 3.000105 | val 3.305745 | tok/s 2626958.11 | norm 0.3829 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22345 | train 2.939703 | val 3.305745 | tok/s 2628955.38 | norm 0.3731 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22346 | train 2.858680 | val 3.305745 | tok/s 2627377.62 | norm 0.3364 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22347 | train 2.901872 | val 3.305745 | tok/s 2631743.38 | norm 0.3150 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22348 | train 2.907748 | val 3.305745 | tok/s 2628047.00 | norm 0.3464 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22349 | train 2.927153 | val 3.305745 | tok/s 2630049.48 | norm 0.3666 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22350 | train 2.916775 | val 3.305745 | tok/s 2626113.96 | norm 0.3524 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22351 | train 2.936856 | val 3.305745 | tok/s 2625842.33 | norm 0.3536 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22352 | train 2.973523 | val 3.305745 | tok/s 2610393.26 | norm 0.3590 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22353 | train 2.912696 | val 3.305745 | tok/s 2628779.84 | norm 0.3670 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22354 | train 2.926095 | val 3.305745 | tok/s 2628174.82 | norm 0.3656 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22355 | train 3.087361 | val 3.305745 | tok/s 2625371.77 | norm 0.3734 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22356 | train 2.949800 | val 3.305745 | tok/s 2630765.81 | norm 0.3713 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22357 | train 2.880882 | val 3.305745 | tok/s 2632967.16 | norm 0.3511 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22358 | train 2.995241 | val 3.305745 | tok/s 2628683.75 | norm 0.3466 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22359 | train 2.820580 | val 3.305745 | tok/s 2630051.30 | norm 0.3682 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22360 | train 2.919598 | val 3.305745 | tok/s 2624063.85 | norm 0.3536 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22361 | train 2.821809 | val 3.305745 | tok/s 2626827.69 | norm 0.3676 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22362 | train 2.954123 | val 3.305745 | tok/s 2629894.48 | norm 0.4185 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22363 | train 2.986297 | val 3.305745 | tok/s 2626829.79 | norm 0.4164 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22364 | train 3.056220 | val 3.305745 | tok/s 2628369.84 | norm 0.3969 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22365 | train 2.860860 | val 3.305745 | tok/s 2626060.89 | norm 0.3750 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22366 | train 2.909305 | val 3.305745 | tok/s 2622714.06 | norm 0.3679 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22367 | train 3.006644 | val 3.305745 | tok/s 2628925.26 | norm 0.3777 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22368 | train 2.943603 | val 3.305745 | tok/s 2618496.22 | norm 0.3754 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22369 | train 2.817944 | val 3.305745 | tok/s 2627818.62 | norm 0.3918 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22370 | train 2.981853 | val 3.305745 | tok/s 2626049.61 | norm 0.3755 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22371 | train 2.935874 | val 3.305745 | tok/s 2625741.71 | norm 0.4085 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22372 | train 2.885034 | val 3.305745 | tok/s 2621582.56 | norm 0.4037 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22373 | train 2.919354 | val 3.305745 | tok/s 2620157.94 | norm 0.3857 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22374 | train 3.083106 | val 3.305745 | tok/s 2632124.03 | norm 0.3838 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22375 | train 3.019285 | val 3.305745 | tok/s 2626137.03 | norm 0.3700 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22376 | train 2.963831 | val 3.305745 | tok/s 2627841.38 | norm 0.3418 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22377 | train 3.051149 | val 3.305745 | tok/s 2628265.66 | norm 0.3511 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22378 | train 2.872579 | val 3.305745 | tok/s 2624131.36 | norm 0.3449 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22379 | train 2.967233 | val 3.305745 | tok/s 2632362.84 | norm 0.3486 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22380 | train 2.894915 | val 3.305745 | tok/s 2627328.43 | norm 0.3506 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22381 | train 2.933426 | val 3.305745 | tok/s 2627551.58 | norm 0.3623 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22382 | train 3.023064 | val 3.305745 | tok/s 2620993.66 | norm 0.3994 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22383 | train 2.837702 | val 3.305745 | tok/s 2623333.46 | norm 0.3938 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22384 | train 2.875227 | val 3.305745 | tok/s 2446925.09 | norm 0.3554 | dt 214.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22385 | train 2.962604 | val 3.305745 | tok/s 2613518.52 | norm 0.3668 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22386 | train 2.998160 | val 3.305745 | tok/s 2625065.23 | norm 0.3881 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22387 | train 3.112319 | val 3.305745 | tok/s 2623562.12 | norm 0.4666 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22388 | train 2.914396 | val 3.305745 | tok/s 2621802.21 | norm 0.3541 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22389 | train 2.886157 | val 3.305745 | tok/s 2627731.38 | norm 0.3441 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22390 | train 2.988293 | val 3.305745 | tok/s 2626707.95 | norm 0.3778 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22391 | train 2.952951 | val 3.305745 | tok/s 2523306.39 | norm 0.3377 | dt 207.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22392 | train 3.029684 | val 3.305745 | tok/s 2625071.25 | norm 0.3500 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22393 | train 2.864405 | val 3.305745 | tok/s 2620276.23 | norm 0.3428 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22394 | train 2.869797 | val 3.305745 | tok/s 2630780.34 | norm 0.3419 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22395 | train 3.048847 | val 3.305745 | tok/s 2628445.04 | norm 0.3692 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22396 | train 2.929035 | val 3.305745 | tok/s 2629388.98 | norm 0.3752 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22397 | train 2.877765 | val 3.305745 | tok/s 2623956.99 | norm 0.3883 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22398 | train 3.004296 | val 3.305745 | tok/s 2624014.13 | norm 0.3844 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22399 | train 3.000425 | val 3.305745 | tok/s 2627523.83 | norm 0.3706 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22400 | train 3.060419 | val 3.305745 | tok/s 2628373.58 | norm 0.4147 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22401 | train 2.828320 | val 3.305745 | tok/s 2628109.56 | norm 0.3598 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22402 | train 2.921217 | val 3.305745 | tok/s 2623797.78 | norm 0.3358 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22403 | train 2.869325 | val 3.305745 | tok/s 2623648.84 | norm 0.3372 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22404 | train 2.974046 | val 3.305745 | tok/s 2630876.44 | norm 0.3611 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22405 | train 2.995550 | val 3.305745 | tok/s 2623575.11 | norm 0.3794 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22406 | train 2.953052 | val 3.305745 | tok/s 2628642.81 | norm 0.3815 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22407 | train 2.939663 | val 3.305745 | tok/s 2626868.24 | norm 0.3830 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22408 | train 3.108294 | val 3.305745 | tok/s 2623021.46 | norm 0.3980 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22409 | train 2.862382 | val 3.305745 | tok/s 2629304.54 | norm 0.3441 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22410 | train 3.004111 | val 3.305745 | tok/s 2617819.75 | norm 0.3714 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22411 | train 2.822525 | val 3.305745 | tok/s 2624869.64 | norm 0.3632 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22412 | train 2.882062 | val 3.305745 | tok/s 2622490.02 | norm 0.3685 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22413 | train 2.936702 | val 3.305745 | tok/s 2627931.32 | norm 0.3787 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22414 | train 3.008413 | val 3.305745 | tok/s 2626131.64 | norm 0.3579 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22415 | train 2.887274 | val 3.305745 | tok/s 2619467.79 | norm 0.3719 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22416 | train 2.993958 | val 3.305745 | tok/s 2629687.98 | norm 0.3970 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22417 | train 3.073041 | val 3.305745 | tok/s 2625038.84 | norm 0.4615 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22418 | train 2.870330 | val 3.305745 | tok/s 2624095.73 | norm 0.3370 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22419 | train 2.875381 | val 3.305745 | tok/s 2621990.38 | norm 0.3547 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22420 | train 2.893165 | val 3.305745 | tok/s 2620336.41 | norm 0.3631 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22421 | train 3.112400 | val 3.305745 | tok/s 2628904.42 | norm 0.3825 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22422 | train 2.967513 | val 3.305745 | tok/s 2628024.20 | norm 0.3592 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22423 | train 2.929179 | val 3.305745 | tok/s 2630249.97 | norm 0.3222 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22424 | train 2.929025 | val 3.305745 | tok/s 2615303.15 | norm 0.3402 | dt 200.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22425 | train 2.806061 | val 3.305745 | tok/s 2574675.69 | norm 0.3562 | dt 203.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22426 | train 2.974414 | val 3.305745 | tok/s 2632630.27 | norm 0.3437 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22427 | train 2.857002 | val 3.305745 | tok/s 2542604.45 | norm 0.3406 | dt 206.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22428 | train 2.865792 | val 3.305745 | tok/s 2628646.66 | norm 0.3576 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22429 | train 2.867542 | val 3.305745 | tok/s 2620463.92 | norm 0.3637 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22430 | train 2.815454 | val 3.305745 | tok/s 2622372.48 | norm 0.3495 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22431 | train 2.833134 | val 3.305745 | tok/s 2629073.70 | norm 0.3214 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22432 | train 2.945219 | val 3.305745 | tok/s 2629448.14 | norm 0.3527 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22433 | train 2.954837 | val 3.305745 | tok/s 2488050.02 | norm 0.3767 | dt 210.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22434 | train 2.993309 | val 3.305745 | tok/s 2622890.62 | norm 0.3833 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22435 | train 2.908538 | val 3.305745 | tok/s 2623251.38 | norm 0.3581 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22436 | train 2.888235 | val 3.305745 | tok/s 2625603.72 | norm 0.3823 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22437 | train 2.958268 | val 3.305745 | tok/s 2623833.83 | norm 0.3575 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22438 | train 2.961710 | val 3.305745 | tok/s 2626067.08 | norm 0.3329 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22439 | train 2.879879 | val 3.305745 | tok/s 2621467.76 | norm 0.3487 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22440 | train 2.953260 | val 3.305745 | tok/s 2627406.78 | norm 0.3705 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22441 | train 2.954366 | val 3.305745 | tok/s 2630284.73 | norm 0.3435 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22442 | train 2.931555 | val 3.305745 | tok/s 2606089.56 | norm 0.3371 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22443 | train 3.064491 | val 3.305745 | tok/s 2592197.23 | norm 0.4399 | dt 202.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22444 | train 2.929456 | val 3.305745 | tok/s 2624462.58 | norm 0.3792 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22445 | train 2.902168 | val 3.305745 | tok/s 2626941.06 | norm 0.3592 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22446 | train 3.045345 | val 3.305745 | tok/s 2632616.56 | norm 0.3442 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22447 | train 2.962317 | val 3.305745 | tok/s 2624105.45 | norm 0.3386 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22448 | train 2.921594 | val 3.305745 | tok/s 2629404.57 | norm 0.3478 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22449 | train 2.976455 | val 3.305745 | tok/s 2624090.78 | norm 0.3535 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22450 | train 3.091496 | val 3.305745 | tok/s 2631221.06 | norm 0.4342 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22451 | train 2.880152 | val 3.305745 | tok/s 2629341.71 | norm 0.3835 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22452 | train 2.948790 | val 3.305745 | tok/s 2626573.11 | norm 0.4644 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22453 | train 2.956838 | val 3.305745 | tok/s 2625664.79 | norm 0.3427 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22454 | train 2.876302 | val 3.305745 | tok/s 2626010.47 | norm 0.3090 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22455 | train 2.928422 | val 3.305745 | tok/s 2626175.51 | norm 0.3605 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22456 | train 2.993290 | val 3.305745 | tok/s 2627729.12 | norm 0.3750 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22457 | train 2.953390 | val 3.305745 | tok/s 2618234.56 | norm 0.3906 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22458 | train 2.848100 | val 3.305745 | tok/s 2627432.49 | norm 0.3573 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22459 | train 2.910566 | val 3.305745 | tok/s 2622083.28 | norm 0.3584 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22460 | train 3.078513 | val 3.305745 | tok/s 2625756.63 | norm 0.4178 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22461 | train 2.979907 | val 3.305745 | tok/s 2625931.76 | norm 0.3623 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22462 | train 2.956060 | val 3.305745 | tok/s 2611583.83 | norm 0.3595 | dt 200.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22463 | train 2.842225 | val 3.305745 | tok/s 2631945.32 | norm 0.3642 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22464 | train 3.074376 | val 3.305745 | tok/s 2627702.73 | norm 0.3650 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22465 | train 2.999645 | val 3.305745 | tok/s 2628953.43 | norm 0.3782 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22466 | train 2.874424 | val 3.305745 | tok/s 2625621.59 | norm 0.3848 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22467 | train 3.028347 | val 3.305745 | tok/s 2604921.59 | norm 0.4539 | dt 201.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22468 | train 2.835065 | val 3.305745 | tok/s 2632757.28 | norm 0.3475 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22469 | train 2.979293 | val 3.305745 | tok/s 2631436.41 | norm 0.3126 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22470 | train 2.952921 | val 3.305745 | tok/s 2627198.86 | norm 0.3393 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22471 | train 2.814789 | val 3.305745 | tok/s 2620425.52 | norm 0.3460 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22472 | train 2.820648 | val 3.305745 | tok/s 2624000.95 | norm 0.3528 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22473 | train 2.839798 | val 3.305745 | tok/s 2627989.08 | norm 0.3436 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22474 | train 2.936103 | val 3.305745 | tok/s 2627172.28 | norm 0.3513 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22475 | train 2.986375 | val 3.305745 | tok/s 2630627.60 | norm 0.3576 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22476 | train 2.917980 | val 3.305745 | tok/s 2620352.33 | norm 0.3770 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22477 | train 3.028725 | val 3.305745 | tok/s 2617970.57 | norm 0.3468 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22478 | train 2.935066 | val 3.305745 | tok/s 2631486.01 | norm 0.3592 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22479 | train 2.797292 | val 3.305745 | tok/s 2631452.68 | norm 0.3697 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22480 | train 2.950078 | val 3.305745 | tok/s 2628517.94 | norm 0.3593 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22481 | train 2.975305 | val 3.305745 | tok/s 2630552.26 | norm 0.3579 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22482 | train 2.982152 | val 3.305745 | tok/s 2611451.69 | norm 0.3425 | dt 200.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22483 | train 2.891579 | val 3.305745 | tok/s 2632770.06 | norm 0.3457 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22484 | train 2.922391 | val 3.305745 | tok/s 2627222.50 | norm 0.3476 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22485 | train 3.086539 | val 3.305745 | tok/s 2602228.62 | norm 0.3490 | dt 201.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22486 | train 3.126448 | val 3.305745 | tok/s 2627690.74 | norm 0.3638 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22487 | train 2.994528 | val 3.305745 | tok/s 2623325.53 | norm 0.3361 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22488 | train 2.832771 | val 3.305745 | tok/s 2631436.60 | norm 0.3426 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22489 | train 2.951102 | val 3.305745 | tok/s 2610054.88 | norm 0.3885 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22490 | train 2.815072 | val 3.305745 | tok/s 2626646.20 | norm 0.3615 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22491 | train 2.961526 | val 3.305745 | tok/s 2628666.84 | norm 0.3536 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22492 | train 2.882186 | val 3.305745 | tok/s 2624158.69 | norm 0.3526 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22493 | train 2.906204 | val 3.305745 | tok/s 2626701.39 | norm 0.3776 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22494 | train 3.024112 | val 3.305745 | tok/s 2618406.81 | norm 0.3812 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22495 | train 2.919508 | val 3.305745 | tok/s 2629427.50 | norm 0.3552 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22496 | train 2.963230 | val 3.305745 | tok/s 2623575.62 | norm 0.3717 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22497 | train 3.006121 | val 3.305745 | tok/s 2628274.58 | norm 0.3731 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22498 | train 2.921017 | val 3.305745 | tok/s 2626963.94 | norm 0.3899 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22499 | train 2.871517 | val 3.305745 | tok/s 2618220.37 | norm 0.4320 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 22500 | val_loss 3.305542 step 22500 | train 3.024085 | val 3.305542 | tok/s 2613487.88 | norm 0.4135 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22501 | train 2.944627 | val 3.305542 | tok/s 2629356.39 | norm 0.3539 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22502 | train 2.979837 | val 3.305542 | tok/s 2623468.37 | norm 0.3694 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22503 | train 2.935216 | val 3.305542 | tok/s 2630481.82 | norm 0.3772 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22504 | train 2.961090 | val 3.305542 | tok/s 2625515.33 | norm 0.3817 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22505 | train 3.014870 | val 3.305542 | tok/s 2624503.96 | norm 0.3729 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22506 | train 2.930896 | val 3.305542 | tok/s 2630976.69 | norm 0.4028 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22507 | train 2.925567 | val 3.305542 | tok/s 2624106.46 | norm 0.4039 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22508 | train 2.945936 | val 3.305542 | tok/s 2624241.15 | norm 0.3806 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22509 | train 2.956907 | val 3.305542 | tok/s 2620318.24 | norm 0.3801 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22510 | train 2.870764 | val 3.305542 | tok/s 2622445.67 | norm 0.3445 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22511 | train 2.805242 | val 3.305542 | tok/s 2627143.96 | norm 0.3504 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22512 | train 2.892667 | val 3.305542 | tok/s 2626241.74 | norm 0.3977 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22513 | train 2.997303 | val 3.305542 | tok/s 2628029.56 | norm 0.3900 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22514 | train 2.892465 | val 3.305542 | tok/s 2623889.98 | norm 0.3580 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22515 | train 2.970678 | val 3.305542 | tok/s 2623154.87 | norm 0.3822 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22516 | train 2.964023 | val 3.305542 | tok/s 2626587.55 | norm 0.4086 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22517 | train 2.830096 | val 3.305542 | tok/s 2628287.83 | norm 0.3392 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22518 | train 2.895653 | val 3.305542 | tok/s 2630171.13 | norm 0.3929 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22519 | train 2.946660 | val 3.305542 | tok/s 2610797.71 | norm 0.4000 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22520 | train 2.953807 | val 3.305542 | tok/s 2625605.42 | norm 0.3871 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22521 | train 2.922618 | val 3.305542 | tok/s 2631903.09 | norm 0.3591 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22522 | train 2.920905 | val 3.305542 | tok/s 2632439.72 | norm 0.3431 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22523 | train 3.178273 | val 3.305542 | tok/s 2626288.80 | norm 0.4206 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22524 | train 3.173450 | val 3.305542 | tok/s 2625095.42 | norm 0.4309 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22525 | train 3.002635 | val 3.305542 | tok/s 2626991.87 | norm 0.3552 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22526 | train 2.921145 | val 3.305542 | tok/s 2630242.03 | norm 0.3387 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22527 | train 2.995332 | val 3.305542 | tok/s 2545805.12 | norm 0.3616 | dt 205.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22528 | train 2.849440 | val 3.305542 | tok/s 2577944.44 | norm 0.3381 | dt 203.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22529 | train 2.747989 | val 3.305542 | tok/s 2621031.37 | norm 0.3406 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22530 | train 2.925431 | val 3.305542 | tok/s 2623380.28 | norm 0.3549 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22531 | train 2.925302 | val 3.305542 | tok/s 2630608.01 | norm 0.3572 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22532 | train 3.042320 | val 3.305542 | tok/s 2624648.49 | norm 0.3595 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22533 | train 3.042361 | val 3.305542 | tok/s 2628239.37 | norm 0.3630 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22534 | train 2.891169 | val 3.305542 | tok/s 2628289.64 | norm 0.3460 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22535 | train 3.217843 | val 3.305542 | tok/s 2614929.19 | norm 0.6124 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22536 | train 3.019023 | val 3.305542 | tok/s 2634853.30 | norm 0.3473 | dt 198.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22537 | train 2.922618 | val 3.305542 | tok/s 2631055.13 | norm 0.2992 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22538 | train 2.943518 | val 3.305542 | tok/s 2627311.15 | norm 0.3419 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22539 | train 2.900277 | val 3.305542 | tok/s 2628364.77 | norm 0.3762 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22540 | train 2.859400 | val 3.305542 | tok/s 2631172.40 | norm 0.3479 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22541 | train 2.916375 | val 3.305542 | tok/s 2630544.66 | norm 0.3199 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22542 | train 2.934543 | val 3.305542 | tok/s 2620113.20 | norm 0.3460 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22543 | train 3.035990 | val 3.305542 | tok/s 2631368.63 | norm 0.3938 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22544 | train 2.975848 | val 3.305542 | tok/s 2624528.94 | norm 0.3909 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22545 | train 3.060237 | val 3.305542 | tok/s 2629316.01 | norm 0.3801 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22546 | train 2.944452 | val 3.305542 | tok/s 2623368.96 | norm 0.3621 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22547 | train 2.797363 | val 3.305542 | tok/s 2618851.00 | norm 0.3669 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22548 | train 3.011342 | val 3.305542 | tok/s 2631932.83 | norm 0.4010 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22549 | train 2.928405 | val 3.305542 | tok/s 2630858.97 | norm 0.3389 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22550 | train 2.988832 | val 3.305542 | tok/s 2632582.97 | norm 0.3467 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22551 | train 2.842032 | val 3.305542 | tok/s 2620997.35 | norm 0.3611 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22552 | train 2.877192 | val 3.305542 | tok/s 2626358.49 | norm 0.3688 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22553 | train 2.968340 | val 3.305542 | tok/s 2634784.42 | norm 0.3587 | dt 198.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22554 | train 3.009028 | val 3.305542 | tok/s 2630819.04 | norm 0.3995 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22555 | train 3.016820 | val 3.305542 | tok/s 2633208.86 | norm 0.3832 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22556 | train 2.999920 | val 3.305542 | tok/s 2617277.77 | norm 0.3313 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22557 | train 3.055496 | val 3.305542 | tok/s 2601754.89 | norm 0.3353 | dt 201.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22558 | train 2.802863 | val 3.305542 | tok/s 2627480.71 | norm 0.3610 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22559 | train 3.166680 | val 3.305542 | tok/s 2626031.81 | norm 0.4342 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22560 | train 2.941252 | val 3.305542 | tok/s 2626890.63 | norm 0.3523 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22561 | train 3.099626 | val 3.305542 | tok/s 2622306.54 | norm 0.3331 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22562 | train 3.136989 | val 3.305542 | tok/s 2612500.34 | norm 0.4160 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22563 | train 2.887802 | val 3.305542 | tok/s 2564410.57 | norm 0.3928 | dt 204.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22564 | train 3.159839 | val 3.305542 | tok/s 2625867.34 | norm 0.3647 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22565 | train 3.066090 | val 3.305542 | tok/s 2626169.15 | norm 0.4721 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22566 | train 2.895158 | val 3.305542 | tok/s 2621269.83 | norm 0.3555 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22567 | train 2.924371 | val 3.305542 | tok/s 2623437.20 | norm 0.3174 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22568 | train 2.951342 | val 3.305542 | tok/s 2631414.79 | norm 0.3607 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22569 | train 3.041289 | val 3.305542 | tok/s 2625049.39 | norm 0.3599 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22570 | train 2.901718 | val 3.305542 | tok/s 2628847.03 | norm 0.3613 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22571 | train 2.934105 | val 3.305542 | tok/s 2622323.92 | norm 0.3526 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22572 | train 2.990424 | val 3.305542 | tok/s 2624217.75 | norm 0.3678 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22573 | train 2.909488 | val 3.305542 | tok/s 2626813.37 | norm 0.3971 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22574 | train 2.938039 | val 3.305542 | tok/s 2627156.63 | norm 0.3642 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22575 | train 2.951457 | val 3.305542 | tok/s 2628430.64 | norm 0.3536 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22576 | train 2.975093 | val 3.305542 | tok/s 2544697.08 | norm 0.3740 | dt 206.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22577 | train 2.915704 | val 3.305542 | tok/s 2577825.26 | norm 0.3746 | dt 203.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22578 | train 2.927253 | val 3.305542 | tok/s 2625524.36 | norm 0.3302 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22579 | train 3.054221 | val 3.305542 | tok/s 2628221.11 | norm 0.3533 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22580 | train 2.919172 | val 3.305542 | tok/s 2624499.54 | norm 0.3362 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22581 | train 2.796609 | val 3.305542 | tok/s 2622150.45 | norm 0.3468 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22582 | train 3.037121 | val 3.305542 | tok/s 2625010.87 | norm 0.3768 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22583 | train 2.909291 | val 3.305542 | tok/s 2629210.08 | norm 0.3533 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22584 | train 2.973545 | val 3.305542 | tok/s 2621187.61 | norm 0.3399 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22585 | train 2.957413 | val 3.305542 | tok/s 2626717.22 | norm 0.3453 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22586 | train 2.973897 | val 3.305542 | tok/s 2624959.74 | norm 0.3282 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22587 | train 2.951676 | val 3.305542 | tok/s 2624960.15 | norm 0.3363 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22588 | train 2.916924 | val 3.305542 | tok/s 2627237.68 | norm 0.3432 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22589 | train 2.961844 | val 3.305542 | tok/s 2617348.76 | norm 0.3707 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22590 | train 2.912535 | val 3.305542 | tok/s 2625728.55 | norm 0.3922 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22591 | train 3.041444 | val 3.305542 | tok/s 2626437.19 | norm 0.3553 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22592 | train 2.955683 | val 3.305542 | tok/s 2627856.24 | norm 0.3424 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22593 | train 2.909920 | val 3.305542 | tok/s 2628329.77 | norm 0.3798 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22594 | train 2.844187 | val 3.305542 | tok/s 2621298.63 | norm 0.3731 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22595 | train 2.812283 | val 3.305542 | tok/s 2627959.97 | norm 0.3550 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22596 | train 2.939789 | val 3.305542 | tok/s 2622513.09 | norm 0.3531 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22597 | train 2.979181 | val 3.305542 | tok/s 2624839.19 | norm 0.3531 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22598 | train 2.882600 | val 3.305542 | tok/s 2630167.53 | norm 0.3560 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22599 | train 3.018667 | val 3.305542 | tok/s 2627562.21 | norm 0.4304 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22600 | train 2.893771 | val 3.305542 | tok/s 2628839.06 | norm 0.3144 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22601 | train 2.853915 | val 3.305542 | tok/s 2626127.61 | norm 0.3007 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22602 | train 2.930744 | val 3.305542 | tok/s 2628696.19 | norm 0.3576 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22603 | train 2.920999 | val 3.305542 | tok/s 2618335.82 | norm 0.3594 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22604 | train 2.880026 | val 3.305542 | tok/s 2623508.58 | norm 0.3577 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22605 | train 2.814332 | val 3.305542 | tok/s 2630041.39 | norm 0.3663 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22606 | train 3.063465 | val 3.305542 | tok/s 2626124.85 | norm 0.3510 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22607 | train 2.976631 | val 3.305542 | tok/s 2627888.22 | norm 0.3572 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22608 | train 2.869505 | val 3.305542 | tok/s 2627703.16 | norm 0.3701 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22609 | train 2.916031 | val 3.305542 | tok/s 2626574.51 | norm 0.4253 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22610 | train 2.918494 | val 3.305542 | tok/s 2629459.83 | norm 0.3961 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22611 | train 2.922016 | val 3.305542 | tok/s 2624006.66 | norm 0.3676 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22612 | train 3.081128 | val 3.305542 | tok/s 2629130.87 | norm 0.3714 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22613 | train 2.931389 | val 3.305542 | tok/s 2619599.76 | norm 0.3533 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22614 | train 3.013842 | val 3.305542 | tok/s 2623004.98 | norm 0.3394 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22615 | train 2.958368 | val 3.305542 | tok/s 2627818.86 | norm 0.3709 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22616 | train 2.936049 | val 3.305542 | tok/s 2529600.90 | norm 0.3854 | dt 207.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22617 | train 3.086751 | val 3.305542 | tok/s 2482660.91 | norm 0.3925 | dt 211.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22618 | train 2.992523 | val 3.305542 | tok/s 2627791.40 | norm 0.3688 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22619 | train 2.981976 | val 3.305542 | tok/s 2628514.81 | norm 0.3695 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22620 | train 3.043310 | val 3.305542 | tok/s 2633936.39 | norm 0.3817 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22621 | train 2.953905 | val 3.305542 | tok/s 2627716.19 | norm 0.3812 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22622 | train 3.062684 | val 3.305542 | tok/s 2628984.08 | norm 0.3589 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22623 | train 2.889566 | val 3.305542 | tok/s 2626115.72 | norm 0.3360 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22624 | train 2.987906 | val 3.305542 | tok/s 2574413.45 | norm 0.3422 | dt 203.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22625 | train 2.944451 | val 3.305542 | tok/s 2602190.94 | norm 0.3949 | dt 201.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22626 | train 2.895304 | val 3.305542 | tok/s 2627470.94 | norm 0.3667 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22627 | train 2.949742 | val 3.305542 | tok/s 2631694.12 | norm 0.3599 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22628 | train 2.840425 | val 3.305542 | tok/s 2628497.42 | norm 0.3811 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22629 | train 2.847605 | val 3.305542 | tok/s 2629910.07 | norm 0.3812 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22630 | train 2.931133 | val 3.305542 | tok/s 2630289.23 | norm 0.3443 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22631 | train 2.818806 | val 3.305542 | tok/s 2631263.76 | norm 0.3673 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22632 | train 2.878566 | val 3.305542 | tok/s 2630676.77 | norm 0.3935 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22633 | train 2.848327 | val 3.305542 | tok/s 2625048.97 | norm 0.3868 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22634 | train 2.881742 | val 3.305542 | tok/s 2600603.97 | norm 0.3442 | dt 201.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22635 | train 2.966780 | val 3.305542 | tok/s 2629783.63 | norm 0.3652 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22636 | train 2.929488 | val 3.305542 | tok/s 2618253.60 | norm 0.3493 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22637 | train 3.007417 | val 3.305542 | tok/s 2630344.93 | norm 0.3630 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22638 | train 2.931612 | val 3.305542 | tok/s 2625964.94 | norm 0.3668 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22639 | train 2.973067 | val 3.305542 | tok/s 2630757.09 | norm 0.3733 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22640 | train 3.424144 | val 3.305542 | tok/s 2617639.89 | norm 0.4680 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22641 | train 2.816795 | val 3.305542 | tok/s 2616330.87 | norm 0.3839 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22642 | train 3.037929 | val 3.305542 | tok/s 2629451.16 | norm 0.3245 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22643 | train 3.000768 | val 3.305542 | tok/s 2622739.92 | norm 0.3286 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22644 | train 2.923375 | val 3.305542 | tok/s 2621467.20 | norm 0.3600 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22645 | train 2.909989 | val 3.305542 | tok/s 2621585.22 | norm 0.3659 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22646 | train 2.978336 | val 3.305542 | tok/s 2619283.91 | norm 0.3594 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22647 | train 2.950210 | val 3.305542 | tok/s 2629705.23 | norm 0.3620 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22648 | train 2.916009 | val 3.305542 | tok/s 2623525.15 | norm 0.3555 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22649 | train 2.974359 | val 3.305542 | tok/s 2627048.93 | norm 0.3488 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22650 | train 2.843425 | val 3.305542 | tok/s 2620113.92 | norm 0.3653 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22651 | train 2.924503 | val 3.305542 | tok/s 2623394.90 | norm 0.3652 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22652 | train 2.976696 | val 3.305542 | tok/s 2631511.77 | norm 0.3432 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22653 | train 2.999931 | val 3.305542 | tok/s 2633999.07 | norm 0.3537 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22654 | train 2.897771 | val 3.305542 | tok/s 2630972.06 | norm 0.3757 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22655 | train 3.016514 | val 3.305542 | tok/s 2619612.43 | norm 0.3942 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22656 | train 2.999497 | val 3.305542 | tok/s 2618593.99 | norm 0.3570 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22657 | train 2.798994 | val 3.305542 | tok/s 2630742.73 | norm 0.3500 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22658 | train 2.918276 | val 3.305542 | tok/s 2626440.56 | norm 0.3884 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22659 | train 2.851864 | val 3.305542 | tok/s 2624030.52 | norm 0.3952 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22660 | train 2.779831 | val 3.305542 | tok/s 2617777.59 | norm 0.3584 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22661 | train 2.838990 | val 3.305542 | tok/s 2624280.92 | norm 0.3618 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22662 | train 3.130166 | val 3.305542 | tok/s 2616902.17 | norm 0.4038 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22663 | train 2.912447 | val 3.305542 | tok/s 2625605.51 | norm 0.3548 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22664 | train 2.738132 | val 3.305542 | tok/s 2532625.53 | norm 0.3827 | dt 207.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22665 | train 3.002223 | val 3.305542 | tok/s 2621651.73 | norm 0.3846 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22666 | train 3.008822 | val 3.305542 | tok/s 2625845.35 | norm 0.3260 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22667 | train 2.856013 | val 3.305542 | tok/s 2630289.53 | norm 0.3131 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22668 | train 2.945627 | val 3.305542 | tok/s 2626585.85 | norm 0.3223 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22669 | train 2.835864 | val 3.305542 | tok/s 2617188.66 | norm 0.3431 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22670 | train 2.920836 | val 3.305542 | tok/s 2628388.09 | norm 0.3650 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22671 | train 2.875502 | val 3.305542 | tok/s 2628834.12 | norm 0.3835 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22672 | train 2.983595 | val 3.305542 | tok/s 2632163.30 | norm 0.3925 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22673 | train 2.899632 | val 3.305542 | tok/s 2625892.65 | norm 0.3785 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22674 | train 2.964460 | val 3.305542 | tok/s 2631980.99 | norm 0.3799 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22675 | train 2.980926 | val 3.305542 | tok/s 2625204.98 | norm 0.3804 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22676 | train 2.925143 | val 3.305542 | tok/s 2625877.17 | norm 0.3954 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22677 | train 2.909179 | val 3.305542 | tok/s 2627186.52 | norm 0.3697 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22678 | train 3.027670 | val 3.305542 | tok/s 2626874.41 | norm 0.3774 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22679 | train 2.784319 | val 3.305542 | tok/s 2626705.39 | norm 0.3863 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22680 | train 2.933928 | val 3.305542 | tok/s 2624994.18 | norm 0.3664 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22681 | train 2.981686 | val 3.305542 | tok/s 2609307.65 | norm 0.3675 | dt 200.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22682 | train 3.004419 | val 3.305542 | tok/s 2627030.73 | norm 0.3532 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22683 | train 2.934390 | val 3.305542 | tok/s 2627058.73 | norm 0.3535 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22684 | train 2.839412 | val 3.305542 | tok/s 2629174.19 | norm 0.3989 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22685 | train 2.981424 | val 3.305542 | tok/s 2628757.07 | norm 0.3797 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22686 | train 2.939245 | val 3.305542 | tok/s 2630585.31 | norm 0.3512 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22687 | train 2.996222 | val 3.305542 | tok/s 2625663.29 | norm 0.3945 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22688 | train 2.950624 | val 3.305542 | tok/s 2617431.73 | norm 0.3732 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22689 | train 2.974936 | val 3.305542 | tok/s 2624832.44 | norm 0.3547 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22690 | train 2.862643 | val 3.305542 | tok/s 2626990.20 | norm 0.3300 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22691 | train 2.924937 | val 3.305542 | tok/s 2628413.05 | norm 0.3358 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22692 | train 2.944168 | val 3.305542 | tok/s 2626573.05 | norm 0.3796 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22693 | train 2.978689 | val 3.305542 | tok/s 2621759.74 | norm 0.3786 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22694 | train 2.929891 | val 3.305542 | tok/s 2630334.53 | norm 0.3399 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22695 | train 2.983623 | val 3.305542 | tok/s 2621686.31 | norm 0.3577 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22696 | train 2.953178 | val 3.305542 | tok/s 2629419.96 | norm 0.3528 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22697 | train 3.247846 | val 3.305542 | tok/s 2620900.64 | norm 0.4542 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22698 | train 2.925672 | val 3.305542 | tok/s 2621292.78 | norm 0.3456 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22699 | train 2.899703 | val 3.305542 | tok/s 2625480.56 | norm 0.3308 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22700 | train 2.865627 | val 3.305542 | tok/s 2625675.00 | norm 0.3005 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22701 | train 2.897277 | val 3.305542 | tok/s 2630029.95 | norm 0.3478 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22702 | train 2.948346 | val 3.305542 | tok/s 2627373.21 | norm 0.3972 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22703 | train 2.958754 | val 3.305542 | tok/s 2626492.33 | norm 0.3797 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22704 | train 2.922805 | val 3.305542 | tok/s 2609391.07 | norm 0.3534 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22705 | train 3.034369 | val 3.305542 | tok/s 2626696.85 | norm 0.3497 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22706 | train 2.970577 | val 3.305542 | tok/s 2626526.33 | norm 0.3740 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22707 | train 2.935243 | val 3.305542 | tok/s 2622439.86 | norm 0.3414 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22708 | train 2.944470 | val 3.305542 | tok/s 2622611.82 | norm 0.3488 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22709 | train 2.820431 | val 3.305542 | tok/s 2630094.34 | norm 0.3761 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22710 | train 2.994098 | val 3.305542 | tok/s 2616534.19 | norm 0.3924 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22711 | train 2.940859 | val 3.305542 | tok/s 2621725.27 | norm 0.3696 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22712 | train 2.956417 | val 3.305542 | tok/s 2603958.17 | norm 0.3533 | dt 201.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22713 | train 2.985217 | val 3.305542 | tok/s 2626812.99 | norm 0.3819 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22714 | train 2.851372 | val 3.305542 | tok/s 2625402.99 | norm 0.3997 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22715 | train 2.852209 | val 3.305542 | tok/s 2628836.13 | norm 0.4035 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22716 | train 3.011958 | val 3.305542 | tok/s 2626336.98 | norm 0.3892 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22717 | train 2.829833 | val 3.305542 | tok/s 2623226.27 | norm 0.3626 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22718 | train 3.014593 | val 3.305542 | tok/s 2631391.74 | norm 0.3820 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22719 | train 2.890330 | val 3.305542 | tok/s 2628670.86 | norm 0.3484 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22720 | train 2.941572 | val 3.305542 | tok/s 2630288.77 | norm 0.3229 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22721 | train 2.912213 | val 3.305542 | tok/s 2627794.59 | norm 0.3922 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22722 | train 2.925703 | val 3.305542 | tok/s 2622824.20 | norm 0.3509 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22723 | train 3.039008 | val 3.305542 | tok/s 2630678.48 | norm 0.3350 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22724 | train 2.996216 | val 3.305542 | tok/s 2625634.81 | norm 0.3422 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22725 | train 2.859393 | val 3.305542 | tok/s 2629827.49 | norm 0.3653 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22726 | train 2.907770 | val 3.305542 | tok/s 2618096.86 | norm 0.3626 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22727 | train 3.107468 | val 3.305542 | tok/s 2625640.68 | norm 0.3561 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22728 | train 2.889873 | val 3.305542 | tok/s 2623756.51 | norm 0.3412 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22729 | train 2.886310 | val 3.305542 | tok/s 2625207.17 | norm 0.3379 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22730 | train 2.946293 | val 3.305542 | tok/s 2628038.44 | norm 0.3865 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22731 | train 2.925034 | val 3.305542 | tok/s 2622212.92 | norm 0.3684 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22732 | train 2.946813 | val 3.305542 | tok/s 2618574.57 | norm 0.3680 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22733 | train 2.947699 | val 3.305542 | tok/s 2628505.55 | norm 0.3831 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22734 | train 2.884730 | val 3.305542 | tok/s 2622971.89 | norm 0.3567 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22735 | train 2.974385 | val 3.305542 | tok/s 2624533.33 | norm 0.3384 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22736 | train 2.932995 | val 3.305542 | tok/s 2618191.49 | norm 0.3477 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22737 | train 2.948858 | val 3.305542 | tok/s 2622374.88 | norm 0.3718 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22738 | train 2.897335 | val 3.305542 | tok/s 2629314.53 | norm 0.3584 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22739 | train 2.887258 | val 3.305542 | tok/s 2621111.05 | norm 0.3580 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22740 | train 2.962898 | val 3.305542 | tok/s 2629491.16 | norm 0.3742 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22741 | train 2.965546 | val 3.305542 | tok/s 2620530.06 | norm 0.3248 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22742 | train 2.850104 | val 3.305542 | tok/s 2623024.71 | norm 0.3441 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22743 | train 2.883168 | val 3.305542 | tok/s 2630956.97 | norm 0.3568 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22744 | train 2.852333 | val 3.305542 | tok/s 2625646.17 | norm 0.3488 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22745 | train 2.969956 | val 3.305542 | tok/s 2630094.71 | norm 0.3869 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22746 | train 3.036382 | val 3.305542 | tok/s 2619307.41 | norm 0.3821 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22747 | train 3.074621 | val 3.305542 | tok/s 2623674.86 | norm 0.3598 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22748 | train 3.052260 | val 3.305542 | tok/s 2627844.33 | norm 0.3768 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22749 | train 2.973717 | val 3.305542 | tok/s 2625178.94 | norm 0.3546 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 22750 | val_loss 3.301970 step 22750 | train 2.995283 | val 3.301970 | tok/s 2606075.01 | norm 0.3561 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22751 | train 2.931134 | val 3.301970 | tok/s 2626930.39 | norm 0.3656 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22752 | train 2.838468 | val 3.301970 | tok/s 2629515.25 | norm 0.3651 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22753 | train 2.915662 | val 3.301970 | tok/s 2626331.75 | norm 0.3839 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22754 | train 2.905277 | val 3.301970 | tok/s 2629006.03 | norm 0.3860 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22755 | train 2.921702 | val 3.301970 | tok/s 2627856.38 | norm 0.3819 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22756 | train 2.932957 | val 3.301970 | tok/s 2628403.40 | norm 0.3661 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22757 | train 3.021628 | val 3.301970 | tok/s 2624398.63 | norm 0.3659 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22758 | train 2.884780 | val 3.301970 | tok/s 2624800.56 | norm 0.3654 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22759 | train 2.952591 | val 3.301970 | tok/s 2628127.23 | norm 0.3593 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22760 | train 3.000022 | val 3.301970 | tok/s 2627588.53 | norm 0.3384 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22761 | train 2.977901 | val 3.301970 | tok/s 2629240.46 | norm 0.3378 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22762 | train 2.980451 | val 3.301970 | tok/s 2616679.36 | norm 0.3582 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22763 | train 2.876979 | val 3.301970 | tok/s 2507174.54 | norm 0.3361 | dt 209.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22764 | train 2.952800 | val 3.301970 | tok/s 2626878.30 | norm 0.3197 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22765 | train 2.904729 | val 3.301970 | tok/s 2629002.19 | norm 0.3603 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22766 | train 2.882075 | val 3.301970 | tok/s 2630544.16 | norm 0.3517 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22767 | train 2.880354 | val 3.301970 | tok/s 2630050.77 | norm 0.3649 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22768 | train 2.994053 | val 3.301970 | tok/s 2628916.60 | norm 0.3848 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22769 | train 2.976212 | val 3.301970 | tok/s 2628230.99 | norm 0.3644 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22770 | train 2.918412 | val 3.301970 | tok/s 2606652.82 | norm 0.3668 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22771 | train 2.950812 | val 3.301970 | tok/s 2616462.81 | norm 0.3389 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22772 | train 2.998972 | val 3.301970 | tok/s 2624364.43 | norm 0.3461 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22773 | train 2.861806 | val 3.301970 | tok/s 2629569.99 | norm 0.3443 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22774 | train 2.951580 | val 3.301970 | tok/s 2626389.42 | norm 0.3736 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22775 | train 2.805660 | val 3.301970 | tok/s 2618563.86 | norm 0.3598 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22776 | train 2.911310 | val 3.301970 | tok/s 2631492.99 | norm 0.3945 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22777 | train 2.987583 | val 3.301970 | tok/s 2631114.70 | norm 0.3277 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22778 | train 2.791173 | val 3.301970 | tok/s 2625662.81 | norm 0.3504 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22779 | train 2.994868 | val 3.301970 | tok/s 2620773.77 | norm 0.3624 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22780 | train 2.960112 | val 3.301970 | tok/s 2622937.79 | norm 0.3523 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22781 | train 3.178144 | val 3.301970 | tok/s 2628713.76 | norm 0.3959 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22782 | train 3.028361 | val 3.301970 | tok/s 2631149.24 | norm 0.3567 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22783 | train 2.895346 | val 3.301970 | tok/s 2629256.35 | norm 0.3212 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22784 | train 2.863944 | val 3.301970 | tok/s 2622571.89 | norm 0.3330 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22785 | train 2.928109 | val 3.301970 | tok/s 2625579.86 | norm 0.3448 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22786 | train 2.893938 | val 3.301970 | tok/s 2632524.27 | norm 0.3459 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22787 | train 2.986921 | val 3.301970 | tok/s 2625068.84 | norm 0.3778 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22788 | train 2.968711 | val 3.301970 | tok/s 2630805.64 | norm 0.3607 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22789 | train 2.992901 | val 3.301970 | tok/s 2622572.35 | norm 0.3606 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22790 | train 2.987718 | val 3.301970 | tok/s 2617318.87 | norm 0.3422 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22791 | train 3.001863 | val 3.301970 | tok/s 2632140.52 | norm 0.3561 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22792 | train 3.022663 | val 3.301970 | tok/s 2629792.25 | norm 0.4592 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22793 | train 2.736358 | val 3.301970 | tok/s 2626091.22 | norm 0.3992 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22794 | train 2.953128 | val 3.301970 | tok/s 2617678.49 | norm 0.3574 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22795 | train 2.969276 | val 3.301970 | tok/s 2625001.60 | norm 0.3728 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22796 | train 2.939205 | val 3.301970 | tok/s 2630362.97 | norm 0.3390 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22797 | train 2.917646 | val 3.301970 | tok/s 2628077.22 | norm 0.3691 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22798 | train 2.998431 | val 3.301970 | tok/s 2627213.22 | norm 0.5488 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22799 | train 2.954401 | val 3.301970 | tok/s 2624721.82 | norm 0.3607 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22800 | train 2.962169 | val 3.301970 | tok/s 2626260.65 | norm 0.3404 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22801 | train 2.997727 | val 3.301970 | tok/s 2628477.39 | norm 0.3597 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22802 | train 2.916524 | val 3.301970 | tok/s 2626718.11 | norm 0.3468 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22803 | train 3.026390 | val 3.301970 | tok/s 2630191.11 | norm 0.3656 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22804 | train 3.057945 | val 3.301970 | tok/s 2409961.27 | norm 0.3924 | dt 217.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22805 | train 3.181793 | val 3.301970 | tok/s 2614918.46 | norm 0.3523 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22806 | train 3.019362 | val 3.301970 | tok/s 2629542.66 | norm 0.3937 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22807 | train 3.022718 | val 3.301970 | tok/s 2534111.50 | norm 0.3855 | dt 206.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22808 | train 3.063157 | val 3.301970 | tok/s 2617992.86 | norm 0.3634 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22809 | train 2.872987 | val 3.301970 | tok/s 2625144.99 | norm 0.3780 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22810 | train 2.986323 | val 3.301970 | tok/s 2626435.70 | norm 0.4107 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22811 | train 2.968269 | val 3.301970 | tok/s 2616868.03 | norm 0.3832 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22812 | train 2.949380 | val 3.301970 | tok/s 2617822.29 | norm 0.3306 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22813 | train 3.003298 | val 3.301970 | tok/s 2628836.90 | norm 0.3466 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22814 | train 2.982529 | val 3.301970 | tok/s 2625748.68 | norm 0.3544 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22815 | train 2.879674 | val 3.301970 | tok/s 2629256.05 | norm 0.3724 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22816 | train 2.911691 | val 3.301970 | tok/s 2630719.52 | norm 0.4009 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22817 | train 3.031488 | val 3.301970 | tok/s 2621330.35 | norm 0.3801 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22818 | train 2.956377 | val 3.301970 | tok/s 2627004.46 | norm 0.3173 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22819 | train 3.054016 | val 3.301970 | tok/s 2619284.80 | norm 0.3195 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22820 | train 2.862501 | val 3.301970 | tok/s 2618096.19 | norm 0.3659 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22821 | train 3.019825 | val 3.301970 | tok/s 2628742.63 | norm 0.3397 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22822 | train 3.002166 | val 3.301970 | tok/s 2627073.76 | norm 0.3741 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22823 | train 2.874260 | val 3.301970 | tok/s 2631069.69 | norm 0.3858 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22824 | train 2.968938 | val 3.301970 | tok/s 2625782.93 | norm 0.3778 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22825 | train 2.919601 | val 3.301970 | tok/s 2626588.85 | norm 0.3973 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22826 | train 2.933709 | val 3.301970 | tok/s 2617788.64 | norm 0.3879 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22827 | train 2.895324 | val 3.301970 | tok/s 2625395.08 | norm 0.3649 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22828 | train 2.824720 | val 3.301970 | tok/s 2628398.94 | norm 0.3590 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22829 | train 3.011419 | val 3.301970 | tok/s 2623870.43 | norm 0.3768 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22830 | train 2.877452 | val 3.301970 | tok/s 2627373.67 | norm 0.3602 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22831 | train 2.878518 | val 3.301970 | tok/s 2619520.69 | norm 0.3388 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22832 | train 2.956681 | val 3.301970 | tok/s 2153229.76 | norm 0.4442 | dt 243.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22833 | train 2.938538 | val 3.301970 | tok/s 2630702.84 | norm 0.3622 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22834 | train 3.002372 | val 3.301970 | tok/s 2621851.09 | norm 0.3707 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22835 | train 2.942416 | val 3.301970 | tok/s 2627789.38 | norm 0.3699 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22836 | train 3.072978 | val 3.301970 | tok/s 2597602.98 | norm 0.4023 | dt 201.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22837 | train 2.868817 | val 3.301970 | tok/s 2624374.09 | norm 0.3895 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22838 | train 2.935096 | val 3.301970 | tok/s 2627601.77 | norm 0.3428 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22839 | train 2.886196 | val 3.301970 | tok/s 2622687.46 | norm 0.3792 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22840 | train 2.934052 | val 3.301970 | tok/s 2628535.07 | norm 0.4051 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22841 | train 2.911411 | val 3.301970 | tok/s 2626259.23 | norm 0.4052 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22842 | train 3.030496 | val 3.301970 | tok/s 2628451.41 | norm 0.4333 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22843 | train 2.975825 | val 3.301970 | tok/s 2628671.22 | norm 0.4034 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22844 | train 2.963706 | val 3.301970 | tok/s 2627169.63 | norm 0.3760 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22845 | train 2.954316 | val 3.301970 | tok/s 2630331.76 | norm 0.3639 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22846 | train 2.962398 | val 3.301970 | tok/s 2625886.57 | norm 0.3722 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22847 | train 2.857153 | val 3.301970 | tok/s 2626263.01 | norm 0.3625 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22848 | train 3.014646 | val 3.301970 | tok/s 2610155.27 | norm 0.3605 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22849 | train 2.968748 | val 3.301970 | tok/s 2623281.82 | norm 0.3490 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22850 | train 2.867167 | val 3.301970 | tok/s 2625909.98 | norm 0.3490 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22851 | train 3.046647 | val 3.301970 | tok/s 2622736.67 | norm 0.3452 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22852 | train 2.927875 | val 3.301970 | tok/s 2628120.70 | norm 0.3293 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22853 | train 2.888722 | val 3.301970 | tok/s 2497040.86 | norm 0.3333 | dt 209.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22854 | train 2.981112 | val 3.301970 | tok/s 2624173.80 | norm 0.3395 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22855 | train 2.980110 | val 3.301970 | tok/s 2628527.76 | norm 0.3410 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22856 | train 2.890327 | val 3.301970 | tok/s 2626392.79 | norm 0.3585 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22857 | train 2.891438 | val 3.301970 | tok/s 2631306.56 | norm 0.3382 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22858 | train 2.911604 | val 3.301970 | tok/s 2623613.23 | norm 0.3613 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22859 | train 3.082097 | val 3.301970 | tok/s 2625574.95 | norm 0.4060 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22860 | train 2.961678 | val 3.301970 | tok/s 2525779.41 | norm 0.3523 | dt 207.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22861 | train 2.862433 | val 3.301970 | tok/s 2471526.86 | norm 0.3681 | dt 212.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22862 | train 2.970836 | val 3.301970 | tok/s 2620580.96 | norm 0.3855 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22863 | train 3.024685 | val 3.301970 | tok/s 2623296.74 | norm 0.3536 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22864 | train 2.879843 | val 3.301970 | tok/s 2622823.74 | norm 0.3781 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22865 | train 3.216448 | val 3.301970 | tok/s 2628867.26 | norm 0.4470 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22866 | train 2.917719 | val 3.301970 | tok/s 2624482.93 | norm 0.3713 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22867 | train 2.924699 | val 3.301970 | tok/s 2629995.84 | norm 0.3686 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22868 | train 2.867790 | val 3.301970 | tok/s 2622040.64 | norm 0.3724 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22869 | train 2.917578 | val 3.301970 | tok/s 2629631.88 | norm 0.3611 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22870 | train 2.958782 | val 3.301970 | tok/s 2623108.44 | norm 0.3704 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22871 | train 2.942066 | val 3.301970 | tok/s 2623494.43 | norm 0.3368 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22872 | train 2.981042 | val 3.301970 | tok/s 2629395.51 | norm 0.4123 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22873 | train 2.804025 | val 3.301970 | tok/s 2627306.12 | norm 0.3994 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22874 | train 2.957070 | val 3.301970 | tok/s 2627982.19 | norm 0.3706 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22875 | train 2.967689 | val 3.301970 | tok/s 2617411.67 | norm 0.3379 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22876 | train 2.929914 | val 3.301970 | tok/s 2620810.47 | norm 0.3649 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22877 | train 2.984825 | val 3.301970 | tok/s 2629741.79 | norm 0.3412 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22878 | train 2.920336 | val 3.301970 | tok/s 2628479.00 | norm 0.3337 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22879 | train 2.982650 | val 3.301970 | tok/s 2625036.86 | norm 0.3467 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22880 | train 2.898443 | val 3.301970 | tok/s 2613231.55 | norm 0.3490 | dt 200.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22881 | train 2.965340 | val 3.301970 | tok/s 2619958.35 | norm 0.3552 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22882 | train 2.877761 | val 3.301970 | tok/s 2630542.00 | norm 0.3603 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22883 | train 2.822361 | val 3.301970 | tok/s 2626905.50 | norm 0.3613 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22884 | train 2.900878 | val 3.301970 | tok/s 2631960.46 | norm 0.3895 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22885 | train 2.869292 | val 3.301970 | tok/s 2615027.43 | norm 0.3733 | dt 200.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22886 | train 2.964800 | val 3.301970 | tok/s 2623811.56 | norm 0.3949 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22887 | train 2.932738 | val 3.301970 | tok/s 2628100.00 | norm 0.3845 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22888 | train 2.911001 | val 3.301970 | tok/s 2627126.68 | norm 0.3868 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22889 | train 2.910233 | val 3.301970 | tok/s 2631745.11 | norm 0.3921 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22890 | train 3.006185 | val 3.301970 | tok/s 2623180.49 | norm 0.3859 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22891 | train 3.003756 | val 3.301970 | tok/s 2618866.74 | norm 0.3760 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22892 | train 3.085670 | val 3.301970 | tok/s 2631346.73 | norm 0.3889 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22893 | train 2.979513 | val 3.301970 | tok/s 2554878.80 | norm 0.3766 | dt 205.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22894 | train 2.894309 | val 3.301970 | tok/s 2625340.89 | norm 0.3622 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22895 | train 2.903012 | val 3.301970 | tok/s 2632353.10 | norm 0.3960 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22896 | train 2.949955 | val 3.301970 | tok/s 2624502.18 | norm 0.3646 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22897 | train 2.926437 | val 3.301970 | tok/s 2627379.52 | norm 0.3401 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22898 | train 2.985027 | val 3.301970 | tok/s 2627556.66 | norm 0.3688 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22899 | train 2.947661 | val 3.301970 | tok/s 2626654.00 | norm 0.4097 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22900 | train 3.013098 | val 3.301970 | tok/s 2628528.79 | norm 0.3735 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22901 | train 2.917170 | val 3.301970 | tok/s 2482091.95 | norm 0.4156 | dt 211.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22902 | train 2.963304 | val 3.301970 | tok/s 2631570.68 | norm 0.4063 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22903 | train 2.916883 | val 3.301970 | tok/s 2609398.57 | norm 0.4073 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22904 | train 2.846740 | val 3.301970 | tok/s 2631413.17 | norm 0.3977 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22905 | train 2.913713 | val 3.301970 | tok/s 2625711.61 | norm 0.3396 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22906 | train 2.835270 | val 3.301970 | tok/s 2631258.49 | norm 0.3463 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22907 | train 2.883365 | val 3.301970 | tok/s 2627953.62 | norm 0.3721 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22908 | train 2.852984 | val 3.301970 | tok/s 2621973.15 | norm 0.3719 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22909 | train 2.979371 | val 3.301970 | tok/s 2631335.64 | norm 0.3328 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22910 | train 2.957799 | val 3.301970 | tok/s 2622871.94 | norm 0.3443 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22911 | train 2.932476 | val 3.301970 | tok/s 2630443.39 | norm 0.3807 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22912 | train 3.029150 | val 3.301970 | tok/s 2629720.85 | norm 0.3989 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22913 | train 2.907743 | val 3.301970 | tok/s 2622794.78 | norm 0.3533 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22914 | train 2.919855 | val 3.301970 | tok/s 2629531.91 | norm 0.3511 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22915 | train 2.827878 | val 3.301970 | tok/s 2592348.25 | norm 0.3765 | dt 202.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22916 | train 2.925226 | val 3.301970 | tok/s 2629289.64 | norm 0.3523 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22917 | train 2.903868 | val 3.301970 | tok/s 2618376.08 | norm 0.3674 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22918 | train 2.872726 | val 3.301970 | tok/s 2606726.85 | norm 0.3759 | dt 201.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22919 | train 2.805079 | val 3.301970 | tok/s 2630053.90 | norm 0.4163 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22920 | train 2.889702 | val 3.301970 | tok/s 2624572.16 | norm 0.3762 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22921 | train 2.927146 | val 3.301970 | tok/s 2628334.62 | norm 0.4081 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22922 | train 2.987050 | val 3.301970 | tok/s 2628977.63 | norm 0.3930 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22923 | train 2.998519 | val 3.301970 | tok/s 2629655.03 | norm 0.3829 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22924 | train 2.927669 | val 3.301970 | tok/s 2630756.51 | norm 0.3844 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22925 | train 2.923247 | val 3.301970 | tok/s 2626018.69 | norm 0.3952 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22926 | train 3.096425 | val 3.301970 | tok/s 2626757.99 | norm 0.3516 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22927 | train 2.879566 | val 3.301970 | tok/s 2620342.90 | norm 0.3586 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22928 | train 2.940387 | val 3.301970 | tok/s 2625839.63 | norm 0.3694 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22929 | train 2.932664 | val 3.301970 | tok/s 2626845.36 | norm 0.3778 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22930 | train 2.884111 | val 3.301970 | tok/s 2626375.47 | norm 0.3731 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22931 | train 2.822677 | val 3.301970 | tok/s 2629616.11 | norm 0.3655 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22932 | train 2.898153 | val 3.301970 | tok/s 2622673.51 | norm 0.3545 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22933 | train 2.948884 | val 3.301970 | tok/s 2626264.54 | norm 0.3844 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22934 | train 2.890111 | val 3.301970 | tok/s 2631719.51 | norm 0.3809 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22935 | train 2.923833 | val 3.301970 | tok/s 2629469.57 | norm 0.3550 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22936 | train 2.995543 | val 3.301970 | tok/s 2629019.62 | norm 0.3623 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22937 | train 2.956140 | val 3.301970 | tok/s 2621415.42 | norm 0.3712 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22938 | train 2.844727 | val 3.301970 | tok/s 2623700.57 | norm 0.3615 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22939 | train 2.922124 | val 3.301970 | tok/s 2545046.08 | norm 0.3570 | dt 206.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22940 | train 2.912401 | val 3.301970 | tok/s 2607556.98 | norm 0.4258 | dt 201.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22941 | train 2.913702 | val 3.301970 | tok/s 2628153.92 | norm 0.3763 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22942 | train 2.902451 | val 3.301970 | tok/s 2625085.99 | norm 0.3786 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22943 | train 2.872755 | val 3.301970 | tok/s 2628360.51 | norm 0.3897 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22944 | train 2.925972 | val 3.301970 | tok/s 2630886.73 | norm 0.3772 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22945 | train 2.983918 | val 3.301970 | tok/s 2629604.05 | norm 0.3915 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22946 | train 3.087731 | val 3.301970 | tok/s 2628494.51 | norm 0.4313 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22947 | train 2.881650 | val 3.301970 | tok/s 2489073.62 | norm 0.3925 | dt 210.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22948 | train 2.910022 | val 3.301970 | tok/s 2581939.31 | norm 0.3362 | dt 203.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22949 | train 2.957584 | val 3.301970 | tok/s 2628750.75 | norm 0.4006 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22950 | train 2.873567 | val 3.301970 | tok/s 2622497.00 | norm 0.3635 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22951 | train 2.957084 | val 3.301970 | tok/s 2628413.55 | norm 0.3680 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22952 | train 2.894040 | val 3.301970 | tok/s 2627009.59 | norm 0.3791 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22953 | train 2.953349 | val 3.301970 | tok/s 2630341.97 | norm 0.3511 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22954 | train 2.921408 | val 3.301970 | tok/s 2620828.52 | norm 0.3708 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22955 | train 2.938152 | val 3.301970 | tok/s 2619485.73 | norm 0.3722 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22956 | train 2.871736 | val 3.301970 | tok/s 2630275.12 | norm 0.3276 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22957 | train 2.909779 | val 3.301970 | tok/s 2626442.35 | norm 0.3310 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22958 | train 2.875645 | val 3.301970 | tok/s 2629657.01 | norm 0.3470 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22959 | train 2.945236 | val 3.301970 | tok/s 2623843.53 | norm 0.3299 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22960 | train 2.977886 | val 3.301970 | tok/s 2624021.38 | norm 0.3755 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22961 | train 2.919610 | val 3.301970 | tok/s 2630768.32 | norm 0.4032 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22962 | train 2.896220 | val 3.301970 | tok/s 2625181.62 | norm 0.3773 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22963 | train 2.937891 | val 3.301970 | tok/s 2630144.01 | norm 0.3638 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22964 | train 2.992890 | val 3.301970 | tok/s 2626676.81 | norm 0.3576 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22965 | train 2.927738 | val 3.301970 | tok/s 2619796.14 | norm 0.3718 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22966 | train 2.870289 | val 3.301970 | tok/s 2630918.29 | norm 0.4080 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22967 | train 2.985898 | val 3.301970 | tok/s 2629647.54 | norm 0.3821 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22968 | train 2.895758 | val 3.301970 | tok/s 2631159.90 | norm 0.3730 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22969 | train 3.038507 | val 3.301970 | tok/s 2621337.30 | norm 0.4234 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22970 | train 2.933380 | val 3.301970 | tok/s 2619988.64 | norm 0.3453 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22971 | train 2.890450 | val 3.301970 | tok/s 2630693.21 | norm 0.3335 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22972 | train 3.021808 | val 3.301970 | tok/s 2626776.06 | norm 0.4031 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22973 | train 2.857963 | val 3.301970 | tok/s 2631117.05 | norm 0.3864 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22974 | train 2.972717 | val 3.301970 | tok/s 2625126.26 | norm 0.3681 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22975 | train 2.979678 | val 3.301970 | tok/s 2623274.63 | norm 0.3459 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22976 | train 2.876164 | val 3.301970 | tok/s 2628511.18 | norm 0.3620 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22977 | train 2.900624 | val 3.301970 | tok/s 2627946.56 | norm 0.4035 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22978 | train 2.956904 | val 3.301970 | tok/s 2625913.11 | norm 0.3585 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22979 | train 2.962297 | val 3.301970 | tok/s 2537773.82 | norm 0.3566 | dt 206.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22980 | train 2.919222 | val 3.301970 | tok/s 2607844.73 | norm 0.3805 | dt 201.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22981 | train 3.059887 | val 3.301970 | tok/s 2628390.98 | norm 0.3531 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22982 | train 2.883450 | val 3.301970 | tok/s 2630319.97 | norm 0.3548 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22983 | train 2.896891 | val 3.301970 | tok/s 2632102.52 | norm 0.3597 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22984 | train 2.869883 | val 3.301970 | tok/s 2629006.86 | norm 0.3677 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22985 | train 2.929050 | val 3.301970 | tok/s 2625747.43 | norm 0.3677 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22986 | train 2.958677 | val 3.301970 | tok/s 2630795.37 | norm 0.3620 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22987 | train 2.785618 | val 3.301970 | tok/s 2508584.72 | norm 0.3536 | dt 209.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22988 | train 2.984949 | val 3.301970 | tok/s 2628506.79 | norm 0.3665 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22989 | train 2.996804 | val 3.301970 | tok/s 2626986.00 | norm 0.3464 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22990 | train 2.849500 | val 3.301970 | tok/s 2625513.04 | norm 0.3500 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22991 | train 2.917590 | val 3.301970 | tok/s 2624562.81 | norm 0.3702 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22992 | train 2.937823 | val 3.301970 | tok/s 2619403.77 | norm 0.3527 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22993 | train 3.011224 | val 3.301970 | tok/s 2631478.19 | norm 0.3497 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22994 | train 2.755723 | val 3.301970 | tok/s 2620781.21 | norm 0.3611 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22995 | train 2.940647 | val 3.301970 | tok/s 2625793.92 | norm 0.3980 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22996 | train 3.052402 | val 3.301970 | tok/s 2628406.24 | norm 0.3933 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22997 | train 2.934880 | val 3.301970 | tok/s 2625310.06 | norm 0.3724 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22998 | train 3.070335 | val 3.301970 | tok/s 2625966.23 | norm 0.3711 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 22999 | train 2.864190 | val 3.301970 | tok/s 2627751.83 | norm 0.3505 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 23000 | val_loss 3.302221 step 23000 | train 2.871871 | val 3.302221 | tok/s 2611420.55 | norm 0.3986 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23001 | train 2.905471 | val 3.302221 | tok/s 2625322.33 | norm 0.3703 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23002 | train 2.966887 | val 3.302221 | tok/s 2627377.52 | norm 0.3331 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23003 | train 3.171615 | val 3.302221 | tok/s 2623008.69 | norm 0.3995 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23004 | train 2.991668 | val 3.302221 | tok/s 2631426.55 | norm 0.3587 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23005 | train 2.980158 | val 3.302221 | tok/s 2629685.37 | norm 0.3607 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23006 | train 2.855157 | val 3.302221 | tok/s 2625000.83 | norm 0.3951 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23007 | train 2.968889 | val 3.302221 | tok/s 2609446.55 | norm 0.3815 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23008 | train 3.079080 | val 3.302221 | tok/s 2621010.77 | norm 0.3598 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23009 | train 2.852894 | val 3.302221 | tok/s 2631779.50 | norm 0.3180 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23010 | train 2.857908 | val 3.302221 | tok/s 2626184.77 | norm 0.3410 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23011 | train 2.789038 | val 3.302221 | tok/s 2626911.45 | norm 0.3438 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23012 | train 3.031576 | val 3.302221 | tok/s 2625921.69 | norm 0.3546 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23013 | train 2.836470 | val 3.302221 | tok/s 2621580.01 | norm 0.3343 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23014 | train 2.929782 | val 3.302221 | tok/s 2627859.89 | norm 0.3217 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23015 | train 3.022956 | val 3.302221 | tok/s 2623187.55 | norm 0.3493 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23016 | train 2.848416 | val 3.302221 | tok/s 2628425.28 | norm 0.3815 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23017 | train 2.892783 | val 3.302221 | tok/s 2622293.18 | norm 0.3419 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23018 | train 2.937268 | val 3.302221 | tok/s 2625065.04 | norm 0.3270 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23019 | train 2.913537 | val 3.302221 | tok/s 2628382.99 | norm 0.3334 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23020 | train 3.036505 | val 3.302221 | tok/s 2617792.85 | norm 0.3747 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23021 | train 2.860010 | val 3.302221 | tok/s 2618845.14 | norm 0.3443 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23022 | train 2.825828 | val 3.302221 | tok/s 2619204.72 | norm 0.3635 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23023 | train 2.989003 | val 3.302221 | tok/s 2620353.29 | norm 0.3959 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23024 | train 2.837571 | val 3.302221 | tok/s 2625314.06 | norm 0.3389 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23025 | train 3.080259 | val 3.302221 | tok/s 2624655.19 | norm 0.3810 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23026 | train 2.958431 | val 3.302221 | tok/s 2626831.74 | norm 0.3922 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23027 | train 2.979405 | val 3.302221 | tok/s 2621581.56 | norm 0.3584 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23028 | train 2.887502 | val 3.302221 | tok/s 2625320.49 | norm 0.3907 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23029 | train 2.874352 | val 3.302221 | tok/s 2631697.30 | norm 0.3991 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23030 | train 2.934372 | val 3.302221 | tok/s 2627832.52 | norm 0.4040 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23031 | train 2.917629 | val 3.302221 | tok/s 2627670.42 | norm 0.3844 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23032 | train 2.931854 | val 3.302221 | tok/s 2626132.79 | norm 0.3675 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23033 | train 2.867043 | val 3.302221 | tok/s 2623893.67 | norm 0.3489 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23034 | train 2.897796 | val 3.302221 | tok/s 2628193.67 | norm 0.3716 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23035 | train 2.761173 | val 3.302221 | tok/s 2628525.41 | norm 0.3952 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23036 | train 3.035268 | val 3.302221 | tok/s 2630875.07 | norm 0.3662 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23037 | train 3.091343 | val 3.302221 | tok/s 2625396.14 | norm 0.3534 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23038 | train 2.921078 | val 3.302221 | tok/s 2626850.45 | norm 0.3769 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23039 | train 2.941019 | val 3.302221 | tok/s 2629397.65 | norm 0.3641 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23040 | train 2.954005 | val 3.302221 | tok/s 2624512.46 | norm 0.3487 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23041 | train 2.934422 | val 3.302221 | tok/s 2628440.81 | norm 0.4558 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23042 | train 3.045912 | val 3.302221 | tok/s 2626775.57 | norm 0.4052 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23043 | train 2.964031 | val 3.302221 | tok/s 2626011.36 | norm 0.3689 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23044 | train 3.062025 | val 3.302221 | tok/s 2627758.31 | norm 0.3445 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23045 | train 2.981035 | val 3.302221 | tok/s 2621070.53 | norm 0.3458 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23046 | train 2.955007 | val 3.302221 | tok/s 2625405.74 | norm 0.4071 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23047 | train 2.925514 | val 3.302221 | tok/s 2619280.83 | norm 0.4966 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23048 | train 2.926139 | val 3.302221 | tok/s 2620789.35 | norm 0.3431 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23049 | train 3.006549 | val 3.302221 | tok/s 2617825.71 | norm 0.3259 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23050 | train 2.992197 | val 3.302221 | tok/s 2613756.25 | norm 0.3439 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23051 | train 3.216117 | val 3.302221 | tok/s 2623543.15 | norm 0.4224 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23052 | train 2.996492 | val 3.302221 | tok/s 2623232.26 | norm 0.3919 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23053 | train 2.996475 | val 3.302221 | tok/s 2624896.04 | norm 0.3444 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23054 | train 3.002028 | val 3.302221 | tok/s 2617606.80 | norm 0.3272 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23055 | train 2.952697 | val 3.302221 | tok/s 2583551.76 | norm 0.3490 | dt 202.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23056 | train 2.899849 | val 3.302221 | tok/s 2630912.97 | norm 0.3823 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23057 | train 2.904793 | val 3.302221 | tok/s 2628380.09 | norm 0.3959 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23058 | train 2.886274 | val 3.302221 | tok/s 2630974.18 | norm 0.3634 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23059 | train 2.893062 | val 3.302221 | tok/s 2628039.20 | norm 0.3461 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23060 | train 2.945925 | val 3.302221 | tok/s 2625793.86 | norm 0.3500 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23061 | train 2.922410 | val 3.302221 | tok/s 2627056.87 | norm 0.3800 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23062 | train 2.970168 | val 3.302221 | tok/s 2625589.01 | norm 0.3739 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23063 | train 2.841138 | val 3.302221 | tok/s 2627726.42 | norm 0.3562 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23064 | train 2.925720 | val 3.302221 | tok/s 2623497.86 | norm 0.3577 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23065 | train 2.887271 | val 3.302221 | tok/s 2610908.11 | norm 0.3598 | dt 200.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23066 | train 2.941582 | val 3.302221 | tok/s 2626586.59 | norm 0.3602 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23067 | train 2.978905 | val 3.302221 | tok/s 2626023.51 | norm 0.4124 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23068 | train 2.946264 | val 3.302221 | tok/s 2630631.78 | norm 0.4019 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23069 | train 2.929753 | val 3.302221 | tok/s 2621985.31 | norm 0.3555 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23070 | train 2.898236 | val 3.302221 | tok/s 2624593.34 | norm 0.3457 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23071 | train 2.870348 | val 3.302221 | tok/s 2631889.71 | norm 0.3716 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23072 | train 2.858927 | val 3.302221 | tok/s 2628947.02 | norm 0.3940 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23073 | train 2.878147 | val 3.302221 | tok/s 2625691.26 | norm 0.4014 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23074 | train 2.973788 | val 3.302221 | tok/s 2532094.18 | norm 0.3711 | dt 207.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23075 | train 2.923079 | val 3.302221 | tok/s 2563265.79 | norm 0.3804 | dt 204.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23076 | train 2.922402 | val 3.302221 | tok/s 2624714.00 | norm 0.3655 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23077 | train 2.853229 | val 3.302221 | tok/s 2620242.77 | norm 0.3321 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23078 | train 2.831524 | val 3.302221 | tok/s 2624042.66 | norm 0.3433 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23079 | train 3.005323 | val 3.302221 | tok/s 2622666.20 | norm 0.3369 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23080 | train 2.863739 | val 3.302221 | tok/s 2625534.52 | norm 0.3334 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23081 | train 3.021183 | val 3.302221 | tok/s 2629662.60 | norm 0.3236 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23082 | train 2.993198 | val 3.302221 | tok/s 2627815.09 | norm 0.3314 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23083 | train 3.045641 | val 3.302221 | tok/s 2628011.37 | norm 0.3403 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23084 | train 3.026492 | val 3.302221 | tok/s 2625148.80 | norm 0.3439 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23085 | train 3.011594 | val 3.302221 | tok/s 2629205.66 | norm 0.3601 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23086 | train 2.855285 | val 3.302221 | tok/s 2615374.74 | norm 0.3273 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23087 | train 2.932375 | val 3.302221 | tok/s 2621849.15 | norm 0.3672 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23088 | train 2.902346 | val 3.302221 | tok/s 2631798.80 | norm 0.3819 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23089 | train 3.063051 | val 3.302221 | tok/s 2628076.27 | norm 0.3995 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23090 | train 2.881559 | val 3.302221 | tok/s 2627099.67 | norm 0.3523 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23091 | train 3.011232 | val 3.302221 | tok/s 2628659.86 | norm 0.3593 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23092 | train 2.978402 | val 3.302221 | tok/s 2617406.81 | norm 0.3650 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23093 | train 3.023723 | val 3.302221 | tok/s 2629539.92 | norm 0.3410 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23094 | train 2.919045 | val 3.302221 | tok/s 2627904.01 | norm 0.3218 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23095 | train 2.899742 | val 3.302221 | tok/s 2629895.02 | norm 0.3510 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23096 | train 2.948711 | val 3.302221 | tok/s 2615236.15 | norm 0.4053 | dt 200.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23097 | train 2.971231 | val 3.302221 | tok/s 2620147.70 | norm 0.3815 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23098 | train 2.896712 | val 3.302221 | tok/s 2626136.03 | norm 0.3670 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23099 | train 3.010209 | val 3.302221 | tok/s 2624121.04 | norm 0.3619 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23100 | train 3.186549 | val 3.302221 | tok/s 2629641.30 | norm 0.4682 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23101 | train 2.905329 | val 3.302221 | tok/s 2620454.98 | norm 0.3391 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23102 | train 2.894530 | val 3.302221 | tok/s 2622604.70 | norm 0.3437 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23103 | train 2.855998 | val 3.302221 | tok/s 2631170.74 | norm 0.3793 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23104 | train 2.926847 | val 3.302221 | tok/s 2627427.05 | norm 0.3699 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23105 | train 2.924347 | val 3.302221 | tok/s 2627582.41 | norm 0.3895 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23106 | train 3.018295 | val 3.302221 | tok/s 2623108.57 | norm 0.3568 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23107 | train 2.944337 | val 3.302221 | tok/s 2624857.41 | norm 0.3819 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23108 | train 2.953635 | val 3.302221 | tok/s 2630972.12 | norm 0.3857 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23109 | train 3.030854 | val 3.302221 | tok/s 2622539.89 | norm 0.3593 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23110 | train 2.884084 | val 3.302221 | tok/s 2627907.20 | norm 0.3744 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23111 | train 2.954658 | val 3.302221 | tok/s 2617822.86 | norm 0.3354 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23112 | train 2.808630 | val 3.302221 | tok/s 2623287.80 | norm 0.3338 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23113 | train 2.985848 | val 3.302221 | tok/s 2626005.20 | norm 0.3582 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23114 | train 2.875290 | val 3.302221 | tok/s 2622896.73 | norm 0.3511 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23115 | train 2.815020 | val 3.302221 | tok/s 2623934.94 | norm 0.3695 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23116 | train 2.887097 | val 3.302221 | tok/s 2623399.61 | norm 0.3998 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23117 | train 2.908054 | val 3.302221 | tok/s 2623509.11 | norm 0.3775 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23118 | train 3.001669 | val 3.302221 | tok/s 2632023.19 | norm 0.3326 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23119 | train 2.981539 | val 3.302221 | tok/s 2626802.68 | norm 0.3470 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23120 | train 2.843235 | val 3.302221 | tok/s 2628449.59 | norm 0.3674 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23121 | train 2.941932 | val 3.302221 | tok/s 2617619.44 | norm 0.3728 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23122 | train 2.938283 | val 3.302221 | tok/s 2481222.52 | norm 0.3751 | dt 211.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23123 | train 2.854163 | val 3.302221 | tok/s 2504400.39 | norm 0.3652 | dt 209.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23124 | train 3.125324 | val 3.302221 | tok/s 2615784.82 | norm 0.3592 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23125 | train 2.839815 | val 3.302221 | tok/s 2614065.80 | norm 0.3596 | dt 200.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23126 | train 2.890647 | val 3.302221 | tok/s 2626459.82 | norm 0.3311 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23127 | train 2.965914 | val 3.302221 | tok/s 2629258.98 | norm 0.3755 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23128 | train 2.945513 | val 3.302221 | tok/s 2621219.35 | norm 0.3600 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23129 | train 2.880026 | val 3.302221 | tok/s 2622141.55 | norm 0.3743 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23130 | train 2.897824 | val 3.302221 | tok/s 2627492.21 | norm 0.3944 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23131 | train 3.029023 | val 3.302221 | tok/s 2628568.90 | norm 0.3988 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23132 | train 2.913519 | val 3.302221 | tok/s 2629988.91 | norm 0.3780 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23133 | train 2.936577 | val 3.302221 | tok/s 2621586.40 | norm 0.3640 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23134 | train 2.969361 | val 3.302221 | tok/s 2626061.04 | norm 0.3384 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23135 | train 2.977376 | val 3.302221 | tok/s 2633992.66 | norm 0.3518 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23136 | train 2.929703 | val 3.302221 | tok/s 2619297.24 | norm 0.3471 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23137 | train 2.922583 | val 3.302221 | tok/s 2621817.91 | norm 0.3489 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23138 | train 2.919972 | val 3.302221 | tok/s 2609143.10 | norm 0.3765 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23139 | train 2.891524 | val 3.302221 | tok/s 2617612.97 | norm 0.3740 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23140 | train 3.004044 | val 3.302221 | tok/s 2625082.60 | norm 0.4177 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23141 | train 2.955018 | val 3.302221 | tok/s 2620539.26 | norm 0.3609 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23142 | train 2.804736 | val 3.302221 | tok/s 2624256.71 | norm 0.3441 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23143 | train 2.972100 | val 3.302221 | tok/s 2613551.49 | norm 0.3772 | dt 200.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23144 | train 2.910070 | val 3.302221 | tok/s 2604179.07 | norm 0.3612 | dt 201.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23145 | train 2.932145 | val 3.302221 | tok/s 2624879.81 | norm 0.3228 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23146 | train 3.039057 | val 3.302221 | tok/s 2621766.54 | norm 0.3306 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23147 | train 2.857966 | val 3.302221 | tok/s 2620473.32 | norm 0.3248 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23148 | train 3.039391 | val 3.302221 | tok/s 2616148.42 | norm 0.3162 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23149 | train 3.015517 | val 3.302221 | tok/s 2621612.62 | norm 0.3523 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23150 | train 3.048548 | val 3.302221 | tok/s 2626904.41 | norm 0.4107 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23151 | train 3.045037 | val 3.302221 | tok/s 2628424.17 | norm 0.3941 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23152 | train 3.030046 | val 3.302221 | tok/s 2627300.90 | norm 0.3820 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23153 | train 2.877355 | val 3.302221 | tok/s 2609923.19 | norm 0.3697 | dt 200.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23154 | train 2.822396 | val 3.302221 | tok/s 2620771.70 | norm 0.3910 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23155 | train 2.920430 | val 3.302221 | tok/s 2632915.10 | norm 0.3480 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23156 | train 2.997429 | val 3.302221 | tok/s 2627540.65 | norm 0.3251 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23157 | train 2.818079 | val 3.302221 | tok/s 2629600.83 | norm 0.3288 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23158 | train 3.011847 | val 3.302221 | tok/s 2624946.50 | norm 0.3796 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23159 | train 2.920322 | val 3.302221 | tok/s 2624592.11 | norm 0.3572 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23160 | train 2.977662 | val 3.302221 | tok/s 2623973.28 | norm 0.3172 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23161 | train 2.901476 | val 3.302221 | tok/s 2623431.45 | norm 0.3570 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23162 | train 2.925208 | val 3.302221 | tok/s 2622287.81 | norm 0.3777 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23163 | train 2.942079 | val 3.302221 | tok/s 2574157.12 | norm 0.3476 | dt 203.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23164 | train 2.839906 | val 3.302221 | tok/s 2449310.40 | norm 0.3646 | dt 214.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23165 | train 2.966212 | val 3.302221 | tok/s 2618547.67 | norm 0.3953 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23166 | train 2.910918 | val 3.302221 | tok/s 2616252.95 | norm 0.3917 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23167 | train 2.949378 | val 3.302221 | tok/s 2629380.29 | norm 0.3700 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23168 | train 2.954373 | val 3.302221 | tok/s 2622702.67 | norm 0.3766 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23169 | train 2.965165 | val 3.302221 | tok/s 2628498.54 | norm 0.3717 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23170 | train 3.072595 | val 3.302221 | tok/s 2621364.87 | norm 0.4090 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23171 | train 2.741584 | val 3.302221 | tok/s 2620044.75 | norm 0.4087 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23172 | train 2.836219 | val 3.302221 | tok/s 2629827.04 | norm 0.3494 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23173 | train 2.936458 | val 3.302221 | tok/s 2612975.77 | norm 0.3365 | dt 200.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23174 | train 2.833694 | val 3.302221 | tok/s 2624612.01 | norm 0.3598 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23175 | train 2.972381 | val 3.302221 | tok/s 2616549.34 | norm 0.3646 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23176 | train 3.000803 | val 3.302221 | tok/s 2624485.43 | norm 0.3629 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23177 | train 2.987898 | val 3.302221 | tok/s 2630802.61 | norm 0.3269 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23178 | train 2.877905 | val 3.302221 | tok/s 2627351.26 | norm 0.3457 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23179 | train 2.970912 | val 3.302221 | tok/s 2626937.14 | norm 0.3663 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23180 | train 2.838765 | val 3.302221 | tok/s 2617983.78 | norm 0.3387 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23181 | train 2.837084 | val 3.302221 | tok/s 2623646.46 | norm 0.3809 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23182 | train 2.894400 | val 3.302221 | tok/s 2631658.27 | norm 0.3644 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23183 | train 3.235528 | val 3.302221 | tok/s 2628028.74 | norm 0.3611 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23184 | train 3.041938 | val 3.302221 | tok/s 2628982.62 | norm 0.4173 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23185 | train 3.027089 | val 3.302221 | tok/s 2612277.50 | norm 0.3447 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23186 | train 2.987472 | val 3.302221 | tok/s 2620921.24 | norm 0.4574 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23187 | train 3.110547 | val 3.302221 | tok/s 2623829.88 | norm 0.3514 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23188 | train 2.941735 | val 3.302221 | tok/s 2620431.81 | norm 0.3579 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23189 | train 3.030633 | val 3.302221 | tok/s 2618669.23 | norm 0.3916 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23190 | train 3.096135 | val 3.302221 | tok/s 2618463.10 | norm 0.3485 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23191 | train 3.014739 | val 3.302221 | tok/s 2617433.06 | norm 0.3296 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23192 | train 2.852683 | val 3.302221 | tok/s 2624811.25 | norm 0.3724 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23193 | train 3.038838 | val 3.302221 | tok/s 2619207.22 | norm 0.3922 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23194 | train 2.904776 | val 3.302221 | tok/s 2620191.58 | norm 0.3504 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23195 | train 2.981764 | val 3.302221 | tok/s 2617799.73 | norm 0.3862 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23196 | train 3.024514 | val 3.302221 | tok/s 2619676.87 | norm 0.3775 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23197 | train 2.994667 | val 3.302221 | tok/s 2623787.57 | norm 0.3294 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23198 | train 2.993313 | val 3.302221 | tok/s 2627452.25 | norm 0.3321 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23199 | train 2.946583 | val 3.302221 | tok/s 2624707.14 | norm 0.3584 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23200 | train 2.995786 | val 3.302221 | tok/s 2618419.33 | norm 0.3928 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23201 | train 2.843558 | val 3.302221 | tok/s 2621627.33 | norm 0.3621 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23202 | train 2.939301 | val 3.302221 | tok/s 2630727.18 | norm 0.3866 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23203 | train 2.967841 | val 3.302221 | tok/s 2468253.73 | norm 0.3874 | dt 212.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23204 | train 3.107530 | val 3.302221 | tok/s 2600068.31 | norm 0.4486 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23205 | train 2.991026 | val 3.302221 | tok/s 2630748.88 | norm 0.3748 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23206 | train 2.880853 | val 3.302221 | tok/s 2626564.45 | norm 0.3559 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23207 | train 2.972230 | val 3.302221 | tok/s 2630605.67 | norm 0.4018 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23208 | train 3.017556 | val 3.302221 | tok/s 2626490.81 | norm 0.3506 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23209 | train 3.042037 | val 3.302221 | tok/s 2630237.74 | norm 0.3376 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23210 | train 2.905993 | val 3.302221 | tok/s 2481277.45 | norm 0.3304 | dt 211.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23211 | train 2.846036 | val 3.302221 | tok/s 2614269.76 | norm 0.3709 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23212 | train 2.921521 | val 3.302221 | tok/s 2587970.75 | norm 0.3769 | dt 202.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23213 | train 3.048758 | val 3.302221 | tok/s 2618837.09 | norm 0.3744 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23214 | train 2.883728 | val 3.302221 | tok/s 2629126.66 | norm 0.3698 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23215 | train 2.797538 | val 3.302221 | tok/s 2627515.84 | norm 0.3700 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23216 | train 3.035879 | val 3.302221 | tok/s 2627911.28 | norm 0.3666 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23217 | train 2.942230 | val 3.302221 | tok/s 2621471.09 | norm 0.3462 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23218 | train 2.998020 | val 3.302221 | tok/s 2592217.45 | norm 0.3299 | dt 202.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23219 | train 3.056170 | val 3.302221 | tok/s 2628176.21 | norm 0.3576 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23220 | train 2.935718 | val 3.302221 | tok/s 2623472.46 | norm 0.3702 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23221 | train 2.894296 | val 3.302221 | tok/s 2626689.37 | norm 0.3435 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23222 | train 3.070141 | val 3.302221 | tok/s 2618892.43 | norm 0.3328 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23223 | train 2.834327 | val 3.302221 | tok/s 2620853.44 | norm 0.3465 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23224 | train 2.886629 | val 3.302221 | tok/s 2629575.63 | norm 0.3322 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23225 | train 2.971645 | val 3.302221 | tok/s 2622162.71 | norm 0.3513 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23226 | train 2.978516 | val 3.302221 | tok/s 2626374.01 | norm 0.3705 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23227 | train 3.026159 | val 3.302221 | tok/s 2624658.07 | norm 0.3813 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23228 | train 2.842156 | val 3.302221 | tok/s 2622494.50 | norm 0.3892 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23229 | train 2.950610 | val 3.302221 | tok/s 2629071.42 | norm 0.3877 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23230 | train 2.869997 | val 3.302221 | tok/s 2627783.81 | norm 0.3430 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23231 | train 2.961210 | val 3.302221 | tok/s 2630408.13 | norm 0.3737 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23232 | train 2.920825 | val 3.302221 | tok/s 2625740.90 | norm 0.3708 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23233 | train 3.025901 | val 3.302221 | tok/s 2625419.99 | norm 0.3445 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23234 | train 2.871820 | val 3.302221 | tok/s 2628705.48 | norm 0.3431 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23235 | train 2.961402 | val 3.302221 | tok/s 2625820.58 | norm 0.3535 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23236 | train 2.935779 | val 3.302221 | tok/s 2629947.21 | norm 0.3556 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23237 | train 2.833999 | val 3.302221 | tok/s 2626046.10 | norm 0.3752 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23238 | train 2.904553 | val 3.302221 | tok/s 2626141.84 | norm 0.3815 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23239 | train 2.855853 | val 3.302221 | tok/s 2631595.89 | norm 0.3755 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23240 | train 2.876724 | val 3.302221 | tok/s 2628825.95 | norm 0.3696 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23241 | train 2.906558 | val 3.302221 | tok/s 2629443.01 | norm 0.3935 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23242 | train 2.929866 | val 3.302221 | tok/s 2618704.79 | norm 0.3830 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23243 | train 3.125525 | val 3.302221 | tok/s 2561881.17 | norm 0.4309 | dt 204.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23244 | train 2.988497 | val 3.302221 | tok/s 2573848.54 | norm 0.3628 | dt 203.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23245 | train 2.923924 | val 3.302221 | tok/s 2626430.14 | norm 0.3406 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23246 | train 2.911213 | val 3.302221 | tok/s 2630957.81 | norm 0.3276 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23247 | train 3.048550 | val 3.302221 | tok/s 2625681.32 | norm 0.3318 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23248 | train 2.824694 | val 3.302221 | tok/s 2626194.16 | norm 0.3326 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23249 | train 2.989408 | val 3.302221 | tok/s 2631094.92 | norm 0.3239 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 23250 | val_loss 3.305943 step 23250 | train 2.986238 | val 3.305943 | tok/s 2579283.28 | norm 0.3466 | dt 203.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23251 | train 2.828399 | val 3.305943 | tok/s 2625745.58 | norm 0.3655 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23252 | train 2.835401 | val 3.305943 | tok/s 2628829.89 | norm 0.3626 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23253 | train 2.828250 | val 3.305943 | tok/s 2631041.66 | norm 0.3686 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23254 | train 3.002811 | val 3.305943 | tok/s 2628922.91 | norm 0.3797 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23255 | train 2.866117 | val 3.305943 | tok/s 2629955.52 | norm 0.3795 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23256 | train 3.041568 | val 3.305943 | tok/s 2627543.36 | norm 0.4293 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23257 | train 2.885904 | val 3.305943 | tok/s 2629446.67 | norm 0.3971 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23258 | train 2.913219 | val 3.305943 | tok/s 2525973.23 | norm 0.3612 | dt 207.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23259 | train 2.950169 | val 3.305943 | tok/s 2549622.82 | norm 0.3772 | dt 205.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23260 | train 2.920007 | val 3.305943 | tok/s 2627725.91 | norm 0.3371 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23261 | train 2.906853 | val 3.305943 | tok/s 2620649.78 | norm 0.3482 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23262 | train 2.844216 | val 3.305943 | tok/s 2631724.50 | norm 0.3716 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23263 | train 3.044142 | val 3.305943 | tok/s 2625344.14 | norm 0.4065 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23264 | train 2.877613 | val 3.305943 | tok/s 2630359.92 | norm 0.3924 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23265 | train 2.873144 | val 3.305943 | tok/s 2618859.56 | norm 0.3671 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23266 | train 2.894460 | val 3.305943 | tok/s 2626606.53 | norm 0.3693 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23267 | train 2.887023 | val 3.305943 | tok/s 2632021.87 | norm 0.3728 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23268 | train 2.926127 | val 3.305943 | tok/s 2627244.24 | norm 0.3513 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23269 | train 2.932541 | val 3.305943 | tok/s 2630311.88 | norm 0.3483 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23270 | train 2.992785 | val 3.305943 | tok/s 2625271.26 | norm 0.3580 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23271 | train 3.113364 | val 3.305943 | tok/s 2627929.88 | norm 0.3917 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23272 | train 2.982228 | val 3.305943 | tok/s 2629921.84 | norm 0.3757 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23273 | train 2.920281 | val 3.305943 | tok/s 2628019.93 | norm 0.3169 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23274 | train 2.891392 | val 3.305943 | tok/s 2629711.31 | norm 0.3537 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23275 | train 2.943175 | val 3.305943 | tok/s 2615208.12 | norm 0.3734 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23276 | train 2.899146 | val 3.305943 | tok/s 2622944.77 | norm 0.3598 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23277 | train 3.007192 | val 3.305943 | tok/s 2630933.21 | norm 0.3476 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23278 | train 2.973397 | val 3.305943 | tok/s 2630764.30 | norm 0.3466 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23279 | train 2.878537 | val 3.305943 | tok/s 2630638.24 | norm 0.3410 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23280 | train 2.823887 | val 3.305943 | tok/s 2623171.80 | norm 0.3384 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23281 | train 2.879969 | val 3.305943 | tok/s 2622736.24 | norm 0.3373 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23282 | train 2.963658 | val 3.305943 | tok/s 2630552.62 | norm 0.3442 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23283 | train 2.889311 | val 3.305943 | tok/s 2624595.83 | norm 0.3343 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23284 | train 2.977786 | val 3.305943 | tok/s 2632001.72 | norm 0.3603 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23285 | train 2.871153 | val 3.305943 | tok/s 2624835.83 | norm 0.3711 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23286 | train 2.988336 | val 3.305943 | tok/s 2623483.97 | norm 0.3577 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23287 | train 2.893368 | val 3.305943 | tok/s 2629516.86 | norm 0.3772 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23288 | train 2.973791 | val 3.305943 | tok/s 2621012.85 | norm 0.3880 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23289 | train 2.911134 | val 3.305943 | tok/s 2624689.10 | norm 0.3235 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23290 | train 2.959029 | val 3.305943 | tok/s 2624511.10 | norm 0.3191 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23291 | train 2.977248 | val 3.305943 | tok/s 2624106.30 | norm 0.3680 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23292 | train 2.985017 | val 3.305943 | tok/s 2624890.93 | norm 0.3464 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23293 | train 2.998025 | val 3.305943 | tok/s 2623535.18 | norm 0.3466 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23294 | train 2.870574 | val 3.305943 | tok/s 2628252.21 | norm 0.3243 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23295 | train 2.821045 | val 3.305943 | tok/s 2626781.47 | norm 0.3439 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23296 | train 2.980815 | val 3.305943 | tok/s 2625910.57 | norm 0.4117 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23297 | train 2.883232 | val 3.305943 | tok/s 2631206.02 | norm 0.3913 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23298 | train 2.946396 | val 3.305943 | tok/s 2428381.17 | norm 0.3796 | dt 215.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23299 | train 2.863153 | val 3.305943 | tok/s 2610156.63 | norm 0.4279 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23300 | train 2.995064 | val 3.305943 | tok/s 2624507.34 | norm 0.4126 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23301 | train 2.978328 | val 3.305943 | tok/s 2628811.54 | norm 0.3624 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23302 | train 3.043270 | val 3.305943 | tok/s 2519241.79 | norm 0.3477 | dt 208.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23303 | train 2.952465 | val 3.305943 | tok/s 2622313.99 | norm 0.4016 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23304 | train 3.093838 | val 3.305943 | tok/s 2630594.86 | norm 0.4861 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23305 | train 2.911829 | val 3.305943 | tok/s 2627567.95 | norm 0.4093 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23306 | train 2.977333 | val 3.305943 | tok/s 2629674.95 | norm 0.3288 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23307 | train 2.956465 | val 3.305943 | tok/s 2578898.12 | norm 0.3399 | dt 203.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23308 | train 2.933012 | val 3.305943 | tok/s 2618718.25 | norm 0.3890 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23309 | train 2.935188 | val 3.305943 | tok/s 2631198.36 | norm 0.3771 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23310 | train 2.945031 | val 3.305943 | tok/s 2628222.44 | norm 0.3417 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23311 | train 2.906518 | val 3.305943 | tok/s 2630580.24 | norm 0.3576 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23312 | train 3.046674 | val 3.305943 | tok/s 2618447.01 | norm 0.4020 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23313 | train 2.978102 | val 3.305943 | tok/s 2623834.68 | norm 0.3522 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23314 | train 2.864651 | val 3.305943 | tok/s 2631085.81 | norm 0.3247 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23315 | train 2.960872 | val 3.305943 | tok/s 2627127.88 | norm 0.3919 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23316 | train 2.878281 | val 3.305943 | tok/s 2630562.38 | norm 0.3665 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23317 | train 3.005654 | val 3.305943 | tok/s 2625725.79 | norm 0.3494 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23318 | train 2.935751 | val 3.305943 | tok/s 2624908.87 | norm 0.3717 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23319 | train 2.947003 | val 3.305943 | tok/s 2630552.13 | norm 0.3499 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23320 | train 2.997554 | val 3.305943 | tok/s 2630416.90 | norm 0.3755 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23321 | train 2.933374 | val 3.305943 | tok/s 2625655.03 | norm 0.3721 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23322 | train 2.868838 | val 3.305943 | tok/s 2620659.65 | norm 0.3327 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23323 | train 2.805531 | val 3.305943 | tok/s 2622615.29 | norm 0.3523 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23324 | train 3.112394 | val 3.305943 | tok/s 2633576.63 | norm 0.3746 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23325 | train 2.973598 | val 3.305943 | tok/s 2625436.36 | norm 0.3939 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23326 | train 2.951501 | val 3.305943 | tok/s 2626873.86 | norm 0.3578 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23327 | train 2.940104 | val 3.305943 | tok/s 2621227.21 | norm 0.3496 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23328 | train 2.837141 | val 3.305943 | tok/s 2624988.22 | norm 0.3639 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23329 | train 2.983888 | val 3.305943 | tok/s 2630685.99 | norm 0.3594 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23330 | train 2.867341 | val 3.305943 | tok/s 2627419.26 | norm 0.3618 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23331 | train 3.052632 | val 3.305943 | tok/s 2631888.57 | norm 0.4022 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23332 | train 3.044289 | val 3.305943 | tok/s 2620056.73 | norm 0.3898 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23333 | train 2.912718 | val 3.305943 | tok/s 2622629.20 | norm 0.3868 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23334 | train 2.888502 | val 3.305943 | tok/s 2631277.33 | norm 0.3842 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23335 | train 3.064747 | val 3.305943 | tok/s 2624524.10 | norm 0.3748 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23336 | train 2.975916 | val 3.305943 | tok/s 2629090.44 | norm 0.3141 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23337 | train 2.977545 | val 3.305943 | tok/s 2624498.43 | norm 0.3342 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23338 | train 2.978458 | val 3.305943 | tok/s 2625538.98 | norm 0.3767 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23339 | train 2.895118 | val 3.305943 | tok/s 2630579.37 | norm 0.3798 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23340 | train 2.970048 | val 3.305943 | tok/s 2630668.00 | norm 0.3580 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23341 | train 2.897090 | val 3.305943 | tok/s 2628944.35 | norm 0.3612 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23342 | train 2.775464 | val 3.305943 | tok/s 2624607.97 | norm 0.3995 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23343 | train 2.989155 | val 3.305943 | tok/s 2626710.22 | norm 0.3563 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23344 | train 2.875172 | val 3.305943 | tok/s 2559129.96 | norm 0.3404 | dt 204.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23345 | train 2.983230 | val 3.305943 | tok/s 2464117.83 | norm 0.3686 | dt 212.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23346 | train 3.038634 | val 3.305943 | tok/s 2629074.40 | norm 0.3504 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23347 | train 2.906251 | val 3.305943 | tok/s 2627117.33 | norm 0.3179 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23348 | train 3.136001 | val 3.305943 | tok/s 2630100.78 | norm 0.3811 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23349 | train 2.956521 | val 3.305943 | tok/s 2629687.77 | norm 0.3498 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23350 | train 2.972967 | val 3.305943 | tok/s 2618624.83 | norm 0.3606 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23351 | train 3.003009 | val 3.305943 | tok/s 2633365.00 | norm 0.4189 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23352 | train 3.034340 | val 3.305943 | tok/s 2524662.10 | norm 0.3737 | dt 207.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23353 | train 2.885001 | val 3.305943 | tok/s 2631718.69 | norm 0.3390 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23354 | train 2.955268 | val 3.305943 | tok/s 2626002.71 | norm 0.3540 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23355 | train 2.973036 | val 3.305943 | tok/s 2625895.67 | norm 0.4070 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23356 | train 2.925766 | val 3.305943 | tok/s 2632363.43 | norm 0.3951 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23357 | train 2.963933 | val 3.305943 | tok/s 2626053.67 | norm 0.4064 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23358 | train 2.964488 | val 3.305943 | tok/s 2624783.19 | norm 0.3938 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23359 | train 2.900352 | val 3.305943 | tok/s 2620708.52 | norm 0.3561 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23360 | train 2.915532 | val 3.305943 | tok/s 2625812.74 | norm 0.3735 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23361 | train 2.921376 | val 3.305943 | tok/s 2627614.70 | norm 0.3650 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23362 | train 2.883355 | val 3.305943 | tok/s 2624788.32 | norm 0.3787 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23363 | train 3.228461 | val 3.305943 | tok/s 2628819.85 | norm 0.4096 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23364 | train 2.901378 | val 3.305943 | tok/s 2619257.49 | norm 0.3514 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23365 | train 2.964558 | val 3.305943 | tok/s 2620659.88 | norm 0.3240 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23366 | train 2.973867 | val 3.305943 | tok/s 2626270.71 | norm 0.3395 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23367 | train 2.899779 | val 3.305943 | tok/s 2625280.83 | norm 0.3759 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23368 | train 2.924532 | val 3.305943 | tok/s 2625442.19 | norm 0.3859 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23369 | train 3.005947 | val 3.305943 | tok/s 2620621.83 | norm 0.3751 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23370 | train 3.027911 | val 3.305943 | tok/s 2623397.87 | norm 0.3543 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23371 | train 2.872497 | val 3.305943 | tok/s 2629996.32 | norm 0.3717 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23372 | train 2.870959 | val 3.305943 | tok/s 2627522.11 | norm 0.3755 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23373 | train 3.037273 | val 3.305943 | tok/s 2630064.86 | norm 0.3979 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23374 | train 2.981539 | val 3.305943 | tok/s 2621295.33 | norm 0.3920 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23375 | train 2.928812 | val 3.305943 | tok/s 2604567.71 | norm 0.3823 | dt 201.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23376 | train 2.878669 | val 3.305943 | tok/s 2627567.62 | norm 0.3498 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23377 | train 2.907538 | val 3.305943 | tok/s 2627848.13 | norm 0.3985 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23378 | train 2.961944 | val 3.305943 | tok/s 2627315.25 | norm 0.3843 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23379 | train 3.012610 | val 3.305943 | tok/s 2623966.07 | norm 0.3294 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23380 | train 2.841034 | val 3.305943 | tok/s 2626001.73 | norm 0.3912 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23381 | train 2.949003 | val 3.305943 | tok/s 2627957.87 | norm 0.4055 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23382 | train 2.922864 | val 3.305943 | tok/s 2627136.60 | norm 0.3693 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23383 | train 2.975714 | val 3.305943 | tok/s 2626596.23 | norm 0.3648 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23384 | train 2.916242 | val 3.305943 | tok/s 2577795.63 | norm 0.3909 | dt 203.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23385 | train 2.811216 | val 3.305943 | tok/s 2628775.55 | norm 0.3628 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23386 | train 2.938228 | val 3.305943 | tok/s 2633283.46 | norm 0.3471 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23387 | train 2.889116 | val 3.305943 | tok/s 2628694.51 | norm 0.3526 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23388 | train 3.169449 | val 3.305943 | tok/s 2627172.36 | norm 0.4993 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23389 | train 2.965401 | val 3.305943 | tok/s 2627898.88 | norm 0.3943 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23390 | train 2.943914 | val 3.305943 | tok/s 2518342.24 | norm 0.3231 | dt 208.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23391 | train 3.088686 | val 3.305943 | tok/s 2623978.67 | norm 0.4292 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23392 | train 2.896714 | val 3.305943 | tok/s 2620294.89 | norm 0.4164 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23393 | train 3.030400 | val 3.305943 | tok/s 2630408.67 | norm 0.3701 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23394 | train 2.924258 | val 3.305943 | tok/s 2628181.86 | norm 0.3611 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23395 | train 2.795483 | val 3.305943 | tok/s 2629967.85 | norm 0.3510 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23396 | train 3.026149 | val 3.305943 | tok/s 2632949.60 | norm 0.3357 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23397 | train 2.943807 | val 3.305943 | tok/s 2523269.43 | norm 0.3374 | dt 207.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23398 | train 2.966381 | val 3.305943 | tok/s 2570334.73 | norm 0.3658 | dt 203.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23399 | train 2.972644 | val 3.305943 | tok/s 2627460.12 | norm 0.3871 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23400 | train 2.859757 | val 3.305943 | tok/s 2626659.82 | norm 0.3534 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23401 | train 2.829120 | val 3.305943 | tok/s 2620540.94 | norm 0.3644 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23402 | train 2.945962 | val 3.305943 | tok/s 2625732.26 | norm 0.3597 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23403 | train 2.926116 | val 3.305943 | tok/s 2630426.40 | norm 0.3342 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23404 | train 3.193501 | val 3.305943 | tok/s 2632426.88 | norm 0.3993 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23405 | train 2.868285 | val 3.305943 | tok/s 2631782.83 | norm 0.3990 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23406 | train 3.062481 | val 3.305943 | tok/s 2628243.51 | norm 0.3384 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23407 | train 2.908063 | val 3.305943 | tok/s 2621395.92 | norm 0.3790 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23408 | train 2.826166 | val 3.305943 | tok/s 2628812.37 | norm 0.3552 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23409 | train 2.826761 | val 3.305943 | tok/s 2622618.53 | norm 0.3880 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23410 | train 2.941990 | val 3.305943 | tok/s 2624956.17 | norm 0.3531 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23411 | train 2.832212 | val 3.305943 | tok/s 2620445.85 | norm 0.3645 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23412 | train 2.970074 | val 3.305943 | tok/s 2626029.21 | norm 0.3699 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23413 | train 2.958419 | val 3.305943 | tok/s 2629444.67 | norm 0.3507 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23414 | train 2.913178 | val 3.305943 | tok/s 2625309.10 | norm 0.3517 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23415 | train 2.903226 | val 3.305943 | tok/s 2629134.60 | norm 0.3950 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23416 | train 2.836703 | val 3.305943 | tok/s 2623779.91 | norm 0.3650 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23417 | train 2.913988 | val 3.305943 | tok/s 2624387.80 | norm 0.3777 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23418 | train 2.959323 | val 3.305943 | tok/s 2626065.31 | norm 0.3926 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23419 | train 2.922823 | val 3.305943 | tok/s 2627059.73 | norm 0.3762 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23420 | train 3.072766 | val 3.305943 | tok/s 2629531.00 | norm 0.3933 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23421 | train 2.888700 | val 3.305943 | tok/s 2627864.78 | norm 0.3573 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23422 | train 2.976136 | val 3.305943 | tok/s 2609140.77 | norm 0.3585 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23423 | train 3.062156 | val 3.305943 | tok/s 2616842.97 | norm 0.4445 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23424 | train 2.987949 | val 3.305943 | tok/s 2630543.01 | norm 0.3825 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23425 | train 2.895130 | val 3.305943 | tok/s 2631926.56 | norm 0.3217 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23426 | train 2.800525 | val 3.305943 | tok/s 2622896.56 | norm 0.3444 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23427 | train 2.900315 | val 3.305943 | tok/s 2627570.24 | norm 0.3623 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23428 | train 2.965445 | val 3.305943 | tok/s 2631448.05 | norm 0.3821 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23429 | train 2.960455 | val 3.305943 | tok/s 2628609.52 | norm 0.3999 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23430 | train 3.049226 | val 3.305943 | tok/s 2632035.36 | norm 0.3709 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23431 | train 2.948844 | val 3.305943 | tok/s 2592194.35 | norm 0.3647 | dt 202.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23432 | train 2.942109 | val 3.305943 | tok/s 2625215.11 | norm 0.4034 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23433 | train 2.932737 | val 3.305943 | tok/s 2631100.00 | norm 0.6955 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23434 | train 3.129688 | val 3.305943 | tok/s 2629110.38 | norm 0.6006 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23435 | train 2.891379 | val 3.305943 | tok/s 2628324.53 | norm 0.3618 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23436 | train 3.016924 | val 3.305943 | tok/s 2630483.34 | norm 0.2939 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23437 | train 2.908050 | val 3.305943 | tok/s 2628504.67 | norm 0.2956 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23438 | train 2.881818 | val 3.305943 | tok/s 2630154.03 | norm 0.3443 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23439 | train 2.971449 | val 3.305943 | tok/s 2622801.56 | norm 0.3698 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23440 | train 2.957146 | val 3.305943 | tok/s 2629579.30 | norm 0.3413 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23441 | train 2.865176 | val 3.305943 | tok/s 2625118.12 | norm 0.3412 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23442 | train 2.948126 | val 3.305943 | tok/s 2628120.70 | norm 0.3608 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23443 | train 2.945965 | val 3.305943 | tok/s 2512302.67 | norm 0.3283 | dt 208.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23444 | train 2.942736 | val 3.305943 | tok/s 2623218.27 | norm 0.3044 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23445 | train 2.878958 | val 3.305943 | tok/s 2631555.94 | norm 0.3178 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23446 | train 2.970699 | val 3.305943 | tok/s 2627131.88 | norm 0.3568 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23447 | train 2.970008 | val 3.305943 | tok/s 2622922.03 | norm 0.3631 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23448 | train 3.035804 | val 3.305943 | tok/s 2626981.84 | norm 0.3803 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23449 | train 2.907985 | val 3.305943 | tok/s 2621550.33 | norm 0.3711 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23450 | train 2.900218 | val 3.305943 | tok/s 2628294.84 | norm 0.3577 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23451 | train 2.936739 | val 3.305943 | tok/s 2630125.69 | norm 0.3557 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23452 | train 3.019822 | val 3.305943 | tok/s 2630275.05 | norm 0.3436 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23453 | train 2.989917 | val 3.305943 | tok/s 2620447.83 | norm 0.3780 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23454 | train 2.950247 | val 3.305943 | tok/s 2604973.29 | norm 0.3528 | dt 201.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23455 | train 2.891548 | val 3.305943 | tok/s 2631704.60 | norm 0.3566 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23456 | train 2.956482 | val 3.305943 | tok/s 2618465.62 | norm 0.3876 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23457 | train 2.976060 | val 3.305943 | tok/s 2621096.11 | norm 0.4033 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23458 | train 2.906532 | val 3.305943 | tok/s 2618544.45 | norm 0.3861 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23459 | train 2.958904 | val 3.305943 | tok/s 2625328.93 | norm 0.3726 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23460 | train 2.937317 | val 3.305943 | tok/s 2625739.53 | norm 0.4045 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23461 | train 2.956564 | val 3.305943 | tok/s 2617532.29 | norm 0.4050 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23462 | train 2.978397 | val 3.305943 | tok/s 2625478.21 | norm 0.3613 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23463 | train 2.928582 | val 3.305943 | tok/s 2620739.62 | norm 0.3627 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23464 | train 2.851516 | val 3.305943 | tok/s 2625363.82 | norm 0.4216 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23465 | train 2.898191 | val 3.305943 | tok/s 2630463.57 | norm 0.3446 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23466 | train 2.894113 | val 3.305943 | tok/s 2630381.38 | norm 0.3288 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23467 | train 2.985616 | val 3.305943 | tok/s 2628264.47 | norm 0.3644 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23468 | train 2.918977 | val 3.305943 | tok/s 2624943.99 | norm 0.3479 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23469 | train 2.800944 | val 3.305943 | tok/s 2622222.45 | norm 0.3507 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23470 | train 2.927593 | val 3.305943 | tok/s 2631060.10 | norm 0.3469 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23471 | train 2.903294 | val 3.305943 | tok/s 2628460.71 | norm 0.3420 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23472 | train 2.967701 | val 3.305943 | tok/s 2625606.83 | norm 0.3752 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23473 | train 2.957106 | val 3.305943 | tok/s 2622942.16 | norm 0.4078 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23474 | train 2.973837 | val 3.305943 | tok/s 2625359.11 | norm 0.4056 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23475 | train 2.994907 | val 3.305943 | tok/s 2631890.26 | norm 0.4038 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23476 | train 2.868046 | val 3.305943 | tok/s 2630619.89 | norm 0.3603 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23477 | train 2.922330 | val 3.305943 | tok/s 2628101.82 | norm 0.3253 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23478 | train 2.961330 | val 3.305943 | tok/s 2625046.26 | norm 0.3357 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23479 | train 2.934393 | val 3.305943 | tok/s 2626974.30 | norm 0.3215 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23480 | train 2.911187 | val 3.305943 | tok/s 2631753.94 | norm 0.3293 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23481 | train 2.883772 | val 3.305943 | tok/s 2628797.44 | norm 0.3342 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23482 | train 2.873098 | val 3.305943 | tok/s 2625798.96 | norm 0.3484 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23483 | train 2.913169 | val 3.305943 | tok/s 2621468.64 | norm 0.3803 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23484 | train 2.866868 | val 3.305943 | tok/s 2623532.03 | norm 0.3669 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23485 | train 2.885925 | val 3.305943 | tok/s 2630443.48 | norm 0.3520 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23486 | train 2.923465 | val 3.305943 | tok/s 2627904.09 | norm 0.3364 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23487 | train 2.875710 | val 3.305943 | tok/s 2587897.13 | norm 0.3167 | dt 202.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23488 | train 2.966017 | val 3.305943 | tok/s 2629888.81 | norm 0.3526 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23489 | train 2.958881 | val 3.305943 | tok/s 2626371.62 | norm 0.3489 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23490 | train 2.769315 | val 3.305943 | tok/s 2628326.56 | norm 0.3632 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23491 | train 2.891407 | val 3.305943 | tok/s 2528379.49 | norm 0.3776 | dt 207.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23492 | train 2.918963 | val 3.305943 | tok/s 2501435.02 | norm 0.3328 | dt 209.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23493 | train 2.918046 | val 3.305943 | tok/s 2624926.76 | norm 0.3371 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23494 | train 2.974737 | val 3.305943 | tok/s 2632634.25 | norm 0.3851 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23495 | train 3.018637 | val 3.305943 | tok/s 2629316.13 | norm 0.3846 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23496 | train 2.967403 | val 3.305943 | tok/s 2624568.21 | norm 0.3711 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23497 | train 3.057393 | val 3.305943 | tok/s 2634007.76 | norm 0.3649 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23498 | train 3.059148 | val 3.305943 | tok/s 2627699.01 | norm 0.3757 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23499 | train 2.865217 | val 3.305943 | tok/s 2625051.55 | norm 0.3769 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 23500 | val_loss 3.300309 step 23500 | train 2.839293 | val 3.300309 | tok/s 2617863.68 | norm 0.3558 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23501 | train 3.037612 | val 3.300309 | tok/s 2626755.69 | norm 0.3971 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23502 | train 2.875074 | val 3.300309 | tok/s 2621189.30 | norm 0.3507 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23503 | train 2.907146 | val 3.300309 | tok/s 2631841.08 | norm 0.3972 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23504 | train 3.244157 | val 3.300309 | tok/s 2628835.94 | norm 0.4626 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23505 | train 2.991121 | val 3.300309 | tok/s 2627652.87 | norm 0.3431 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23506 | train 2.922234 | val 3.300309 | tok/s 2548729.04 | norm 0.4373 | dt 205.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23507 | train 2.905456 | val 3.300309 | tok/s 2622633.99 | norm 0.3558 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23508 | train 2.878111 | val 3.300309 | tok/s 2631421.27 | norm 0.3543 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23509 | train 2.883029 | val 3.300309 | tok/s 2626377.12 | norm 0.3375 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23510 | train 2.947871 | val 3.300309 | tok/s 2629990.68 | norm 0.3243 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23511 | train 2.965944 | val 3.300309 | tok/s 2620104.98 | norm 0.3584 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23512 | train 2.937980 | val 3.300309 | tok/s 2605376.06 | norm 0.3409 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23513 | train 2.962278 | val 3.300309 | tok/s 2613722.49 | norm 0.3528 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23514 | train 2.987161 | val 3.300309 | tok/s 2627685.17 | norm 0.4046 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23515 | train 3.051178 | val 3.300309 | tok/s 2627726.03 | norm 0.4019 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23516 | train 2.894394 | val 3.300309 | tok/s 2623190.85 | norm 0.3987 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23517 | train 2.890445 | val 3.300309 | tok/s 2623356.45 | norm 0.4247 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23518 | train 2.945020 | val 3.300309 | tok/s 2631347.09 | norm 0.3816 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23519 | train 2.881565 | val 3.300309 | tok/s 2625595.62 | norm 0.3686 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23520 | train 2.880189 | val 3.300309 | tok/s 2611406.00 | norm 0.3725 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23521 | train 2.952851 | val 3.300309 | tok/s 2623984.82 | norm 0.3142 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23522 | train 2.894952 | val 3.300309 | tok/s 2627082.97 | norm 0.3159 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23523 | train 2.886873 | val 3.300309 | tok/s 2629911.76 | norm 0.3532 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23524 | train 2.985353 | val 3.300309 | tok/s 2632830.16 | norm 0.3780 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23525 | train 2.989825 | val 3.300309 | tok/s 2631506.49 | norm 0.3566 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23526 | train 2.848038 | val 3.300309 | tok/s 2625629.24 | norm 0.3271 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23527 | train 2.973144 | val 3.300309 | tok/s 2625532.93 | norm 0.4173 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23528 | train 2.965966 | val 3.300309 | tok/s 2632195.40 | norm 0.3191 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23529 | train 2.925669 | val 3.300309 | tok/s 2628746.32 | norm 0.3407 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23530 | train 2.992298 | val 3.300309 | tok/s 2628766.60 | norm 0.6465 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23531 | train 2.954837 | val 3.300309 | tok/s 2623794.75 | norm 0.3151 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23532 | train 2.875273 | val 3.300309 | tok/s 2625596.75 | norm 0.4298 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23533 | train 3.256611 | val 3.300309 | tok/s 2630622.62 | norm 0.4462 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23534 | train 2.974308 | val 3.300309 | tok/s 2627105.14 | norm 0.3653 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23535 | train 2.978182 | val 3.300309 | tok/s 2629462.64 | norm 0.3522 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23536 | train 2.963631 | val 3.300309 | tok/s 2627214.18 | norm 0.3772 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23537 | train 2.867992 | val 3.300309 | tok/s 2628352.37 | norm 0.3525 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23538 | train 2.828177 | val 3.300309 | tok/s 2632144.14 | norm 0.3557 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23539 | train 2.997878 | val 3.300309 | tok/s 2630259.43 | norm 0.4766 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23540 | train 2.925921 | val 3.300309 | tok/s 2630150.97 | norm 0.3269 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23541 | train 2.896344 | val 3.300309 | tok/s 2624428.19 | norm 0.3345 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23542 | train 3.223197 | val 3.300309 | tok/s 2618746.42 | norm 0.4065 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23543 | train 3.005680 | val 3.300309 | tok/s 2630540.42 | norm 0.3542 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23544 | train 2.859223 | val 3.300309 | tok/s 2627514.14 | norm 0.3327 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23545 | train 2.999746 | val 3.300309 | tok/s 2630485.34 | norm 0.3955 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23546 | train 2.871051 | val 3.300309 | tok/s 2622505.41 | norm 0.3655 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23547 | train 3.021935 | val 3.300309 | tok/s 2628297.06 | norm 0.3794 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23548 | train 3.064202 | val 3.300309 | tok/s 2624344.80 | norm 0.3851 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23549 | train 3.024280 | val 3.300309 | tok/s 2624764.36 | norm 0.3750 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23550 | train 2.804374 | val 3.300309 | tok/s 2626093.11 | norm 0.4140 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23551 | train 2.969838 | val 3.300309 | tok/s 2630049.60 | norm 0.3820 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23552 | train 2.758793 | val 3.300309 | tok/s 2630970.51 | norm 0.3588 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23553 | train 2.832251 | val 3.300309 | tok/s 2629488.14 | norm 0.3803 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23554 | train 2.950977 | val 3.300309 | tok/s 2627965.59 | norm 0.3775 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23555 | train 2.869953 | val 3.300309 | tok/s 2628655.12 | norm 0.3715 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23556 | train 2.950373 | val 3.300309 | tok/s 2630541.85 | norm 0.3507 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23557 | train 2.935068 | val 3.300309 | tok/s 2626580.65 | norm 0.3726 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23558 | train 2.894688 | val 3.300309 | tok/s 2546410.04 | norm 0.3885 | dt 205.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23559 | train 2.973202 | val 3.300309 | tok/s 2624110.48 | norm 0.3865 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23560 | train 2.950249 | val 3.300309 | tok/s 2626185.89 | norm 0.3873 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23561 | train 3.091568 | val 3.300309 | tok/s 2627432.75 | norm 0.4067 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23562 | train 2.946572 | val 3.300309 | tok/s 2627328.46 | norm 0.3512 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23563 | train 2.938231 | val 3.300309 | tok/s 2617728.88 | norm 0.3606 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23564 | train 2.954391 | val 3.300309 | tok/s 2622477.62 | norm 0.3812 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23565 | train 3.139902 | val 3.300309 | tok/s 2631217.59 | norm 0.3716 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23566 | train 3.131471 | val 3.300309 | tok/s 2591071.82 | norm 0.4152 | dt 202.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23567 | train 2.950224 | val 3.300309 | tok/s 2629148.60 | norm 0.4232 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23568 | train 2.950353 | val 3.300309 | tok/s 2602499.59 | norm 0.3630 | dt 201.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23569 | train 2.874306 | val 3.300309 | tok/s 2612041.68 | norm 0.3252 | dt 200.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23570 | train 2.911026 | val 3.300309 | tok/s 2630671.84 | norm 0.3723 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23571 | train 2.923048 | val 3.300309 | tok/s 2626300.51 | norm 0.3889 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23572 | train 2.945997 | val 3.300309 | tok/s 2628838.72 | norm 0.3565 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23573 | train 2.914819 | val 3.300309 | tok/s 2623243.10 | norm 0.3597 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23574 | train 2.944142 | val 3.300309 | tok/s 2625961.14 | norm 0.3397 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23575 | train 2.894869 | val 3.300309 | tok/s 2629761.96 | norm 0.3550 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23576 | train 2.977233 | val 3.300309 | tok/s 2628080.23 | norm 0.3498 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23577 | train 2.972745 | val 3.300309 | tok/s 2630390.72 | norm 0.3391 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23578 | train 2.917965 | val 3.300309 | tok/s 2628069.21 | norm 0.3610 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23579 | train 2.953416 | val 3.300309 | tok/s 2620867.38 | norm 0.3947 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23580 | train 2.983466 | val 3.300309 | tok/s 2630229.93 | norm 0.3598 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23581 | train 2.912269 | val 3.300309 | tok/s 2627304.58 | norm 0.3449 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23582 | train 3.061194 | val 3.300309 | tok/s 2628813.26 | norm 0.3867 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23583 | train 2.901034 | val 3.300309 | tok/s 2623117.65 | norm 0.3702 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23584 | train 2.975943 | val 3.300309 | tok/s 2627409.37 | norm 0.3416 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23585 | train 2.935579 | val 3.300309 | tok/s 2628343.86 | norm 0.3703 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23586 | train 2.993247 | val 3.300309 | tok/s 2625727.91 | norm 0.3762 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23587 | train 2.956918 | val 3.300309 | tok/s 2630243.23 | norm 0.3478 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23588 | train 2.952172 | val 3.300309 | tok/s 2624116.87 | norm 0.3468 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23589 | train 2.927726 | val 3.300309 | tok/s 2622285.21 | norm 0.3508 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23590 | train 3.042556 | val 3.300309 | tok/s 2624806.13 | norm 0.3816 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23591 | train 2.886976 | val 3.300309 | tok/s 2624706.37 | norm 0.3611 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23592 | train 2.978966 | val 3.300309 | tok/s 2630555.33 | norm 0.3463 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23593 | train 2.901962 | val 3.300309 | tok/s 2627161.04 | norm 0.3794 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23594 | train 3.010199 | val 3.300309 | tok/s 2628749.21 | norm 0.3540 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23595 | train 2.856316 | val 3.300309 | tok/s 2629078.17 | norm 0.3367 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23596 | train 2.923598 | val 3.300309 | tok/s 2626720.52 | norm 0.4012 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23597 | train 2.944525 | val 3.300309 | tok/s 2629461.73 | norm 0.3655 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23598 | train 2.979316 | val 3.300309 | tok/s 2626164.14 | norm 0.3409 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23599 | train 2.951064 | val 3.300309 | tok/s 2626635.95 | norm 0.3528 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23600 | train 3.018977 | val 3.300309 | tok/s 2623151.34 | norm 0.3531 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23601 | train 2.864868 | val 3.300309 | tok/s 2622291.71 | norm 0.3692 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23602 | train 2.857289 | val 3.300309 | tok/s 2630972.65 | norm 0.3606 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23603 | train 2.908648 | val 3.300309 | tok/s 2625904.31 | norm 0.3662 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23604 | train 2.918920 | val 3.300309 | tok/s 2626186.90 | norm 0.3998 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23605 | train 2.846683 | val 3.300309 | tok/s 2623930.98 | norm 0.3777 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23606 | train 2.941306 | val 3.300309 | tok/s 2625024.51 | norm 0.3538 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23607 | train 2.958273 | val 3.300309 | tok/s 2633336.84 | norm 0.3600 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23608 | train 2.935499 | val 3.300309 | tok/s 2626922.55 | norm 0.3774 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23609 | train 3.062202 | val 3.300309 | tok/s 2623836.91 | norm 0.3855 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23610 | train 2.940266 | val 3.300309 | tok/s 2633029.97 | norm 0.3774 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23611 | train 2.924654 | val 3.300309 | tok/s 2624324.36 | norm 0.3698 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23612 | train 2.898405 | val 3.300309 | tok/s 2629200.90 | norm 0.3707 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23613 | train 2.831379 | val 3.300309 | tok/s 2626677.13 | norm 0.3723 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23614 | train 2.988971 | val 3.300309 | tok/s 2629136.78 | norm 0.3825 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23615 | train 2.901342 | val 3.300309 | tok/s 2627082.65 | norm 0.3872 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23616 | train 2.881860 | val 3.300309 | tok/s 2601505.30 | norm 0.3686 | dt 201.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23617 | train 2.821818 | val 3.300309 | tok/s 2631992.41 | norm 0.3760 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23618 | train 3.005217 | val 3.300309 | tok/s 2486741.02 | norm 0.4117 | dt 210.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23619 | train 2.998442 | val 3.300309 | tok/s 2627286.74 | norm 0.3585 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23620 | train 3.055946 | val 3.300309 | tok/s 2624249.97 | norm 0.4021 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23621 | train 2.887134 | val 3.300309 | tok/s 2623311.50 | norm 0.3999 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23622 | train 2.932618 | val 3.300309 | tok/s 2628138.32 | norm 0.3710 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23623 | train 2.906488 | val 3.300309 | tok/s 2628712.29 | norm 0.3889 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23624 | train 2.949159 | val 3.300309 | tok/s 2628864.48 | norm 0.3808 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23625 | train 2.993814 | val 3.300309 | tok/s 2619042.36 | norm 0.3763 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23626 | train 2.924315 | val 3.300309 | tok/s 2624940.08 | norm 0.3363 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23627 | train 3.075788 | val 3.300309 | tok/s 2632635.45 | norm 0.4889 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23628 | train 2.962723 | val 3.300309 | tok/s 2628316.91 | norm 0.3723 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23629 | train 2.941246 | val 3.300309 | tok/s 2628509.88 | norm 0.3083 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23630 | train 2.827862 | val 3.300309 | tok/s 2624711.62 | norm 0.3261 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23631 | train 2.981696 | val 3.300309 | tok/s 2626655.14 | norm 0.3260 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23632 | train 2.920463 | val 3.300309 | tok/s 2631643.13 | norm 0.3460 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23633 | train 2.921498 | val 3.300309 | tok/s 2629515.40 | norm 0.3671 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23634 | train 2.892392 | val 3.300309 | tok/s 2629397.93 | norm 0.3526 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23635 | train 2.927490 | val 3.300309 | tok/s 2624026.40 | norm 0.3707 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23636 | train 3.035767 | val 3.300309 | tok/s 2625909.11 | norm 0.4001 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23637 | train 2.935940 | val 3.300309 | tok/s 2619717.97 | norm 0.3361 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23638 | train 2.940963 | val 3.300309 | tok/s 2628962.72 | norm 0.3398 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23639 | train 2.950751 | val 3.300309 | tok/s 2625020.46 | norm 0.3473 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23640 | train 2.927406 | val 3.300309 | tok/s 2626682.49 | norm 0.3459 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23641 | train 2.897236 | val 3.300309 | tok/s 2627839.80 | norm 0.3492 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23642 | train 2.871623 | val 3.300309 | tok/s 2631295.12 | norm 0.3831 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23643 | train 2.911800 | val 3.300309 | tok/s 2627638.57 | norm 0.3750 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23644 | train 2.971325 | val 3.300309 | tok/s 2629713.12 | norm 0.3717 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23645 | train 3.087679 | val 3.300309 | tok/s 2623840.13 | norm 0.4332 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23646 | train 2.922974 | val 3.300309 | tok/s 2626506.62 | norm 0.4093 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23647 | train 2.989243 | val 3.300309 | tok/s 2629555.23 | norm 0.3667 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23648 | train 2.891551 | val 3.300309 | tok/s 2627806.47 | norm 0.3712 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23649 | train 3.050273 | val 3.300309 | tok/s 2630072.37 | norm 0.3545 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23650 | train 2.909077 | val 3.300309 | tok/s 2627186.69 | norm 0.3596 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23651 | train 2.986491 | val 3.300309 | tok/s 2628636.91 | norm 0.3452 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23652 | train 2.853600 | val 3.300309 | tok/s 2629953.38 | norm 0.3642 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23653 | train 3.036739 | val 3.300309 | tok/s 2628040.07 | norm 0.3808 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23654 | train 2.918041 | val 3.300309 | tok/s 2630757.99 | norm 0.3351 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23655 | train 2.964260 | val 3.300309 | tok/s 2626779.74 | norm 0.3450 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23656 | train 2.905472 | val 3.300309 | tok/s 2628908.60 | norm 0.3717 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23657 | train 2.826202 | val 3.300309 | tok/s 2633633.55 | norm 0.4018 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23658 | train 2.729013 | val 3.300309 | tok/s 2622993.14 | norm 0.3985 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23659 | train 2.812346 | val 3.300309 | tok/s 2628736.37 | norm 0.3942 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23660 | train 2.894126 | val 3.300309 | tok/s 2558206.01 | norm 0.3780 | dt 204.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23661 | train 2.951780 | val 3.300309 | tok/s 2461604.11 | norm 0.3853 | dt 212.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23662 | train 2.965900 | val 3.300309 | tok/s 2624758.56 | norm 0.3442 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23663 | train 2.824253 | val 3.300309 | tok/s 2627079.27 | norm 0.2984 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23664 | train 2.835935 | val 3.300309 | tok/s 2630582.97 | norm 0.3037 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23665 | train 2.977755 | val 3.300309 | tok/s 2627309.42 | norm 0.3251 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23666 | train 3.104861 | val 3.300309 | tok/s 2626143.24 | norm 0.3405 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23667 | train 2.897047 | val 3.300309 | tok/s 2620081.99 | norm 0.3241 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23668 | train 2.879022 | val 3.300309 | tok/s 2621071.73 | norm 0.3403 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23669 | train 2.875455 | val 3.300309 | tok/s 2630678.96 | norm 0.3611 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23670 | train 2.882979 | val 3.300309 | tok/s 2623906.45 | norm 0.3333 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23671 | train 2.816583 | val 3.300309 | tok/s 2628191.72 | norm 0.3435 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23672 | train 3.001397 | val 3.300309 | tok/s 2620836.66 | norm 0.3627 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23673 | train 2.917817 | val 3.300309 | tok/s 2625188.03 | norm 0.3481 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23674 | train 2.945593 | val 3.300309 | tok/s 2629102.84 | norm 0.3960 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23675 | train 3.058363 | val 3.300309 | tok/s 2629820.34 | norm 0.4337 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23676 | train 3.002024 | val 3.300309 | tok/s 2624949.44 | norm 0.3952 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23677 | train 2.910160 | val 3.300309 | tok/s 2619674.35 | norm 0.3431 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23678 | train 2.873899 | val 3.300309 | tok/s 2622895.50 | norm 0.3695 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23679 | train 2.942497 | val 3.300309 | tok/s 2629560.20 | norm 0.3616 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23680 | train 2.916602 | val 3.300309 | tok/s 2630708.29 | norm 0.3440 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23681 | train 2.948255 | val 3.300309 | tok/s 2627662.83 | norm 0.3687 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23682 | train 3.018872 | val 3.300309 | tok/s 2631467.19 | norm 0.4014 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23683 | train 2.961139 | val 3.300309 | tok/s 2625349.97 | norm 0.4041 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23684 | train 2.878727 | val 3.300309 | tok/s 2632881.04 | norm 0.3638 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23685 | train 3.355304 | val 3.300309 | tok/s 2629281.08 | norm 0.4267 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23686 | train 3.031168 | val 3.300309 | tok/s 2631057.45 | norm 0.3813 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23687 | train 2.839278 | val 3.300309 | tok/s 2628776.55 | norm 0.3704 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23688 | train 2.898824 | val 3.300309 | tok/s 2626723.85 | norm 0.3816 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23689 | train 2.941980 | val 3.300309 | tok/s 2630052.71 | norm 0.3901 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23690 | train 2.930914 | val 3.300309 | tok/s 2626015.62 | norm 0.3736 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23691 | train 2.802058 | val 3.300309 | tok/s 2629079.76 | norm 0.3999 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23692 | train 2.751029 | val 3.300309 | tok/s 2617806.48 | norm 0.3711 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23693 | train 2.963789 | val 3.300309 | tok/s 2623517.09 | norm 0.4068 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23694 | train 2.952703 | val 3.300309 | tok/s 2630219.25 | norm 0.4080 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23695 | train 2.885310 | val 3.300309 | tok/s 2626795.08 | norm 0.3974 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23696 | train 2.942346 | val 3.300309 | tok/s 2630981.35 | norm 0.3578 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23697 | train 2.890134 | val 3.300309 | tok/s 2624526.84 | norm 0.3484 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23698 | train 2.880067 | val 3.300309 | tok/s 2630214.83 | norm 0.3291 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23699 | train 2.914606 | val 3.300309 | tok/s 2627605.87 | norm 0.3146 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23700 | train 2.984189 | val 3.300309 | tok/s 2620094.37 | norm 0.3325 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23701 | train 2.999653 | val 3.300309 | tok/s 2632211.18 | norm 0.3444 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23702 | train 2.907043 | val 3.300309 | tok/s 2490385.11 | norm 0.3389 | dt 210.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23703 | train 2.912827 | val 3.300309 | tok/s 2599526.11 | norm 0.3499 | dt 201.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23704 | train 3.136982 | val 3.300309 | tok/s 2630926.87 | norm 0.4054 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23705 | train 2.839875 | val 3.300309 | tok/s 2579608.36 | norm 0.3794 | dt 203.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23706 | train 3.249157 | val 3.300309 | tok/s 2628328.88 | norm 0.4686 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23707 | train 2.993916 | val 3.300309 | tok/s 2629234.73 | norm 0.4121 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23708 | train 3.107190 | val 3.300309 | tok/s 2629243.74 | norm 0.3742 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23709 | train 2.866599 | val 3.300309 | tok/s 2506279.20 | norm 0.3662 | dt 209.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23710 | train 2.912433 | val 3.300309 | tok/s 2625628.51 | norm 0.3884 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23711 | train 2.885747 | val 3.300309 | tok/s 2630045.40 | norm 0.3848 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23712 | train 2.956193 | val 3.300309 | tok/s 2628591.46 | norm 0.3796 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23713 | train 2.920989 | val 3.300309 | tok/s 2627013.25 | norm 0.3766 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23714 | train 2.963161 | val 3.300309 | tok/s 2619720.79 | norm 0.3630 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23715 | train 2.990954 | val 3.300309 | tok/s 2623703.44 | norm 0.3695 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23716 | train 2.819207 | val 3.300309 | tok/s 2610054.12 | norm 0.3491 | dt 200.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23717 | train 2.875376 | val 3.300309 | tok/s 2628792.88 | norm 0.3402 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23718 | train 2.960539 | val 3.300309 | tok/s 2629700.56 | norm 0.3860 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23719 | train 2.860999 | val 3.300309 | tok/s 2609084.15 | norm 0.3891 | dt 200.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23720 | train 2.922037 | val 3.300309 | tok/s 2625982.08 | norm 0.3766 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23721 | train 2.976489 | val 3.300309 | tok/s 2632884.60 | norm 0.3686 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23722 | train 2.826248 | val 3.300309 | tok/s 2626746.33 | norm 0.4486 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23723 | train 3.021655 | val 3.300309 | tok/s 2631395.87 | norm 0.3391 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23724 | train 2.991261 | val 3.300309 | tok/s 2622723.80 | norm 0.3368 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23725 | train 2.933121 | val 3.300309 | tok/s 2625807.82 | norm 0.3463 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23726 | train 2.914313 | val 3.300309 | tok/s 2627269.85 | norm 0.3418 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23727 | train 2.887406 | val 3.300309 | tok/s 2625872.59 | norm 0.3440 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23728 | train 2.931400 | val 3.300309 | tok/s 2631987.02 | norm 0.3784 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23729 | train 3.037271 | val 3.300309 | tok/s 2630607.71 | norm 0.3607 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23730 | train 3.145700 | val 3.300309 | tok/s 2625192.16 | norm 0.3725 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23731 | train 2.946312 | val 3.300309 | tok/s 2628538.14 | norm 0.3368 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23732 | train 2.946030 | val 3.300309 | tok/s 2626173.43 | norm 0.3368 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23733 | train 3.013002 | val 3.300309 | tok/s 2631199.66 | norm 0.3384 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23734 | train 3.033082 | val 3.300309 | tok/s 2629659.97 | norm 0.4275 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23735 | train 3.047847 | val 3.300309 | tok/s 2627520.35 | norm 0.3751 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23736 | train 2.984967 | val 3.300309 | tok/s 2630731.90 | norm 0.3677 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23737 | train 2.918494 | val 3.300309 | tok/s 2626517.79 | norm 0.3629 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23738 | train 3.101815 | val 3.300309 | tok/s 2628800.98 | norm 0.3595 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23739 | train 3.070240 | val 3.300309 | tok/s 2627253.39 | norm 0.3167 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23740 | train 3.015522 | val 3.300309 | tok/s 2627924.40 | norm 0.3661 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23741 | train 2.923395 | val 3.300309 | tok/s 2625749.84 | norm 0.3470 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23742 | train 2.988267 | val 3.300309 | tok/s 2621478.95 | norm 0.3078 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23743 | train 2.997558 | val 3.300309 | tok/s 2630108.55 | norm 0.3531 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23744 | train 2.988256 | val 3.300309 | tok/s 2626831.16 | norm 0.3898 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23745 | train 2.923531 | val 3.300309 | tok/s 2626090.06 | norm 0.3607 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23746 | train 2.830600 | val 3.300309 | tok/s 2625237.54 | norm 0.3584 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23747 | train 3.008337 | val 3.300309 | tok/s 2438108.99 | norm 0.4286 | dt 215.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23748 | train 2.865409 | val 3.300309 | tok/s 2617942.89 | norm 0.4125 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23749 | train 2.885763 | val 3.300309 | tok/s 2631336.63 | norm 0.3797 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 23750 | val_loss 3.301326 step 23750 | train 3.004402 | val 3.301326 | tok/s 2609558.35 | norm 0.4117 | dt 200.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23751 | train 3.102041 | val 3.301326 | tok/s 2630413.58 | norm 0.4192 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23752 | train 2.852264 | val 3.301326 | tok/s 2619154.34 | norm 0.3218 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23753 | train 2.928245 | val 3.301326 | tok/s 2623703.52 | norm 0.3010 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23754 | train 3.130255 | val 3.301326 | tok/s 2630214.45 | norm 0.3349 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23755 | train 2.975357 | val 3.301326 | tok/s 2627991.77 | norm 0.3298 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23756 | train 2.951074 | val 3.301326 | tok/s 2628977.99 | norm 0.3318 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23757 | train 2.862484 | val 3.301326 | tok/s 2624575.62 | norm 0.3583 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23758 | train 2.925451 | val 3.301326 | tok/s 2620190.03 | norm 0.3674 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23759 | train 2.895495 | val 3.301326 | tok/s 2630075.01 | norm 0.3512 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23760 | train 2.973933 | val 3.301326 | tok/s 2626637.72 | norm 0.3814 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23761 | train 3.015824 | val 3.301326 | tok/s 2628016.20 | norm 0.4028 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23762 | train 3.018471 | val 3.301326 | tok/s 2619061.64 | norm 0.3654 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23763 | train 2.920089 | val 3.301326 | tok/s 2598261.84 | norm 0.3794 | dt 201.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23764 | train 3.014305 | val 3.301326 | tok/s 2629197.13 | norm 0.3895 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23765 | train 2.945851 | val 3.301326 | tok/s 2627297.70 | norm 0.3856 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23766 | train 2.999193 | val 3.301326 | tok/s 2628831.84 | norm 0.3880 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23767 | train 2.868865 | val 3.301326 | tok/s 2627565.05 | norm 0.3587 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23768 | train 2.961874 | val 3.301326 | tok/s 2627536.45 | norm 0.3470 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23769 | train 2.860275 | val 3.301326 | tok/s 2628719.07 | norm 0.3513 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23770 | train 3.038747 | val 3.301326 | tok/s 2617607.47 | norm 0.3732 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23771 | train 2.940341 | val 3.301326 | tok/s 2624693.40 | norm 0.3715 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23772 | train 2.936096 | val 3.301326 | tok/s 2627637.06 | norm 0.3448 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23773 | train 2.934752 | val 3.301326 | tok/s 2624318.15 | norm 0.4371 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23774 | train 3.044900 | val 3.301326 | tok/s 2631563.72 | norm 0.3595 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23775 | train 2.928561 | val 3.301326 | tok/s 2626137.00 | norm 0.3201 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23776 | train 2.956942 | val 3.301326 | tok/s 2629998.86 | norm 0.3750 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23777 | train 2.955747 | val 3.301326 | tok/s 2623911.08 | norm 0.3370 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23778 | train 2.916408 | val 3.301326 | tok/s 2624903.09 | norm 0.3512 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23779 | train 2.952103 | val 3.301326 | tok/s 2634054.45 | norm 0.3777 | dt 199.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23780 | train 2.808793 | val 3.301326 | tok/s 2626442.41 | norm 0.3879 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23781 | train 2.968779 | val 3.301326 | tok/s 2630808.88 | norm 0.3796 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23782 | train 2.908006 | val 3.301326 | tok/s 2610530.61 | norm 0.3679 | dt 200.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23783 | train 2.915062 | val 3.301326 | tok/s 2623289.72 | norm 0.3292 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23784 | train 2.976990 | val 3.301326 | tok/s 2628370.87 | norm 0.3026 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23785 | train 2.860643 | val 3.301326 | tok/s 2626779.23 | norm 0.3384 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23786 | train 2.909156 | val 3.301326 | tok/s 2629681.92 | norm 0.3553 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23787 | train 2.897643 | val 3.301326 | tok/s 2629501.61 | norm 0.3378 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23788 | train 2.986638 | val 3.301326 | tok/s 2628973.94 | norm 0.3715 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23789 | train 2.976143 | val 3.301326 | tok/s 2629716.51 | norm 0.4118 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23790 | train 2.951070 | val 3.301326 | tok/s 2625248.50 | norm 0.3973 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23791 | train 3.016688 | val 3.301326 | tok/s 2628689.91 | norm 0.3856 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23792 | train 3.181954 | val 3.301326 | tok/s 2628905.10 | norm 0.3667 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23793 | train 2.936386 | val 3.301326 | tok/s 2630012.90 | norm 0.4096 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23794 | train 2.916833 | val 3.301326 | tok/s 2631584.83 | norm 0.4020 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23795 | train 2.860188 | val 3.301326 | tok/s 2628562.93 | norm 0.3345 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23796 | train 2.804308 | val 3.301326 | tok/s 2628302.30 | norm 0.3473 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23797 | train 2.884426 | val 3.301326 | tok/s 2624709.09 | norm 0.3855 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23798 | train 2.872220 | val 3.301326 | tok/s 2627786.25 | norm 0.3550 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23799 | train 2.958969 | val 3.301326 | tok/s 2630055.80 | norm 0.3463 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23800 | train 2.963509 | val 3.301326 | tok/s 2625561.99 | norm 0.3603 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23801 | train 3.129259 | val 3.301326 | tok/s 2628079.26 | norm 0.4536 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23802 | train 2.933165 | val 3.301326 | tok/s 2625457.58 | norm 0.3377 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23803 | train 2.940106 | val 3.301326 | tok/s 2628889.91 | norm 0.3362 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23804 | train 2.905666 | val 3.301326 | tok/s 2617891.34 | norm 0.3749 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23805 | train 2.765850 | val 3.301326 | tok/s 2619654.72 | norm 0.3833 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23806 | train 2.816337 | val 3.301326 | tok/s 2557904.80 | norm 0.3687 | dt 204.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23807 | train 2.905656 | val 3.301326 | tok/s 2565987.08 | norm 0.3647 | dt 204.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23808 | train 2.859562 | val 3.301326 | tok/s 2591861.56 | norm 0.3726 | dt 202.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23809 | train 2.929168 | val 3.301326 | tok/s 2570274.85 | norm 0.3845 | dt 203.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23810 | train 2.951326 | val 3.301326 | tok/s 2622366.68 | norm 0.3921 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23811 | train 3.154785 | val 3.301326 | tok/s 2632177.65 | norm 0.3777 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23812 | train 2.918159 | val 3.301326 | tok/s 2627675.75 | norm 0.3566 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23813 | train 3.085357 | val 3.301326 | tok/s 2629693.28 | norm 0.3397 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23814 | train 2.920036 | val 3.301326 | tok/s 2476431.71 | norm 0.3638 | dt 211.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23815 | train 3.050590 | val 3.301326 | tok/s 2540848.49 | norm 0.3314 | dt 206.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23816 | train 2.931942 | val 3.301326 | tok/s 2631601.24 | norm 0.3403 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23817 | train 2.895293 | val 3.301326 | tok/s 2618795.68 | norm 0.3684 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23818 | train 2.918400 | val 3.301326 | tok/s 2628734.70 | norm 0.3814 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23819 | train 2.976785 | val 3.301326 | tok/s 2624617.09 | norm 0.3582 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23820 | train 2.823921 | val 3.301326 | tok/s 2622495.40 | norm 0.3719 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23821 | train 3.007560 | val 3.301326 | tok/s 2631236.36 | norm 0.3732 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23822 | train 3.008095 | val 3.301326 | tok/s 2610353.98 | norm 0.3615 | dt 200.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23823 | train 2.963967 | val 3.301326 | tok/s 2626609.25 | norm 0.3730 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23824 | train 2.829852 | val 3.301326 | tok/s 2627568.70 | norm 0.4003 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23825 | train 2.888772 | val 3.301326 | tok/s 2627350.80 | norm 0.3878 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23826 | train 2.880285 | val 3.301326 | tok/s 2628386.66 | norm 0.3614 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23827 | train 2.798624 | val 3.301326 | tok/s 2629184.50 | norm 0.3926 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23828 | train 2.953526 | val 3.301326 | tok/s 2630036.70 | norm 0.3711 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23829 | train 2.927938 | val 3.301326 | tok/s 2629716.30 | norm 0.3368 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23830 | train 2.954772 | val 3.301326 | tok/s 2628171.07 | norm 0.3589 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23831 | train 2.865228 | val 3.301326 | tok/s 2631368.06 | norm 0.3586 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23832 | train 2.997267 | val 3.301326 | tok/s 2629673.78 | norm 0.3622 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23833 | train 2.880832 | val 3.301326 | tok/s 2629628.76 | norm 0.3372 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23834 | train 2.927337 | val 3.301326 | tok/s 2630242.84 | norm 0.3419 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23835 | train 2.796656 | val 3.301326 | tok/s 2625304.87 | norm 0.3769 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23836 | train 2.973508 | val 3.301326 | tok/s 2633161.86 | norm 0.3606 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23837 | train 2.865435 | val 3.301326 | tok/s 2627954.19 | norm 0.3613 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23838 | train 2.898406 | val 3.301326 | tok/s 2631172.84 | norm 0.3777 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23839 | train 2.857635 | val 3.301326 | tok/s 2626011.32 | norm 0.3789 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23840 | train 2.967115 | val 3.301326 | tok/s 2628741.29 | norm 0.3911 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23841 | train 2.893723 | val 3.301326 | tok/s 2631057.57 | norm 0.3908 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23842 | train 2.991025 | val 3.301326 | tok/s 2625113.90 | norm 0.3730 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23843 | train 2.953129 | val 3.301326 | tok/s 2629959.42 | norm 0.3437 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23844 | train 3.029194 | val 3.301326 | tok/s 2624393.02 | norm 0.3672 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23845 | train 2.941938 | val 3.301326 | tok/s 2630244.15 | norm 0.3714 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23846 | train 2.914489 | val 3.301326 | tok/s 2626876.14 | norm 0.3656 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23847 | train 2.874745 | val 3.301326 | tok/s 2622445.66 | norm 0.3544 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23848 | train 2.810052 | val 3.301326 | tok/s 2632869.88 | norm 0.3610 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23849 | train 3.018343 | val 3.301326 | tok/s 2623823.45 | norm 0.3849 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23850 | train 2.927585 | val 3.301326 | tok/s 2625280.48 | norm 0.3847 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23851 | train 2.972550 | val 3.301326 | tok/s 2626925.07 | norm 0.3703 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23852 | train 2.911115 | val 3.301326 | tok/s 2618358.86 | norm 0.3893 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23853 | train 2.934771 | val 3.301326 | tok/s 2631567.13 | norm 0.3531 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23854 | train 2.979905 | val 3.301326 | tok/s 2622217.97 | norm 0.3346 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23855 | train 2.962876 | val 3.301326 | tok/s 2631603.05 | norm 0.3810 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23856 | train 2.862181 | val 3.301326 | tok/s 2538586.00 | norm 0.3531 | dt 206.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23857 | train 2.913210 | val 3.301326 | tok/s 2622720.53 | norm 0.3438 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23858 | train 2.913862 | val 3.301326 | tok/s 2628749.37 | norm 0.3663 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23859 | train 2.851007 | val 3.301326 | tok/s 2624978.42 | norm 0.3702 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23860 | train 3.022511 | val 3.301326 | tok/s 2628812.86 | norm 0.3732 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23861 | train 2.943981 | val 3.301326 | tok/s 2618067.73 | norm 0.3523 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23862 | train 3.012584 | val 3.301326 | tok/s 2622346.75 | norm 0.3668 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23863 | train 3.012547 | val 3.301326 | tok/s 2491803.24 | norm 0.5208 | dt 210.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23864 | train 3.043576 | val 3.301326 | tok/s 2621338.42 | norm 0.4020 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23865 | train 2.950949 | val 3.301326 | tok/s 2628806.34 | norm 0.3561 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23866 | train 3.014444 | val 3.301326 | tok/s 2621810.15 | norm 0.3255 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23867 | train 3.027728 | val 3.301326 | tok/s 2624324.67 | norm 0.3412 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23868 | train 2.967968 | val 3.301326 | tok/s 2596863.08 | norm 0.3318 | dt 201.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23869 | train 2.924025 | val 3.301326 | tok/s 2558779.92 | norm 0.3461 | dt 204.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23870 | train 2.783292 | val 3.301326 | tok/s 2629878.66 | norm 0.3656 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23871 | train 2.988603 | val 3.301326 | tok/s 2625509.46 | norm 0.3877 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23872 | train 2.861143 | val 3.301326 | tok/s 2626428.86 | norm 0.3533 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23873 | train 3.015314 | val 3.301326 | tok/s 2632948.47 | norm 0.3705 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23874 | train 2.969580 | val 3.301326 | tok/s 2627479.08 | norm 0.3857 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23875 | train 2.814686 | val 3.301326 | tok/s 2628921.70 | norm 0.3405 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23876 | train 2.787595 | val 3.301326 | tok/s 2627911.13 | norm 0.3573 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23877 | train 2.974959 | val 3.301326 | tok/s 2628382.77 | norm 0.4070 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23878 | train 3.016368 | val 3.301326 | tok/s 2631238.72 | norm 0.3838 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23879 | train 2.866496 | val 3.301326 | tok/s 2628660.71 | norm 0.3442 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23880 | train 2.967946 | val 3.301326 | tok/s 2630255.39 | norm 0.3779 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23881 | train 3.045567 | val 3.301326 | tok/s 2623137.51 | norm 0.3977 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23882 | train 2.901396 | val 3.301326 | tok/s 2626846.48 | norm 0.3792 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23883 | train 2.807862 | val 3.301326 | tok/s 2629992.78 | norm 0.5183 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23884 | train 2.771007 | val 3.301326 | tok/s 2628443.84 | norm 0.3842 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23885 | train 2.897074 | val 3.301326 | tok/s 2628967.85 | norm 0.3652 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23886 | train 2.876347 | val 3.301326 | tok/s 2625995.45 | norm 0.3469 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23887 | train 2.891821 | val 3.301326 | tok/s 2628397.97 | norm 0.3890 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23888 | train 3.012788 | val 3.301326 | tok/s 2631407.94 | norm 0.3613 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23889 | train 2.808663 | val 3.301326 | tok/s 2625859.39 | norm 0.3220 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23890 | train 2.907283 | val 3.301326 | tok/s 2629360.83 | norm 0.3682 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23891 | train 2.903524 | val 3.301326 | tok/s 2620154.15 | norm 0.3816 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23892 | train 2.973072 | val 3.301326 | tok/s 2627940.18 | norm 0.3817 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23893 | train 2.932840 | val 3.301326 | tok/s 2624028.84 | norm 0.3822 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23894 | train 3.045521 | val 3.301326 | tok/s 2617473.34 | norm 0.3883 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23895 | train 2.982924 | val 3.301326 | tok/s 2628598.62 | norm 0.3594 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23896 | train 2.938973 | val 3.301326 | tok/s 2631206.76 | norm 0.3484 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23897 | train 2.978873 | val 3.301326 | tok/s 2628531.71 | norm 0.3861 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23898 | train 2.852621 | val 3.301326 | tok/s 2632673.06 | norm 0.3691 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23899 | train 2.864311 | val 3.301326 | tok/s 2625971.01 | norm 0.3685 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23900 | train 2.960232 | val 3.301326 | tok/s 2629533.77 | norm 0.3637 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23901 | train 2.790633 | val 3.301326 | tok/s 2622785.41 | norm 0.3549 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23902 | train 2.995817 | val 3.301326 | tok/s 2624251.53 | norm 0.3633 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23903 | train 2.947390 | val 3.301326 | tok/s 2619899.56 | norm 0.3223 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23904 | train 2.854252 | val 3.301326 | tok/s 2625797.10 | norm 0.3077 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23905 | train 2.942972 | val 3.301326 | tok/s 2629250.25 | norm 0.3426 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23906 | train 3.010001 | val 3.301326 | tok/s 2627923.94 | norm 0.3740 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23907 | train 2.877020 | val 3.301326 | tok/s 2589277.15 | norm 0.3600 | dt 202.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23908 | train 2.844949 | val 3.301326 | tok/s 2613949.48 | norm 0.3595 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23909 | train 2.997674 | val 3.301326 | tok/s 2624714.91 | norm 0.3480 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23910 | train 3.008209 | val 3.301326 | tok/s 2632439.54 | norm 0.3556 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23911 | train 2.933398 | val 3.301326 | tok/s 2629461.73 | norm 0.3427 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23912 | train 2.905483 | val 3.301326 | tok/s 2629025.20 | norm 0.3413 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23913 | train 2.886944 | val 3.301326 | tok/s 2618058.11 | norm 0.3129 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23914 | train 2.958965 | val 3.301326 | tok/s 2624805.53 | norm 0.3235 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23915 | train 2.877200 | val 3.301326 | tok/s 2630626.56 | norm 0.3605 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23916 | train 2.994064 | val 3.301326 | tok/s 2629215.08 | norm 0.3590 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23917 | train 2.934440 | val 3.301326 | tok/s 2629117.22 | norm 0.3409 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23918 | train 2.857289 | val 3.301326 | tok/s 2623537.41 | norm 0.3344 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23919 | train 3.001239 | val 3.301326 | tok/s 2626588.56 | norm 0.3517 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23920 | train 3.005544 | val 3.301326 | tok/s 2627007.13 | norm 0.3789 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23921 | train 3.015425 | val 3.301326 | tok/s 2625340.09 | norm 0.3822 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23922 | train 2.915038 | val 3.301326 | tok/s 2564754.07 | norm 0.3743 | dt 204.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23923 | train 2.941157 | val 3.301326 | tok/s 2622353.26 | norm 0.3517 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23924 | train 2.952049 | val 3.301326 | tok/s 2625209.45 | norm 0.3462 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23925 | train 2.925484 | val 3.301326 | tok/s 2631462.82 | norm 0.3939 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23926 | train 2.949022 | val 3.301326 | tok/s 2624646.35 | norm 0.3796 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23927 | train 2.948670 | val 3.301326 | tok/s 2629086.67 | norm 0.3612 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23928 | train 3.003034 | val 3.301326 | tok/s 2628170.55 | norm 0.3851 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23929 | train 2.806883 | val 3.301326 | tok/s 2625121.39 | norm 0.3793 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23930 | train 3.039287 | val 3.301326 | tok/s 2630340.33 | norm 0.3611 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23931 | train 2.915345 | val 3.301326 | tok/s 2623086.51 | norm 0.3387 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23932 | train 2.996385 | val 3.301326 | tok/s 2629614.41 | norm 0.3620 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23933 | train 2.839055 | val 3.301326 | tok/s 2623156.92 | norm 0.3767 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23934 | train 3.058308 | val 3.301326 | tok/s 2623003.36 | norm 0.3908 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23935 | train 2.906924 | val 3.301326 | tok/s 2628929.28 | norm 0.4065 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23936 | train 2.857897 | val 3.301326 | tok/s 2619060.05 | norm 0.3809 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23937 | train 2.922458 | val 3.301326 | tok/s 2632533.36 | norm 0.3843 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23938 | train 2.962567 | val 3.301326 | tok/s 2629441.71 | norm 0.3803 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23939 | train 3.028177 | val 3.301326 | tok/s 2631094.22 | norm 0.3425 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23940 | train 2.907013 | val 3.301326 | tok/s 2621978.20 | norm 0.3370 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23941 | train 2.944752 | val 3.301326 | tok/s 2624941.20 | norm 0.3210 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23942 | train 2.925987 | val 3.301326 | tok/s 2630879.78 | norm 0.3706 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23943 | train 2.883953 | val 3.301326 | tok/s 2627771.31 | norm 0.4041 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23944 | train 2.928160 | val 3.301326 | tok/s 2628418.13 | norm 0.3879 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23945 | train 2.822731 | val 3.301326 | tok/s 2631001.36 | norm 0.3925 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23946 | train 3.189113 | val 3.301326 | tok/s 2621712.41 | norm 0.3799 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23947 | train 2.984287 | val 3.301326 | tok/s 2630210.23 | norm 0.3529 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23948 | train 2.912458 | val 3.301326 | tok/s 2625204.73 | norm 0.3319 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23949 | train 2.969566 | val 3.301326 | tok/s 2626960.78 | norm 0.3051 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23950 | train 2.898084 | val 3.301326 | tok/s 2617255.30 | norm 0.3627 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23951 | train 2.958572 | val 3.301326 | tok/s 2610432.61 | norm 0.3629 | dt 200.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23952 | train 2.948889 | val 3.301326 | tok/s 2629130.85 | norm 0.3674 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23953 | train 2.925097 | val 3.301326 | tok/s 2626913.94 | norm 0.3428 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23954 | train 2.941050 | val 3.301326 | tok/s 2629858.29 | norm 0.3837 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23955 | train 2.847790 | val 3.301326 | tok/s 2613731.01 | norm 0.3397 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23956 | train 2.914544 | val 3.301326 | tok/s 2625778.82 | norm 0.3252 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23957 | train 3.061936 | val 3.301326 | tok/s 2629561.65 | norm 0.3670 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23958 | train 2.961871 | val 3.301326 | tok/s 2627581.00 | norm 0.3632 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23959 | train 2.887102 | val 3.301326 | tok/s 2626444.18 | norm 0.3452 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23960 | train 2.892234 | val 3.301326 | tok/s 2617448.57 | norm 0.3657 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23961 | train 3.061396 | val 3.301326 | tok/s 2620600.52 | norm 0.3998 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23962 | train 2.901464 | val 3.301326 | tok/s 2631448.17 | norm 0.3616 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23963 | train 2.960513 | val 3.301326 | tok/s 2627198.95 | norm 0.3879 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23964 | train 3.001151 | val 3.301326 | tok/s 2626575.47 | norm 0.3986 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23965 | train 3.066465 | val 3.301326 | tok/s 2629857.64 | norm 0.3660 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23966 | train 2.975102 | val 3.301326 | tok/s 2628265.24 | norm 0.3562 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23967 | train 2.958626 | val 3.301326 | tok/s 2632220.92 | norm 0.3913 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23968 | train 2.939466 | val 3.301326 | tok/s 2475157.67 | norm 0.4137 | dt 211.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23969 | train 2.945845 | val 3.301326 | tok/s 2577682.77 | norm 0.3644 | dt 203.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23970 | train 2.923077 | val 3.301326 | tok/s 2624886.80 | norm 0.3294 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23971 | train 2.878350 | val 3.301326 | tok/s 2626472.03 | norm 0.3863 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23972 | train 3.052482 | val 3.301326 | tok/s 2633335.24 | norm 0.4166 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23973 | train 3.198710 | val 3.301326 | tok/s 2628566.74 | norm 0.4424 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23974 | train 2.950583 | val 3.301326 | tok/s 2631333.70 | norm 0.3877 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23975 | train 2.878656 | val 3.301326 | tok/s 2625101.78 | norm 0.3564 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23976 | train 3.010581 | val 3.301326 | tok/s 2624735.43 | norm 0.3746 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23977 | train 2.979632 | val 3.301326 | tok/s 2631440.67 | norm 0.3615 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23978 | train 2.909043 | val 3.301326 | tok/s 2626401.93 | norm 0.3559 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23979 | train 3.017650 | val 3.301326 | tok/s 2629114.54 | norm 0.3441 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23980 | train 2.936387 | val 3.301326 | tok/s 2627888.42 | norm 0.3274 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23981 | train 2.849016 | val 3.301326 | tok/s 2627441.96 | norm 0.3450 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23982 | train 2.880639 | val 3.301326 | tok/s 2630357.60 | norm 0.3602 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23983 | train 2.879833 | val 3.301326 | tok/s 2628495.39 | norm 0.3689 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23984 | train 2.839533 | val 3.301326 | tok/s 2629285.05 | norm 0.3957 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23985 | train 2.883759 | val 3.301326 | tok/s 2626132.00 | norm 0.3861 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23986 | train 2.906519 | val 3.301326 | tok/s 2628475.40 | norm 0.3597 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23987 | train 3.131033 | val 3.301326 | tok/s 2627808.98 | norm 0.4048 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23988 | train 2.969606 | val 3.301326 | tok/s 2624021.63 | norm 0.3541 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23989 | train 2.956297 | val 3.301326 | tok/s 2625806.27 | norm 0.3092 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23990 | train 2.945628 | val 3.301326 | tok/s 2630343.95 | norm 0.3353 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23991 | train 3.047320 | val 3.301326 | tok/s 2614388.58 | norm 0.3936 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23992 | train 2.967794 | val 3.301326 | tok/s 2624975.78 | norm 0.3897 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23993 | train 2.906122 | val 3.301326 | tok/s 2624571.01 | norm 0.3650 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23994 | train 2.942620 | val 3.301326 | tok/s 2632652.49 | norm 0.3849 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23995 | train 2.949350 | val 3.301326 | tok/s 2630614.40 | norm 0.3883 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23996 | train 2.919109 | val 3.301326 | tok/s 2630312.07 | norm 0.3748 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23997 | train 2.940935 | val 3.301326 | tok/s 2624245.71 | norm 0.3489 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23998 | train 2.888601 | val 3.301326 | tok/s 2620544.74 | norm 0.3485 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 23999 | train 2.848471 | val 3.301326 | tok/s 2625707.13 | norm 0.3790 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 24000 | val_loss 3.293882 step 24000 | train 2.946185 | val 3.293882 | tok/s 2615053.93 | norm 0.3853 | dt 200.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24001 | train 2.920233 | val 3.293882 | tok/s 2627176.32 | norm 0.3872 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24002 | train 2.895488 | val 3.293882 | tok/s 2627620.87 | norm 0.3981 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24003 | train 2.852437 | val 3.293882 | tok/s 2625764.91 | norm 0.3794 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24004 | train 2.889119 | val 3.293882 | tok/s 2620124.37 | norm 0.3875 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24005 | train 2.836309 | val 3.293882 | tok/s 2631629.70 | norm 0.3766 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24006 | train 2.942590 | val 3.293882 | tok/s 2628644.04 | norm 0.3376 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24007 | train 2.883533 | val 3.293882 | tok/s 2626860.00 | norm 0.3549 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24008 | train 2.894927 | val 3.293882 | tok/s 2629888.64 | norm 0.3898 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24009 | train 2.974288 | val 3.293882 | tok/s 2625713.19 | norm 0.4063 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24010 | train 3.020716 | val 3.293882 | tok/s 2632226.63 | norm 0.4532 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24011 | train 3.009247 | val 3.293882 | tok/s 2621427.90 | norm 0.3503 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24012 | train 2.939641 | val 3.293882 | tok/s 2628008.33 | norm 0.3152 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24013 | train 2.909468 | val 3.293882 | tok/s 2616235.30 | norm 0.3444 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24014 | train 2.938717 | val 3.293882 | tok/s 2611028.78 | norm 0.3436 | dt 200.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24015 | train 2.969562 | val 3.293882 | tok/s 2630420.04 | norm 0.3518 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24016 | train 2.826917 | val 3.293882 | tok/s 2626933.46 | norm 0.3575 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24017 | train 2.925087 | val 3.293882 | tok/s 2629348.95 | norm 0.3441 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24018 | train 2.879128 | val 3.293882 | tok/s 2618755.95 | norm 0.3447 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24019 | train 2.903913 | val 3.293882 | tok/s 2624810.68 | norm 0.3571 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24020 | train 2.937524 | val 3.293882 | tok/s 2602092.49 | norm 0.3582 | dt 201.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24021 | train 2.956150 | val 3.293882 | tok/s 2627669.57 | norm 0.3501 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24022 | train 2.883055 | val 3.293882 | tok/s 2631448.60 | norm 0.3453 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24023 | train 2.975310 | val 3.293882 | tok/s 2622502.74 | norm 0.3494 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24024 | train 2.913016 | val 3.293882 | tok/s 2483830.24 | norm 0.3366 | dt 211.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24025 | train 2.873643 | val 3.293882 | tok/s 2504315.52 | norm 0.3354 | dt 209.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24026 | train 2.823102 | val 3.293882 | tok/s 2628070.00 | norm 0.3496 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24027 | train 2.911278 | val 3.293882 | tok/s 2630362.04 | norm 0.3574 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24028 | train 2.891147 | val 3.293882 | tok/s 2623721.01 | norm 0.3634 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24029 | train 2.940145 | val 3.293882 | tok/s 2628077.33 | norm 0.3644 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24030 | train 2.889246 | val 3.293882 | tok/s 2627385.92 | norm 0.3566 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24031 | train 2.907416 | val 3.293882 | tok/s 2627610.19 | norm 0.3713 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24032 | train 2.930460 | val 3.293882 | tok/s 2630897.21 | norm 0.3685 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24033 | train 2.882563 | val 3.293882 | tok/s 2628403.80 | norm 0.3619 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24034 | train 2.900629 | val 3.293882 | tok/s 2425989.75 | norm 0.3796 | dt 216.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24035 | train 2.959650 | val 3.293882 | tok/s 2629170.93 | norm 0.3833 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24036 | train 2.935120 | val 3.293882 | tok/s 2624245.50 | norm 0.3891 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24037 | train 2.809114 | val 3.293882 | tok/s 2630608.76 | norm 0.3693 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24038 | train 2.880728 | val 3.293882 | tok/s 2629393.84 | norm 0.3780 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24039 | train 2.888014 | val 3.293882 | tok/s 2626873.85 | norm 0.3791 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24040 | train 2.924176 | val 3.293882 | tok/s 2626631.38 | norm 0.3776 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24041 | train 2.932975 | val 3.293882 | tok/s 2621935.01 | norm 0.3705 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24042 | train 2.908921 | val 3.293882 | tok/s 2631712.14 | norm 0.3735 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24043 | train 2.850236 | val 3.293882 | tok/s 2629482.97 | norm 0.3816 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24044 | train 2.872011 | val 3.293882 | tok/s 2629627.56 | norm 0.3830 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24045 | train 2.946087 | val 3.293882 | tok/s 2620451.20 | norm 0.3584 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24046 | train 3.067352 | val 3.293882 | tok/s 2622405.96 | norm 0.3571 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24047 | train 2.898035 | val 3.293882 | tok/s 2629491.50 | norm 0.3832 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24048 | train 2.843062 | val 3.293882 | tok/s 2608652.45 | norm 0.3609 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24049 | train 2.917718 | val 3.293882 | tok/s 2621396.20 | norm 0.3771 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24050 | train 3.016171 | val 3.293882 | tok/s 2559746.98 | norm 0.3852 | dt 204.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24051 | train 2.876008 | val 3.293882 | tok/s 2625348.71 | norm 0.3453 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24052 | train 2.976556 | val 3.293882 | tok/s 2624821.61 | norm 0.3851 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24053 | train 2.876372 | val 3.293882 | tok/s 2625158.48 | norm 0.4132 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24054 | train 2.915649 | val 3.293882 | tok/s 2630973.89 | norm 0.3897 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24055 | train 3.023807 | val 3.293882 | tok/s 2626155.26 | norm 0.4042 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24056 | train 3.028475 | val 3.293882 | tok/s 2626927.98 | norm 0.3852 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24057 | train 2.745996 | val 3.293882 | tok/s 2622934.60 | norm 0.3844 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24058 | train 3.024059 | val 3.293882 | tok/s 2622127.25 | norm 0.4051 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24059 | train 3.093128 | val 3.293882 | tok/s 2622427.62 | norm 0.4426 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24060 | train 2.943460 | val 3.293882 | tok/s 2624302.97 | norm 0.3637 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24061 | train 2.918390 | val 3.293882 | tok/s 2614491.47 | norm 0.3401 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24062 | train 2.983113 | val 3.293882 | tok/s 2626087.72 | norm 0.3727 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24063 | train 2.951116 | val 3.293882 | tok/s 2632225.02 | norm 0.3572 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24064 | train 2.935229 | val 3.293882 | tok/s 2516350.68 | norm 0.3779 | dt 208.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24065 | train 2.977690 | val 3.293882 | tok/s 2622324.59 | norm 0.3513 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24066 | train 2.858124 | val 3.293882 | tok/s 2626565.78 | norm 0.3399 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24067 | train 2.889966 | val 3.293882 | tok/s 2627777.41 | norm 0.3698 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24068 | train 2.901910 | val 3.293882 | tok/s 2630143.57 | norm 0.3612 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24069 | train 2.932887 | val 3.293882 | tok/s 2622837.02 | norm 0.3609 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24070 | train 2.957576 | val 3.293882 | tok/s 2605484.40 | norm 0.3567 | dt 201.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24071 | train 2.923416 | val 3.293882 | tok/s 2627862.11 | norm 0.3640 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24072 | train 2.943679 | val 3.293882 | tok/s 2632237.23 | norm 0.3426 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24073 | train 2.980457 | val 3.293882 | tok/s 2630801.93 | norm 0.3324 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24074 | train 3.016567 | val 3.293882 | tok/s 2620035.24 | norm 0.3587 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24075 | train 2.920756 | val 3.293882 | tok/s 2622068.26 | norm 0.3629 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24076 | train 2.883296 | val 3.293882 | tok/s 2631068.58 | norm 0.3539 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24077 | train 2.921633 | val 3.293882 | tok/s 2628502.83 | norm 0.3585 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24078 | train 2.999575 | val 3.293882 | tok/s 2630973.89 | norm 0.3751 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24079 | train 2.987989 | val 3.293882 | tok/s 2620856.43 | norm 0.3776 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24080 | train 3.045199 | val 3.293882 | tok/s 2623461.40 | norm 0.3494 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24081 | train 2.981496 | val 3.293882 | tok/s 2631735.47 | norm 0.3249 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24082 | train 3.205607 | val 3.293882 | tok/s 2626507.72 | norm 0.3942 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24083 | train 2.948342 | val 3.293882 | tok/s 2626160.23 | norm 0.4757 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24084 | train 3.061788 | val 3.293882 | tok/s 2533288.07 | norm 0.3730 | dt 206.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24085 | train 2.919978 | val 3.293882 | tok/s 2626259.08 | norm 0.3322 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24086 | train 2.945577 | val 3.293882 | tok/s 2630892.05 | norm 0.3662 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24087 | train 2.905612 | val 3.293882 | tok/s 2630647.62 | norm 0.3597 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24088 | train 2.998057 | val 3.293882 | tok/s 2628952.97 | norm 0.3431 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24089 | train 2.912792 | val 3.293882 | tok/s 2624562.90 | norm 0.3547 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24090 | train 3.043581 | val 3.293882 | tok/s 2478606.94 | norm 0.3749 | dt 211.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24091 | train 3.035268 | val 3.293882 | tok/s 2543366.95 | norm 0.3605 | dt 206.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24092 | train 2.909865 | val 3.293882 | tok/s 2630334.63 | norm 0.3295 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24093 | train 3.034045 | val 3.293882 | tok/s 2630933.99 | norm 0.3390 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24094 | train 3.068425 | val 3.293882 | tok/s 2626387.85 | norm 0.3804 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24095 | train 3.025009 | val 3.293882 | tok/s 2624028.92 | norm 0.3599 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24096 | train 2.959932 | val 3.293882 | tok/s 2632070.96 | norm 0.3462 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24097 | train 3.015966 | val 3.293882 | tok/s 2626515.22 | norm 0.3386 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24098 | train 2.985713 | val 3.293882 | tok/s 2620434.62 | norm 0.3637 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24099 | train 3.009957 | val 3.293882 | tok/s 2619945.09 | norm 0.3777 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24100 | train 2.906054 | val 3.293882 | tok/s 2626946.45 | norm 0.3630 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24101 | train 2.863591 | val 3.293882 | tok/s 2629703.07 | norm 0.4020 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24102 | train 2.883304 | val 3.293882 | tok/s 2620157.39 | norm 0.3989 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24103 | train 2.912077 | val 3.293882 | tok/s 2628182.05 | norm 0.3157 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24104 | train 2.937745 | val 3.293882 | tok/s 2627701.85 | norm 0.3452 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24105 | train 2.966295 | val 3.293882 | tok/s 2628427.94 | norm 0.3936 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24106 | train 2.884705 | val 3.293882 | tok/s 2622197.08 | norm 0.3629 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24107 | train 2.978076 | val 3.293882 | tok/s 2616477.55 | norm 0.3822 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24108 | train 2.875056 | val 3.293882 | tok/s 2631467.52 | norm 0.3822 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24109 | train 2.931698 | val 3.293882 | tok/s 2629230.78 | norm 0.3709 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24110 | train 2.938002 | val 3.293882 | tok/s 2629406.50 | norm 0.3640 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24111 | train 2.926717 | val 3.293882 | tok/s 2626540.91 | norm 0.3641 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24112 | train 2.957211 | val 3.293882 | tok/s 2619055.78 | norm 0.3696 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24113 | train 3.011946 | val 3.293882 | tok/s 2628334.03 | norm 0.3902 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24114 | train 2.984760 | val 3.293882 | tok/s 2627550.14 | norm 0.3824 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24115 | train 2.928429 | val 3.293882 | tok/s 2625754.35 | norm 0.3630 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24116 | train 2.985436 | val 3.293882 | tok/s 2629065.67 | norm 0.3465 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24117 | train 2.963577 | val 3.293882 | tok/s 2619233.39 | norm 0.3985 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24118 | train 3.086873 | val 3.293882 | tok/s 2626983.13 | norm 0.4806 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24119 | train 2.905498 | val 3.293882 | tok/s 2621135.29 | norm 0.3473 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24120 | train 2.909790 | val 3.293882 | tok/s 2627491.34 | norm 0.3279 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24121 | train 2.969142 | val 3.293882 | tok/s 2617733.47 | norm 0.4046 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24122 | train 2.887210 | val 3.293882 | tok/s 2620990.66 | norm 0.3791 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24123 | train 3.033763 | val 3.293882 | tok/s 2624851.56 | norm 0.3768 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24124 | train 2.908579 | val 3.293882 | tok/s 2593696.94 | norm 0.4414 | dt 202.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24125 | train 2.921347 | val 3.293882 | tok/s 2627215.43 | norm 0.3579 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24126 | train 2.827442 | val 3.293882 | tok/s 2621127.12 | norm 0.3593 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24127 | train 2.832718 | val 3.293882 | tok/s 2621108.55 | norm 0.3664 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24128 | train 3.055864 | val 3.293882 | tok/s 2628101.84 | norm 0.3754 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24129 | train 2.919061 | val 3.293882 | tok/s 2626170.88 | norm 0.3469 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24130 | train 2.903570 | val 3.293882 | tok/s 2626549.28 | norm 0.3542 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24131 | train 2.933675 | val 3.293882 | tok/s 2620744.76 | norm 0.3814 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24132 | train 2.904533 | val 3.293882 | tok/s 2626368.50 | norm 0.3840 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24133 | train 2.893133 | val 3.293882 | tok/s 2492405.84 | norm 0.3670 | dt 210.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24134 | train 2.763347 | val 3.293882 | tok/s 2627766.56 | norm 0.3441 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24135 | train 2.848900 | val 3.293882 | tok/s 2630150.62 | norm 0.3439 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24136 | train 3.024810 | val 3.293882 | tok/s 2620663.04 | norm 0.3649 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24137 | train 3.006807 | val 3.293882 | tok/s 2625540.06 | norm 0.3501 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24138 | train 2.980598 | val 3.293882 | tok/s 2629190.31 | norm 0.3482 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24139 | train 2.854044 | val 3.293882 | tok/s 2627675.61 | norm 0.3319 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24140 | train 2.873550 | val 3.293882 | tok/s 2581738.39 | norm 0.3562 | dt 203.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24141 | train 2.768579 | val 3.293882 | tok/s 2627851.64 | norm 0.3958 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24142 | train 2.895811 | val 3.293882 | tok/s 2624192.50 | norm 0.3858 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24143 | train 2.944038 | val 3.293882 | tok/s 2632279.49 | norm 0.3618 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24144 | train 2.937978 | val 3.293882 | tok/s 2628483.53 | norm 0.3687 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24145 | train 2.931883 | val 3.293882 | tok/s 2626101.33 | norm 0.3741 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24146 | train 2.874072 | val 3.293882 | tok/s 2622891.09 | norm 0.3830 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24147 | train 3.175857 | val 3.293882 | tok/s 2619858.27 | norm 0.3871 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24148 | train 2.877812 | val 3.293882 | tok/s 2627919.37 | norm 0.3371 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24149 | train 2.814892 | val 3.293882 | tok/s 2621281.04 | norm 0.3443 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24150 | train 3.053601 | val 3.293882 | tok/s 2625855.74 | norm 0.3858 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24151 | train 2.891588 | val 3.293882 | tok/s 2623686.77 | norm 0.3251 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24152 | train 2.935933 | val 3.293882 | tok/s 2627425.46 | norm 0.3629 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24153 | train 2.856754 | val 3.293882 | tok/s 2629612.56 | norm 0.3908 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24154 | train 2.824648 | val 3.293882 | tok/s 2620049.79 | norm 0.3576 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24155 | train 2.923391 | val 3.293882 | tok/s 2628624.62 | norm 0.3686 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24156 | train 2.958113 | val 3.293882 | tok/s 2629271.79 | norm 0.3508 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24157 | train 2.985210 | val 3.293882 | tok/s 2632011.47 | norm 0.3519 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24158 | train 3.084996 | val 3.293882 | tok/s 2623941.19 | norm 0.3661 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24159 | train 2.751529 | val 3.293882 | tok/s 2621066.11 | norm 0.3131 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24160 | train 2.971485 | val 3.293882 | tok/s 2630345.42 | norm 0.3313 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24161 | train 2.928953 | val 3.293882 | tok/s 2624767.81 | norm 0.3675 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24162 | train 2.861832 | val 3.293882 | tok/s 2627526.88 | norm 0.4066 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24163 | train 2.931089 | val 3.293882 | tok/s 2627861.98 | norm 0.3321 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24164 | train 2.960764 | val 3.293882 | tok/s 2624454.38 | norm 0.3459 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24165 | train 3.032676 | val 3.293882 | tok/s 2628071.52 | norm 0.3824 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24166 | train 2.961333 | val 3.293882 | tok/s 2627500.11 | norm 0.3529 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24167 | train 2.913063 | val 3.293882 | tok/s 2627327.37 | norm 0.3369 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24168 | train 2.945462 | val 3.293882 | tok/s 2620003.39 | norm 0.3605 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24169 | train 2.916738 | val 3.293882 | tok/s 2623368.25 | norm 0.3462 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24170 | train 2.878964 | val 3.293882 | tok/s 2629854.26 | norm 0.3435 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24171 | train 2.871379 | val 3.293882 | tok/s 2626746.92 | norm 0.3579 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24172 | train 2.960045 | val 3.293882 | tok/s 2626560.64 | norm 0.3778 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24173 | train 2.906701 | val 3.293882 | tok/s 2615985.70 | norm 0.3913 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24174 | train 2.933748 | val 3.293882 | tok/s 2616159.79 | norm 0.3495 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24175 | train 2.928989 | val 3.293882 | tok/s 2631362.05 | norm 0.3638 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24176 | train 2.921247 | val 3.293882 | tok/s 2626418.47 | norm 0.3729 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24177 | train 3.104131 | val 3.293882 | tok/s 2627739.35 | norm 0.3537 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24178 | train 2.936959 | val 3.293882 | tok/s 2617844.36 | norm 0.3801 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24179 | train 2.916997 | val 3.293882 | tok/s 2444632.47 | norm 0.3291 | dt 214.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24180 | train 2.872798 | val 3.293882 | tok/s 2612499.26 | norm 0.3459 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24181 | train 3.112433 | val 3.293882 | tok/s 2628255.42 | norm 0.3638 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24182 | train 2.979322 | val 3.293882 | tok/s 2627587.95 | norm 0.3513 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24183 | train 2.886775 | val 3.293882 | tok/s 2621777.52 | norm 0.3516 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24184 | train 2.833055 | val 3.293882 | tok/s 2627359.92 | norm 0.3758 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24185 | train 2.874423 | val 3.293882 | tok/s 2631909.12 | norm 0.3650 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24186 | train 2.885452 | val 3.293882 | tok/s 2631017.15 | norm 0.3669 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24187 | train 2.955205 | val 3.293882 | tok/s 2513922.81 | norm 0.3377 | dt 208.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24188 | train 3.029756 | val 3.293882 | tok/s 2629667.11 | norm 0.3253 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24189 | train 2.934923 | val 3.293882 | tok/s 2629537.77 | norm 0.3175 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24190 | train 2.825679 | val 3.293882 | tok/s 2629790.33 | norm 0.3380 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24191 | train 2.957479 | val 3.293882 | tok/s 2621158.75 | norm 0.3768 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24192 | train 2.857533 | val 3.293882 | tok/s 2618931.60 | norm 0.3704 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24193 | train 2.925267 | val 3.293882 | tok/s 2623579.29 | norm 0.3521 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24194 | train 2.968771 | val 3.293882 | tok/s 2626205.18 | norm 0.3247 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24195 | train 2.940644 | val 3.293882 | tok/s 2629814.10 | norm 0.3079 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24196 | train 2.916633 | val 3.293882 | tok/s 2616522.85 | norm 0.3303 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24197 | train 2.930386 | val 3.293882 | tok/s 2629904.33 | norm 0.3336 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24198 | train 2.972409 | val 3.293882 | tok/s 2625161.97 | norm 0.3396 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24199 | train 2.881568 | val 3.293882 | tok/s 2622667.28 | norm 0.3638 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24200 | train 2.846165 | val 3.293882 | tok/s 2619718.22 | norm 0.3140 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24201 | train 2.903218 | val 3.293882 | tok/s 2621245.23 | norm 0.3289 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24202 | train 3.057590 | val 3.293882 | tok/s 2625306.58 | norm 0.3573 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24203 | train 2.967143 | val 3.293882 | tok/s 2625169.17 | norm 0.3370 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24204 | train 2.922160 | val 3.293882 | tok/s 2625924.45 | norm 0.3560 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24205 | train 3.060338 | val 3.293882 | tok/s 2621208.85 | norm 0.3738 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24206 | train 2.892217 | val 3.293882 | tok/s 2619822.60 | norm 0.3580 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24207 | train 3.033605 | val 3.293882 | tok/s 2627965.87 | norm 0.3750 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24208 | train 2.954422 | val 3.293882 | tok/s 2628070.10 | norm 0.3618 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24209 | train 2.902576 | val 3.293882 | tok/s 2627671.05 | norm 0.3433 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24210 | train 2.982976 | val 3.293882 | tok/s 2623262.76 | norm 0.3654 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24211 | train 3.032178 | val 3.293882 | tok/s 2624722.28 | norm 0.3636 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24212 | train 2.926175 | val 3.293882 | tok/s 2628707.79 | norm 0.3386 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24213 | train 2.936708 | val 3.293882 | tok/s 2627093.56 | norm 0.3639 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24214 | train 2.852263 | val 3.293882 | tok/s 2625435.10 | norm 0.3945 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24215 | train 2.835588 | val 3.293882 | tok/s 2620452.12 | norm 0.3567 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24216 | train 3.059078 | val 3.293882 | tok/s 2615131.90 | norm 0.3440 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24217 | train 3.139819 | val 3.293882 | tok/s 2627674.21 | norm 0.4152 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24218 | train 2.904384 | val 3.293882 | tok/s 2624489.17 | norm 0.3759 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24219 | train 3.221740 | val 3.293882 | tok/s 2614430.67 | norm 0.3710 | dt 200.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24220 | train 2.970006 | val 3.293882 | tok/s 2556713.62 | norm 0.3629 | dt 205.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24221 | train 2.877811 | val 3.293882 | tok/s 2623919.70 | norm 0.3739 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24222 | train 2.919070 | val 3.293882 | tok/s 2631115.86 | norm 0.3943 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24223 | train 2.986687 | val 3.293882 | tok/s 2625292.20 | norm 0.3562 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24224 | train 2.927130 | val 3.293882 | tok/s 2625202.37 | norm 0.3684 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24225 | train 2.923843 | val 3.293882 | tok/s 2618935.42 | norm 0.3788 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24226 | train 2.935296 | val 3.293882 | tok/s 2492656.62 | norm 0.3954 | dt 210.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24227 | train 2.978495 | val 3.293882 | tok/s 2628022.10 | norm 0.4076 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24228 | train 2.890834 | val 3.293882 | tok/s 2629812.64 | norm 0.3832 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24229 | train 2.797073 | val 3.293882 | tok/s 2626752.33 | norm 0.3966 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24230 | train 2.916148 | val 3.293882 | tok/s 2624525.79 | norm 0.3640 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24231 | train 3.032628 | val 3.293882 | tok/s 2625317.53 | norm 0.3412 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24232 | train 2.844100 | val 3.293882 | tok/s 2628573.37 | norm 0.3503 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24233 | train 2.886767 | val 3.293882 | tok/s 2629259.26 | norm 0.3232 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24234 | train 3.057816 | val 3.293882 | tok/s 2629149.26 | norm 0.3747 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24235 | train 2.879864 | val 3.293882 | tok/s 2608640.91 | norm 0.3817 | dt 200.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24236 | train 2.975891 | val 3.293882 | tok/s 2629195.09 | norm 0.3636 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24237 | train 2.919722 | val 3.293882 | tok/s 2597651.44 | norm 0.3313 | dt 201.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24238 | train 2.898263 | val 3.293882 | tok/s 2630531.80 | norm 0.3641 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24239 | train 2.936457 | val 3.293882 | tok/s 2628317.14 | norm 0.4077 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24240 | train 2.881288 | val 3.293882 | tok/s 2628392.06 | norm 0.3938 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24241 | train 2.898913 | val 3.293882 | tok/s 2630769.35 | norm 0.4150 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24242 | train 2.952641 | val 3.293882 | tok/s 2631101.96 | norm 0.4120 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24243 | train 3.008825 | val 3.293882 | tok/s 2622312.21 | norm 0.3822 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24244 | train 2.858187 | val 3.293882 | tok/s 2631624.66 | norm 0.3683 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24245 | train 2.998299 | val 3.293882 | tok/s 2625094.63 | norm 0.3640 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24246 | train 2.956691 | val 3.293882 | tok/s 2616387.06 | norm 0.3795 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24247 | train 3.101115 | val 3.293882 | tok/s 2626960.23 | norm 0.3845 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24248 | train 2.913987 | val 3.293882 | tok/s 2625273.06 | norm 0.3451 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24249 | train 2.906006 | val 3.293882 | tok/s 2628466.71 | norm 0.3759 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 24250 | val_loss 3.295383 step 24250 | train 2.853797 | val 3.295383 | tok/s 2611044.48 | norm 0.3995 | dt 200.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24251 | train 2.966205 | val 3.295383 | tok/s 2619589.65 | norm 0.3802 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24252 | train 2.997056 | val 3.295383 | tok/s 2628982.93 | norm 0.3543 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24253 | train 2.848662 | val 3.295383 | tok/s 2624544.27 | norm 0.3817 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24254 | train 2.983222 | val 3.295383 | tok/s 2620363.19 | norm 0.4041 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24255 | train 3.009521 | val 3.295383 | tok/s 2627204.85 | norm 0.3325 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24256 | train 2.859138 | val 3.295383 | tok/s 2628821.72 | norm 0.3360 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24257 | train 2.963258 | val 3.295383 | tok/s 2627479.83 | norm 0.3426 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24258 | train 2.944905 | val 3.295383 | tok/s 2624819.27 | norm 0.3430 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24259 | train 2.922483 | val 3.295383 | tok/s 2619404.57 | norm 0.3413 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24260 | train 2.947647 | val 3.295383 | tok/s 2633420.01 | norm 0.3685 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24261 | train 2.965213 | val 3.295383 | tok/s 2626189.66 | norm 0.3591 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24262 | train 2.995885 | val 3.295383 | tok/s 2629675.86 | norm 0.3654 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24263 | train 2.956093 | val 3.295383 | tok/s 2627194.48 | norm 0.3731 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24264 | train 2.853063 | val 3.295383 | tok/s 2620073.61 | norm 0.3816 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24265 | train 3.001721 | val 3.295383 | tok/s 2629716.60 | norm 0.3987 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24266 | train 3.017407 | val 3.295383 | tok/s 2624963.75 | norm 0.4197 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24267 | train 3.032748 | val 3.295383 | tok/s 2629731.73 | norm 0.3560 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24268 | train 2.962794 | val 3.295383 | tok/s 2616268.87 | norm 0.3666 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24269 | train 2.846894 | val 3.295383 | tok/s 2621218.02 | norm 0.3801 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24270 | train 2.950734 | val 3.295383 | tok/s 2629827.73 | norm 0.3796 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24271 | train 2.922469 | val 3.295383 | tok/s 2629007.65 | norm 0.3616 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24272 | train 3.017379 | val 3.295383 | tok/s 2626639.05 | norm 0.4021 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24273 | train 3.106627 | val 3.295383 | tok/s 2624318.35 | norm 0.4640 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24274 | train 2.914932 | val 3.295383 | tok/s 2625898.68 | norm 0.3568 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24275 | train 2.958101 | val 3.295383 | tok/s 2627314.44 | norm 0.3064 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24276 | train 2.916455 | val 3.295383 | tok/s 2625230.32 | norm 0.3392 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24277 | train 3.002464 | val 3.295383 | tok/s 2629234.21 | norm 0.3483 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24278 | train 2.879859 | val 3.295383 | tok/s 2576944.26 | norm 0.3502 | dt 203.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24279 | train 2.955948 | val 3.295383 | tok/s 2607454.25 | norm 0.3497 | dt 201.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24280 | train 3.051231 | val 3.295383 | tok/s 2566825.48 | norm 0.4052 | dt 204.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24281 | train 2.876346 | val 3.295383 | tok/s 2626988.46 | norm 0.3714 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24282 | train 2.812682 | val 3.295383 | tok/s 2626764.60 | norm 0.3573 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24283 | train 2.830696 | val 3.295383 | tok/s 2620075.23 | norm 0.3790 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24284 | train 2.957793 | val 3.295383 | tok/s 2625385.04 | norm 0.3959 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24285 | train 2.920371 | val 3.295383 | tok/s 2632323.46 | norm 0.3647 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24286 | train 3.069993 | val 3.295383 | tok/s 2625171.55 | norm 0.3543 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24287 | train 2.912736 | val 3.295383 | tok/s 2629701.87 | norm 0.3899 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24288 | train 2.780635 | val 3.295383 | tok/s 2625524.62 | norm 0.3787 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24289 | train 2.883448 | val 3.295383 | tok/s 2627685.68 | norm 0.3586 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24290 | train 2.987372 | val 3.295383 | tok/s 2630478.97 | norm 0.3589 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24291 | train 2.913709 | val 3.295383 | tok/s 2622628.13 | norm 0.3268 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24292 | train 2.942045 | val 3.295383 | tok/s 2625295.52 | norm 0.3361 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24293 | train 2.972993 | val 3.295383 | tok/s 2542746.28 | norm 0.3506 | dt 206.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24294 | train 2.765061 | val 3.295383 | tok/s 2585729.24 | norm 0.3945 | dt 202.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24295 | train 3.038922 | val 3.295383 | tok/s 2628441.67 | norm 0.3657 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24296 | train 3.137044 | val 3.295383 | tok/s 2625284.41 | norm 0.3773 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24297 | train 3.032906 | val 3.295383 | tok/s 2628554.68 | norm 0.3481 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24298 | train 2.960519 | val 3.295383 | tok/s 2626591.67 | norm 0.3817 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24299 | train 2.960274 | val 3.295383 | tok/s 2627349.66 | norm 0.4023 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24300 | train 2.999654 | val 3.295383 | tok/s 2623732.60 | norm 0.4096 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24301 | train 2.891006 | val 3.295383 | tok/s 2618730.77 | norm 0.3683 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24302 | train 2.874347 | val 3.295383 | tok/s 2629099.20 | norm 0.3626 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24303 | train 2.991559 | val 3.295383 | tok/s 2624413.87 | norm 0.3450 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24304 | train 2.889600 | val 3.295383 | tok/s 2630956.25 | norm 0.3746 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24305 | train 2.826817 | val 3.295383 | tok/s 2625866.12 | norm 0.3716 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24306 | train 2.894309 | val 3.295383 | tok/s 2622123.18 | norm 0.3394 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24307 | train 2.931724 | val 3.295383 | tok/s 2626988.24 | norm 0.3374 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24308 | train 3.073708 | val 3.295383 | tok/s 2626737.46 | norm 0.3720 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24309 | train 3.000255 | val 3.295383 | tok/s 2628642.30 | norm 0.3481 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24310 | train 2.982927 | val 3.295383 | tok/s 2630729.63 | norm 0.3454 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24311 | train 2.921152 | val 3.295383 | tok/s 2627989.57 | norm 0.3877 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24312 | train 2.995287 | val 3.295383 | tok/s 2629670.82 | norm 0.3840 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24313 | train 2.891313 | val 3.295383 | tok/s 2629921.96 | norm 0.3304 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24314 | train 2.979892 | val 3.295383 | tok/s 2629049.09 | norm 0.3715 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24315 | train 3.034362 | val 3.295383 | tok/s 2621039.95 | norm 0.4044 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24316 | train 2.998063 | val 3.295383 | tok/s 2627868.84 | norm 0.3063 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24317 | train 2.977854 | val 3.295383 | tok/s 2633288.08 | norm 0.3078 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24318 | train 2.873146 | val 3.295383 | tok/s 2625046.10 | norm 0.3203 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24319 | train 2.936845 | val 3.295383 | tok/s 2630300.84 | norm 0.3501 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24320 | train 2.892929 | val 3.295383 | tok/s 2616784.96 | norm 0.3887 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24321 | train 2.873119 | val 3.295383 | tok/s 2622878.54 | norm 0.4178 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24322 | train 2.956921 | val 3.295383 | tok/s 2631128.87 | norm 0.4268 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24323 | train 2.847361 | val 3.295383 | tok/s 2624992.98 | norm 0.3710 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24324 | train 2.980280 | val 3.295383 | tok/s 2626052.86 | norm 0.3647 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24325 | train 2.961413 | val 3.295383 | tok/s 2621502.97 | norm 0.3534 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24326 | train 3.020112 | val 3.295383 | tok/s 2627302.63 | norm 0.3528 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24327 | train 2.893296 | val 3.295383 | tok/s 2634476.48 | norm 0.3566 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24328 | train 2.999331 | val 3.295383 | tok/s 2627843.34 | norm 0.3689 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24329 | train 2.915331 | val 3.295383 | tok/s 2631848.04 | norm 0.3918 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24330 | train 3.060021 | val 3.295383 | tok/s 2622864.77 | norm 0.4006 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24331 | train 2.953997 | val 3.295383 | tok/s 2624275.08 | norm 0.3580 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24332 | train 2.905846 | val 3.295383 | tok/s 2628034.18 | norm 0.3705 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24333 | train 2.919929 | val 3.295383 | tok/s 2629714.90 | norm 0.4024 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24334 | train 2.947289 | val 3.295383 | tok/s 2629006.07 | norm 0.3892 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24335 | train 3.058258 | val 3.295383 | tok/s 2625225.25 | norm 0.3787 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24336 | train 2.995031 | val 3.295383 | tok/s 2625985.16 | norm 0.3509 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24337 | train 2.953531 | val 3.295383 | tok/s 2627747.20 | norm 0.3720 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24338 | train 2.853688 | val 3.295383 | tok/s 2629284.73 | norm 0.3661 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24339 | train 2.835873 | val 3.295383 | tok/s 2629217.09 | norm 0.3539 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24340 | train 3.049524 | val 3.295383 | tok/s 2622750.34 | norm 0.3556 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24341 | train 3.063802 | val 3.295383 | tok/s 2624799.97 | norm 0.3697 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24342 | train 2.883085 | val 3.295383 | tok/s 2631685.45 | norm 0.3881 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24343 | train 2.774261 | val 3.295383 | tok/s 2629485.92 | norm 0.3876 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24344 | train 3.089847 | val 3.295383 | tok/s 2630974.63 | norm 0.3736 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24345 | train 2.876836 | val 3.295383 | tok/s 2626958.53 | norm 0.3915 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24346 | train 2.937434 | val 3.295383 | tok/s 2626526.60 | norm 0.4079 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24347 | train 2.949627 | val 3.295383 | tok/s 2632033.54 | norm 0.3903 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24348 | train 2.903686 | val 3.295383 | tok/s 2625689.97 | norm 0.3956 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24349 | train 2.937770 | val 3.295383 | tok/s 2619179.41 | norm 0.4054 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24350 | train 3.082921 | val 3.295383 | tok/s 2612827.73 | norm 0.3556 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24351 | train 2.879241 | val 3.295383 | tok/s 2624123.93 | norm 0.3369 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24352 | train 2.796592 | val 3.295383 | tok/s 2631730.74 | norm 0.3282 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24353 | train 3.056696 | val 3.295383 | tok/s 2523589.58 | norm 0.3402 | dt 207.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24354 | train 3.032210 | val 3.295383 | tok/s 2627366.77 | norm 0.3441 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24355 | train 2.745720 | val 3.295383 | tok/s 2624067.60 | norm 0.3599 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24356 | train 2.927492 | val 3.295383 | tok/s 2628354.36 | norm 0.3720 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24357 | train 2.908823 | val 3.295383 | tok/s 2621353.43 | norm 0.3738 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24358 | train 2.949899 | val 3.295383 | tok/s 2607305.91 | norm 0.3832 | dt 201.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24359 | train 2.942738 | val 3.295383 | tok/s 2632220.67 | norm 0.3826 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24360 | train 2.857660 | val 3.295383 | tok/s 2629260.54 | norm 0.3446 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24361 | train 2.984779 | val 3.295383 | tok/s 2626075.47 | norm 0.3473 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24362 | train 2.641793 | val 3.295383 | tok/s 2628824.93 | norm 0.3940 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24363 | train 2.965461 | val 3.295383 | tok/s 2620637.93 | norm 0.3760 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24364 | train 2.779441 | val 3.295383 | tok/s 2629871.22 | norm 0.3374 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24365 | train 2.907974 | val 3.295383 | tok/s 2627304.21 | norm 0.3413 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24366 | train 2.893980 | val 3.295383 | tok/s 2628134.77 | norm 0.3833 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24367 | train 2.963139 | val 3.295383 | tok/s 2629318.75 | norm 0.3903 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24368 | train 2.949956 | val 3.295383 | tok/s 2624790.54 | norm 0.3853 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24369 | train 2.950302 | val 3.295383 | tok/s 2629864.51 | norm 0.3912 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24370 | train 2.993117 | val 3.295383 | tok/s 2628263.23 | norm 0.3761 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24371 | train 2.949376 | val 3.295383 | tok/s 2625172.36 | norm 0.3711 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24372 | train 2.888717 | val 3.295383 | tok/s 2571514.77 | norm 0.3553 | dt 203.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24373 | train 3.018559 | val 3.295383 | tok/s 2622513.37 | norm 0.3893 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24374 | train 2.962953 | val 3.295383 | tok/s 2631529.67 | norm 0.3608 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24375 | train 2.944016 | val 3.295383 | tok/s 2632254.45 | norm 0.3640 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24376 | train 2.843923 | val 3.295383 | tok/s 2633782.34 | norm 0.3726 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24377 | train 2.946416 | val 3.295383 | tok/s 2625751.77 | norm 0.3510 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24378 | train 3.046129 | val 3.295383 | tok/s 2624072.47 | norm 0.3589 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24379 | train 2.889708 | val 3.295383 | tok/s 2630106.82 | norm 0.3735 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24380 | train 2.897970 | val 3.295383 | tok/s 2627766.60 | norm 0.3433 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24381 | train 2.841136 | val 3.295383 | tok/s 2627676.17 | norm 0.3643 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24382 | train 2.874410 | val 3.295383 | tok/s 2619407.21 | norm 0.3958 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24383 | train 2.996604 | val 3.295383 | tok/s 2625065.48 | norm 0.3823 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24384 | train 2.990881 | val 3.295383 | tok/s 2630828.47 | norm 0.3784 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24385 | train 3.000869 | val 3.295383 | tok/s 2625996.26 | norm 0.3789 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24386 | train 2.866291 | val 3.295383 | tok/s 2628173.14 | norm 0.3659 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24387 | train 2.985847 | val 3.295383 | tok/s 2623299.11 | norm 0.3494 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24388 | train 2.936717 | val 3.295383 | tok/s 2499700.57 | norm 0.3660 | dt 209.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24389 | train 3.038896 | val 3.295383 | tok/s 2482604.31 | norm 0.3486 | dt 211.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24390 | train 2.875080 | val 3.295383 | tok/s 2623265.94 | norm 0.3616 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24391 | train 2.913318 | val 3.295383 | tok/s 2630340.51 | norm 0.3501 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24392 | train 2.927064 | val 3.295383 | tok/s 2630102.28 | norm 0.3394 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24393 | train 2.955352 | val 3.295383 | tok/s 2629553.33 | norm 0.3846 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24394 | train 2.970587 | val 3.295383 | tok/s 2631576.42 | norm 0.3210 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24395 | train 2.863304 | val 3.295383 | tok/s 2618901.22 | norm 0.3195 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24396 | train 2.826442 | val 3.295383 | tok/s 2613793.63 | norm 0.3639 | dt 200.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24397 | train 3.124434 | val 3.295383 | tok/s 2622559.33 | norm 0.3962 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24398 | train 2.951851 | val 3.295383 | tok/s 2627316.03 | norm 0.3388 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24399 | train 3.092089 | val 3.295383 | tok/s 2505814.04 | norm 0.3575 | dt 209.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24400 | train 2.887188 | val 3.295383 | tok/s 2622149.97 | norm 0.3673 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24401 | train 2.766577 | val 3.295383 | tok/s 2630715.69 | norm 0.3672 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24402 | train 2.963850 | val 3.295383 | tok/s 2625278.17 | norm 0.3739 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24403 | train 2.894578 | val 3.295383 | tok/s 2625485.87 | norm 0.3743 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24404 | train 2.886105 | val 3.295383 | tok/s 2627833.80 | norm 0.3688 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24405 | train 2.926653 | val 3.295383 | tok/s 2623492.36 | norm 0.4041 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24406 | train 2.901002 | val 3.295383 | tok/s 2630374.17 | norm 0.3672 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24407 | train 2.855588 | val 3.295383 | tok/s 2628623.04 | norm 0.3372 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24408 | train 3.091801 | val 3.295383 | tok/s 2627617.29 | norm 0.4372 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24409 | train 2.832681 | val 3.295383 | tok/s 2615456.99 | norm 0.3253 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24410 | train 2.944126 | val 3.295383 | tok/s 2622509.07 | norm 0.3664 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24411 | train 2.891591 | val 3.295383 | tok/s 2633254.56 | norm 0.3593 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24412 | train 2.905237 | val 3.295383 | tok/s 2626061.19 | norm 0.3627 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24413 | train 2.916087 | val 3.295383 | tok/s 2627591.88 | norm 0.3481 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24414 | train 2.909010 | val 3.295383 | tok/s 2619791.67 | norm 0.3182 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24415 | train 2.967118 | val 3.295383 | tok/s 2620619.03 | norm 0.3436 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24416 | train 2.840983 | val 3.295383 | tok/s 2632878.49 | norm 0.3471 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24417 | train 2.970459 | val 3.295383 | tok/s 2632435.68 | norm 0.3488 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24418 | train 2.902457 | val 3.295383 | tok/s 2629775.72 | norm 0.3596 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24419 | train 2.977926 | val 3.295383 | tok/s 2621438.34 | norm 0.3621 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24420 | train 2.916237 | val 3.295383 | tok/s 2624750.52 | norm 0.3654 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24421 | train 2.993513 | val 3.295383 | tok/s 2627746.21 | norm 0.3577 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24422 | train 2.837797 | val 3.295383 | tok/s 2626362.50 | norm 0.3725 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24423 | train 2.898085 | val 3.295383 | tok/s 2628740.80 | norm 0.3655 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24424 | train 2.917582 | val 3.295383 | tok/s 2624669.35 | norm 0.3456 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24425 | train 3.072793 | val 3.295383 | tok/s 2594618.47 | norm 0.3472 | dt 202.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24426 | train 2.945784 | val 3.295383 | tok/s 2629029.31 | norm 0.3981 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24427 | train 3.063896 | val 3.295383 | tok/s 2494038.04 | norm 0.3812 | dt 210.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24428 | train 2.967999 | val 3.295383 | tok/s 2623258.05 | norm 0.3437 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24429 | train 2.906255 | val 3.295383 | tok/s 2626453.39 | norm 0.3503 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24430 | train 3.068371 | val 3.295383 | tok/s 2624296.93 | norm 0.4363 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24431 | train 2.859092 | val 3.295383 | tok/s 2631109.33 | norm 0.3812 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24432 | train 2.814553 | val 3.295383 | tok/s 2623179.27 | norm 0.3641 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24433 | train 2.984874 | val 3.295383 | tok/s 2630292.37 | norm 0.3824 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24434 | train 2.837603 | val 3.295383 | tok/s 2628430.31 | norm 0.3972 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24435 | train 2.951943 | val 3.295383 | tok/s 2629105.66 | norm 0.3991 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24436 | train 3.019777 | val 3.295383 | tok/s 2576981.34 | norm 0.3632 | dt 203.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24437 | train 2.937876 | val 3.295383 | tok/s 2597592.32 | norm 0.3922 | dt 201.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24438 | train 2.940397 | val 3.295383 | tok/s 2628722.19 | norm 0.3795 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24439 | train 2.946939 | val 3.295383 | tok/s 2624147.32 | norm 0.3774 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24440 | train 2.844130 | val 3.295383 | tok/s 2629359.00 | norm 0.3561 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24441 | train 3.190336 | val 3.295383 | tok/s 2624807.52 | norm 0.4002 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24442 | train 2.965078 | val 3.295383 | tok/s 2624041.82 | norm 0.3867 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24443 | train 2.898058 | val 3.295383 | tok/s 2628332.70 | norm 0.3168 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24444 | train 2.821363 | val 3.295383 | tok/s 2626632.04 | norm 0.3540 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24445 | train 2.836902 | val 3.295383 | tok/s 2626981.06 | norm 0.3840 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24446 | train 2.953118 | val 3.295383 | tok/s 2620345.63 | norm 0.3653 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24447 | train 2.969580 | val 3.295383 | tok/s 2623567.20 | norm 0.3426 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24448 | train 2.820543 | val 3.295383 | tok/s 2632988.58 | norm 0.3390 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24449 | train 3.027545 | val 3.295383 | tok/s 2633483.00 | norm 0.3581 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24450 | train 2.872953 | val 3.295383 | tok/s 2630929.87 | norm 0.3331 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24451 | train 2.866397 | val 3.295383 | tok/s 2620020.03 | norm 0.3698 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24452 | train 2.932881 | val 3.295383 | tok/s 2622470.34 | norm 0.4501 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24453 | train 2.974849 | val 3.295383 | tok/s 2628402.08 | norm 0.3355 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24454 | train 2.884243 | val 3.295383 | tok/s 2624957.93 | norm 0.3766 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24455 | train 2.822889 | val 3.295383 | tok/s 2631791.57 | norm 0.3701 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24456 | train 2.891902 | val 3.295383 | tok/s 2618437.93 | norm 0.3327 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24457 | train 2.855233 | val 3.295383 | tok/s 2622746.99 | norm 0.3373 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24458 | train 2.942321 | val 3.295383 | tok/s 2615837.63 | norm 0.3353 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24459 | train 2.795125 | val 3.295383 | tok/s 2625750.71 | norm 0.3433 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24460 | train 2.895385 | val 3.295383 | tok/s 2629156.36 | norm 0.3441 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24461 | train 3.043736 | val 3.295383 | tok/s 2621202.65 | norm 0.3834 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24462 | train 3.006968 | val 3.295383 | tok/s 2624823.02 | norm 0.3941 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24463 | train 2.972442 | val 3.295383 | tok/s 2632180.07 | norm 0.3698 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24464 | train 2.947487 | val 3.295383 | tok/s 2631915.22 | norm 0.3833 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24465 | train 2.928640 | val 3.295383 | tok/s 2629917.13 | norm 0.3648 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24466 | train 2.893157 | val 3.295383 | tok/s 2624104.66 | norm 0.3460 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24467 | train 2.929524 | val 3.295383 | tok/s 2629393.31 | norm 0.3668 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24468 | train 3.049190 | val 3.295383 | tok/s 2633789.75 | norm 0.3596 | dt 199.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24469 | train 3.069981 | val 3.295383 | tok/s 2627739.11 | norm 0.3321 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24470 | train 2.961761 | val 3.295383 | tok/s 2631159.08 | norm 0.3571 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24471 | train 2.873473 | val 3.295383 | tok/s 2629116.47 | norm 0.3944 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24472 | train 2.885762 | val 3.295383 | tok/s 2626478.31 | norm 0.3739 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24473 | train 2.807856 | val 3.295383 | tok/s 2627834.89 | norm 0.3659 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24474 | train 3.025111 | val 3.295383 | tok/s 2627329.58 | norm 0.3496 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24475 | train 2.828336 | val 3.295383 | tok/s 2623982.10 | norm 0.3355 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24476 | train 2.951866 | val 3.295383 | tok/s 2476532.17 | norm 0.3540 | dt 211.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24477 | train 2.894815 | val 3.295383 | tok/s 2625989.49 | norm 0.3860 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24478 | train 3.012004 | val 3.295383 | tok/s 2621719.92 | norm 0.3838 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24479 | train 2.847825 | val 3.295383 | tok/s 2617933.04 | norm 0.3628 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24480 | train 2.998883 | val 3.295383 | tok/s 2626415.47 | norm 0.3380 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24481 | train 2.867611 | val 3.295383 | tok/s 2626451.24 | norm 0.3561 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24482 | train 2.858147 | val 3.295383 | tok/s 2629648.21 | norm 0.3509 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24483 | train 3.050027 | val 3.295383 | tok/s 2626211.23 | norm 0.3598 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24484 | train 3.010111 | val 3.295383 | tok/s 2619406.29 | norm 0.3944 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24485 | train 2.944951 | val 3.295383 | tok/s 2631007.98 | norm 0.3680 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24486 | train 2.823046 | val 3.295383 | tok/s 2625841.57 | norm 0.3617 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24487 | train 2.825340 | val 3.295383 | tok/s 2624785.84 | norm 0.3636 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24488 | train 2.925919 | val 3.295383 | tok/s 2584847.78 | norm 0.3669 | dt 202.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24489 | train 2.935022 | val 3.295383 | tok/s 2619466.55 | norm 0.3956 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24490 | train 2.949394 | val 3.295383 | tok/s 2628742.13 | norm 0.4043 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24491 | train 2.931836 | val 3.295383 | tok/s 2627876.33 | norm 0.4210 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24492 | train 2.983415 | val 3.295383 | tok/s 2627550.29 | norm 0.4040 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24493 | train 2.999416 | val 3.295383 | tok/s 2606009.67 | norm 0.3822 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24494 | train 2.837275 | val 3.295383 | tok/s 2623491.01 | norm 0.3834 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24495 | train 2.852731 | val 3.295383 | tok/s 2624406.87 | norm 0.3830 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24496 | train 2.857265 | val 3.295383 | tok/s 2630626.94 | norm 0.3733 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24497 | train 2.804626 | val 3.295383 | tok/s 2622760.25 | norm 0.3725 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24498 | train 2.987258 | val 3.295383 | tok/s 2619417.05 | norm 0.4289 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24499 | train 2.958381 | val 3.295383 | tok/s 2621067.40 | norm 0.3682 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 24500 | val_loss 3.304849 step 24500 | train 2.911670 | val 3.304849 | tok/s 2620432.86 | norm 0.3623 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24501 | train 3.045890 | val 3.304849 | tok/s 2628893.82 | norm 0.4318 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24502 | train 2.888955 | val 3.304849 | tok/s 2626103.62 | norm 0.3292 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24503 | train 2.880797 | val 3.304849 | tok/s 2626724.01 | norm 0.3327 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24504 | train 2.956136 | val 3.304849 | tok/s 2629839.35 | norm 0.3861 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24505 | train 2.842754 | val 3.304849 | tok/s 2623464.66 | norm 0.3817 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24506 | train 2.808108 | val 3.304849 | tok/s 2630595.68 | norm 0.3499 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24507 | train 3.016961 | val 3.304849 | tok/s 2626459.24 | norm 0.3583 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24508 | train 3.184651 | val 3.304849 | tok/s 2575786.64 | norm 0.4241 | dt 203.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24509 | train 3.081412 | val 3.304849 | tok/s 2615942.35 | norm 0.3472 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24510 | train 2.907652 | val 3.304849 | tok/s 2624981.97 | norm 0.3300 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24511 | train 3.163881 | val 3.304849 | tok/s 2630001.56 | norm 0.5040 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24512 | train 2.931630 | val 3.304849 | tok/s 2628159.06 | norm 0.3748 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24513 | train 2.913501 | val 3.304849 | tok/s 2629476.21 | norm 0.3519 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24514 | train 2.875949 | val 3.304849 | tok/s 2502851.15 | norm 0.3639 | dt 209.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24515 | train 2.902646 | val 3.304849 | tok/s 2619720.60 | norm 0.3509 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24516 | train 2.899735 | val 3.304849 | tok/s 2484653.92 | norm 0.3540 | dt 211.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24517 | train 2.926473 | val 3.304849 | tok/s 2621851.77 | norm 0.3362 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24518 | train 2.891747 | val 3.304849 | tok/s 2628284.95 | norm 0.3680 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24519 | train 2.914484 | val 3.304849 | tok/s 2623988.59 | norm 0.3589 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24520 | train 2.974687 | val 3.304849 | tok/s 2627905.57 | norm 0.3461 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24521 | train 2.915707 | val 3.304849 | tok/s 2623947.08 | norm 0.3510 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24522 | train 2.945183 | val 3.304849 | tok/s 2619392.70 | norm 0.3532 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24523 | train 3.029828 | val 3.304849 | tok/s 2607706.71 | norm 0.3779 | dt 201.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24524 | train 2.907428 | val 3.304849 | tok/s 2619988.45 | norm 0.3270 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24525 | train 2.941415 | val 3.304849 | tok/s 2631122.23 | norm 0.3067 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24526 | train 3.163709 | val 3.304849 | tok/s 2617780.50 | norm 0.3492 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24527 | train 2.817715 | val 3.304849 | tok/s 2600478.04 | norm 0.3399 | dt 201.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24528 | train 2.982719 | val 3.304849 | tok/s 2624726.97 | norm 0.3198 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24529 | train 2.998486 | val 3.304849 | tok/s 2626113.39 | norm 0.3124 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24530 | train 3.012095 | val 3.304849 | tok/s 2624162.41 | norm 0.3213 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24531 | train 2.923489 | val 3.304849 | tok/s 2606117.19 | norm 0.3129 | dt 201.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24532 | train 2.764912 | val 3.304849 | tok/s 2622364.57 | norm 0.3485 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24533 | train 2.943905 | val 3.304849 | tok/s 2626682.71 | norm 0.3976 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24534 | train 2.991146 | val 3.304849 | tok/s 2628839.88 | norm 0.3609 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24535 | train 2.927183 | val 3.304849 | tok/s 2625254.63 | norm 0.3571 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24536 | train 3.057153 | val 3.304849 | tok/s 2620206.29 | norm 0.3849 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24537 | train 2.902885 | val 3.304849 | tok/s 2620886.87 | norm 0.3443 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24538 | train 2.993383 | val 3.304849 | tok/s 2628680.92 | norm 0.3444 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24539 | train 2.984567 | val 3.304849 | tok/s 2625753.31 | norm 0.3635 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24540 | train 2.988691 | val 3.304849 | tok/s 2625768.79 | norm 0.3528 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24541 | train 3.024272 | val 3.304849 | tok/s 2623206.81 | norm 0.3643 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24542 | train 3.199879 | val 3.304849 | tok/s 2623853.47 | norm 0.3521 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24543 | train 2.963404 | val 3.304849 | tok/s 2630829.73 | norm 0.3691 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24544 | train 3.021446 | val 3.304849 | tok/s 2625369.28 | norm 0.3665 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24545 | train 2.906635 | val 3.304849 | tok/s 2628636.50 | norm 0.3935 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24546 | train 2.945667 | val 3.304849 | tok/s 2623369.36 | norm 0.3683 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24547 | train 2.967454 | val 3.304849 | tok/s 2623159.57 | norm 0.3740 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24548 | train 2.849303 | val 3.304849 | tok/s 2628135.02 | norm 0.3562 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24549 | train 2.866537 | val 3.304849 | tok/s 2631381.85 | norm 0.3351 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24550 | train 2.962714 | val 3.304849 | tok/s 2630960.41 | norm 0.3360 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24551 | train 2.855847 | val 3.304849 | tok/s 2625856.03 | norm 0.3496 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24552 | train 2.990159 | val 3.304849 | tok/s 2629092.75 | norm 0.3329 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24553 | train 2.842897 | val 3.304849 | tok/s 2630502.17 | norm 0.3334 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24554 | train 2.913296 | val 3.304849 | tok/s 2625543.04 | norm 0.4167 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24555 | train 2.853390 | val 3.304849 | tok/s 2629019.73 | norm 0.3486 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24556 | train 2.862651 | val 3.304849 | tok/s 2625580.32 | norm 0.3796 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24557 | train 2.826091 | val 3.304849 | tok/s 2624967.08 | norm 0.3863 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24558 | train 2.980936 | val 3.304849 | tok/s 2522185.88 | norm 0.3751 | dt 207.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24559 | train 2.799798 | val 3.304849 | tok/s 2600059.64 | norm 0.3707 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24560 | train 3.061500 | val 3.304849 | tok/s 2584046.58 | norm 0.3814 | dt 202.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24561 | train 2.907599 | val 3.304849 | tok/s 2624258.56 | norm 0.3387 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24562 | train 2.879567 | val 3.304849 | tok/s 2628717.43 | norm 0.4861 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24563 | train 3.209681 | val 3.304849 | tok/s 2621660.14 | norm 0.4560 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24564 | train 2.885849 | val 3.304849 | tok/s 2623328.13 | norm 0.3522 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24565 | train 2.834766 | val 3.304849 | tok/s 2632312.53 | norm 0.3001 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24566 | train 3.050414 | val 3.304849 | tok/s 2628201.77 | norm 0.3485 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24567 | train 2.867957 | val 3.304849 | tok/s 2632091.49 | norm 0.3869 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24568 | train 2.925342 | val 3.304849 | tok/s 2621939.24 | norm 0.3643 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24569 | train 2.942716 | val 3.304849 | tok/s 2621732.52 | norm 0.3419 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24570 | train 2.955622 | val 3.304849 | tok/s 2630630.98 | norm 0.3660 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24571 | train 2.984574 | val 3.304849 | tok/s 2626731.76 | norm 0.3821 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24572 | train 2.818216 | val 3.304849 | tok/s 2625200.69 | norm 0.3743 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24573 | train 2.980316 | val 3.304849 | tok/s 2612242.83 | norm 0.4054 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24574 | train 2.856570 | val 3.304849 | tok/s 2625349.50 | norm 0.4278 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24575 | train 3.026364 | val 3.304849 | tok/s 2629029.34 | norm 0.4189 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24576 | train 2.896965 | val 3.304849 | tok/s 2624150.66 | norm 0.3326 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24577 | train 2.900868 | val 3.304849 | tok/s 2631919.29 | norm 0.3373 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24578 | train 2.944869 | val 3.304849 | tok/s 2633298.22 | norm 0.3480 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24579 | train 3.060917 | val 3.304849 | tok/s 2621418.33 | norm 0.3713 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24580 | train 2.984081 | val 3.304849 | tok/s 2615867.90 | norm 0.3812 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24581 | train 2.960658 | val 3.304849 | tok/s 2631042.41 | norm 0.3625 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24582 | train 3.036695 | val 3.304849 | tok/s 2633513.06 | norm 0.3241 | dt 199.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24583 | train 2.890760 | val 3.304849 | tok/s 2620518.52 | norm 0.3052 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24584 | train 2.975415 | val 3.304849 | tok/s 2624154.43 | norm 0.3246 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24585 | train 2.984492 | val 3.304849 | tok/s 2628270.85 | norm 0.3296 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24586 | train 3.138745 | val 3.304849 | tok/s 2624961.20 | norm 0.4447 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24587 | train 2.904002 | val 3.304849 | tok/s 2628808.27 | norm 0.4096 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24588 | train 3.023913 | val 3.304849 | tok/s 2619013.26 | norm 0.3737 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24589 | train 3.035048 | val 3.304849 | tok/s 2628618.92 | norm 0.3778 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24590 | train 2.937964 | val 3.304849 | tok/s 2634226.48 | norm 0.3698 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24591 | train 2.916857 | val 3.304849 | tok/s 2625064.93 | norm 0.3245 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24592 | train 2.985073 | val 3.304849 | tok/s 2629344.35 | norm 0.3468 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24593 | train 3.055928 | val 3.304849 | tok/s 2626306.91 | norm 0.3821 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24594 | train 2.805115 | val 3.304849 | tok/s 2621254.72 | norm 0.3773 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24595 | train 2.918440 | val 3.304849 | tok/s 2632309.82 | norm 0.3755 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24596 | train 3.040679 | val 3.304849 | tok/s 2630673.39 | norm 0.3800 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24597 | train 2.923445 | val 3.304849 | tok/s 2627057.53 | norm 0.3356 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24598 | train 2.909375 | val 3.304849 | tok/s 2625240.98 | norm 0.3684 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24599 | train 2.920667 | val 3.304849 | tok/s 2624761.61 | norm 0.3855 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24600 | train 2.937020 | val 3.304849 | tok/s 2629349.90 | norm 0.3488 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24601 | train 2.940512 | val 3.304849 | tok/s 2622460.92 | norm 0.3916 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24602 | train 2.882084 | val 3.304849 | tok/s 2627564.46 | norm 0.3864 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24603 | train 2.893398 | val 3.304849 | tok/s 2625566.75 | norm 0.3419 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24604 | train 2.919111 | val 3.304849 | tok/s 2627221.02 | norm 0.3389 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24605 | train 2.857393 | val 3.304849 | tok/s 2632079.33 | norm 0.3619 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24606 | train 2.872818 | val 3.304849 | tok/s 2626888.35 | norm 0.3593 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24607 | train 2.877375 | val 3.304849 | tok/s 2628433.92 | norm 0.3395 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24608 | train 2.868366 | val 3.304849 | tok/s 2625363.97 | norm 0.3379 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24609 | train 2.831098 | val 3.304849 | tok/s 2627480.54 | norm 0.3733 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24610 | train 2.902615 | val 3.304849 | tok/s 2627795.20 | norm 0.3892 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24611 | train 3.069359 | val 3.304849 | tok/s 2625710.30 | norm 0.3680 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24612 | train 2.997404 | val 3.304849 | tok/s 2625546.64 | norm 0.3521 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24613 | train 3.097904 | val 3.304849 | tok/s 2626651.45 | norm 0.4637 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24614 | train 2.908526 | val 3.304849 | tok/s 2625948.40 | norm 0.3385 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24615 | train 3.005691 | val 3.304849 | tok/s 2630799.34 | norm 0.3034 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24616 | train 3.009691 | val 3.304849 | tok/s 2548509.92 | norm 0.4782 | dt 205.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24617 | train 3.056625 | val 3.304849 | tok/s 2628665.58 | norm 0.3636 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24618 | train 2.976301 | val 3.304849 | tok/s 2626332.99 | norm 0.3104 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24619 | train 3.011642 | val 3.304849 | tok/s 2628842.39 | norm 0.3564 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24620 | train 2.933420 | val 3.304849 | tok/s 2624939.57 | norm 0.3792 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24621 | train 2.802386 | val 3.304849 | tok/s 2618584.60 | norm 0.3908 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24622 | train 2.976169 | val 3.304849 | tok/s 2630297.86 | norm 0.3788 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24623 | train 2.881532 | val 3.304849 | tok/s 2628127.16 | norm 0.3630 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24624 | train 3.020240 | val 3.304849 | tok/s 2627448.36 | norm 0.4081 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24625 | train 2.952460 | val 3.304849 | tok/s 2614900.68 | norm 0.3901 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24626 | train 2.858106 | val 3.304849 | tok/s 2619806.14 | norm 0.3622 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24627 | train 2.920821 | val 3.304849 | tok/s 2629809.40 | norm 0.3774 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24628 | train 2.943729 | val 3.304849 | tok/s 2630296.04 | norm 0.4790 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24629 | train 2.880711 | val 3.304849 | tok/s 2629598.69 | norm 0.4056 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24630 | train 2.967381 | val 3.304849 | tok/s 2620328.59 | norm 0.3546 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24631 | train 3.021544 | val 3.304849 | tok/s 2623303.52 | norm 0.3537 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24632 | train 3.101948 | val 3.304849 | tok/s 2630663.50 | norm 0.3947 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24633 | train 2.890307 | val 3.304849 | tok/s 2626437.81 | norm 0.3856 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24634 | train 3.005195 | val 3.304849 | tok/s 2627558.55 | norm 0.3615 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24635 | train 2.989148 | val 3.304849 | tok/s 2622346.73 | norm 0.3624 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24636 | train 2.959386 | val 3.304849 | tok/s 2622599.87 | norm 0.3580 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24637 | train 2.930297 | val 3.304849 | tok/s 2629002.40 | norm 0.3582 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24638 | train 2.949908 | val 3.304849 | tok/s 2626633.45 | norm 0.3572 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24639 | train 2.999141 | val 3.304849 | tok/s 2630190.29 | norm 0.3463 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24640 | train 2.987425 | val 3.304849 | tok/s 2628754.55 | norm 0.3359 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24641 | train 2.960120 | val 3.304849 | tok/s 2625133.13 | norm 0.3714 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24642 | train 2.968410 | val 3.304849 | tok/s 2631728.23 | norm 0.3785 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24643 | train 2.865378 | val 3.304849 | tok/s 2628989.22 | norm 0.3885 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24644 | train 2.896590 | val 3.304849 | tok/s 2629828.23 | norm 0.3929 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24645 | train 2.872802 | val 3.304849 | tok/s 2623757.88 | norm 0.3608 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24646 | train 2.934325 | val 3.304849 | tok/s 2628087.31 | norm 0.3762 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24647 | train 2.952855 | val 3.304849 | tok/s 2630792.55 | norm 0.3588 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24648 | train 2.864275 | val 3.304849 | tok/s 2627259.70 | norm 0.3438 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24649 | train 2.859142 | val 3.304849 | tok/s 2621582.20 | norm 0.3799 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24650 | train 2.811306 | val 3.304849 | tok/s 2615154.35 | norm 0.3967 | dt 200.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24651 | train 3.122034 | val 3.304849 | tok/s 2620767.54 | norm 0.4032 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24652 | train 2.915678 | val 3.304849 | tok/s 2601642.36 | norm 0.3618 | dt 201.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24653 | train 2.983924 | val 3.304849 | tok/s 2626550.01 | norm 0.3415 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24654 | train 3.023131 | val 3.304849 | tok/s 2623199.80 | norm 0.3613 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24655 | train 2.886087 | val 3.304849 | tok/s 2625628.43 | norm 0.3489 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24656 | train 2.852877 | val 3.304849 | tok/s 2627537.40 | norm 0.3790 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24657 | train 3.006978 | val 3.304849 | tok/s 2620002.84 | norm 0.3825 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24658 | train 2.986659 | val 3.304849 | tok/s 2495854.53 | norm 0.3386 | dt 210.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24659 | train 2.968489 | val 3.304849 | tok/s 2458694.75 | norm 0.3323 | dt 213.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24660 | train 3.047413 | val 3.304849 | tok/s 2623322.05 | norm 0.3401 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24661 | train 2.937154 | val 3.304849 | tok/s 2627503.62 | norm 0.3903 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24662 | train 2.870550 | val 3.304849 | tok/s 2627749.67 | norm 0.3840 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24663 | train 2.792340 | val 3.304849 | tok/s 2624935.39 | norm 0.3967 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24664 | train 2.979664 | val 3.304849 | tok/s 2629369.96 | norm 0.4090 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24665 | train 3.083226 | val 3.304849 | tok/s 2627736.57 | norm 0.3635 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24666 | train 2.870786 | val 3.304849 | tok/s 2626200.17 | norm 0.3351 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24667 | train 2.898044 | val 3.304849 | tok/s 2216619.92 | norm 0.3614 | dt 236.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24668 | train 2.942619 | val 3.304849 | tok/s 2624656.36 | norm 0.4152 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24669 | train 2.868887 | val 3.304849 | tok/s 2630787.42 | norm 0.3763 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24670 | train 3.017197 | val 3.304849 | tok/s 2629904.23 | norm 0.3450 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24671 | train 2.917895 | val 3.304849 | tok/s 2628178.40 | norm 0.3538 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24672 | train 2.846836 | val 3.304849 | tok/s 2626823.87 | norm 0.3610 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24673 | train 2.985427 | val 3.304849 | tok/s 2626046.62 | norm 0.3358 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24674 | train 2.850180 | val 3.304849 | tok/s 2628493.44 | norm 0.3831 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24675 | train 2.849607 | val 3.304849 | tok/s 2623727.12 | norm 0.3929 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24676 | train 2.855239 | val 3.304849 | tok/s 2629584.15 | norm 0.4019 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24677 | train 2.905027 | val 3.304849 | tok/s 2548869.72 | norm 0.3861 | dt 205.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24678 | train 2.933604 | val 3.304849 | tok/s 2626715.27 | norm 0.3816 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24679 | train 2.972239 | val 3.304849 | tok/s 2628987.71 | norm 0.4046 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24680 | train 3.083428 | val 3.304849 | tok/s 2624386.31 | norm 0.4138 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24681 | train 2.921327 | val 3.304849 | tok/s 2628002.57 | norm 0.3585 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24682 | train 2.946785 | val 3.304849 | tok/s 2623121.56 | norm 0.3618 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24683 | train 2.895521 | val 3.304849 | tok/s 2624357.54 | norm 0.3717 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24684 | train 2.979535 | val 3.304849 | tok/s 2470309.51 | norm 0.3566 | dt 212.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24685 | train 2.956057 | val 3.304849 | tok/s 2533522.62 | norm 0.3824 | dt 206.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24686 | train 2.989918 | val 3.304849 | tok/s 2628921.42 | norm 0.3450 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24687 | train 2.910125 | val 3.304849 | tok/s 2625976.53 | norm 0.3267 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24688 | train 3.003509 | val 3.304849 | tok/s 2629646.97 | norm 0.3590 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24689 | train 2.988721 | val 3.304849 | tok/s 2618975.15 | norm 0.3468 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24690 | train 2.873923 | val 3.304849 | tok/s 2579232.99 | norm 0.3531 | dt 203.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24691 | train 2.878660 | val 3.304849 | tok/s 2628960.92 | norm 0.3679 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24692 | train 3.060083 | val 3.304849 | tok/s 2625605.86 | norm 0.3435 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24693 | train 3.000775 | val 3.304849 | tok/s 2627080.79 | norm 0.3650 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24694 | train 2.901733 | val 3.304849 | tok/s 2618706.56 | norm 0.3664 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24695 | train 2.855576 | val 3.304849 | tok/s 2622951.00 | norm 0.3604 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24696 | train 2.917990 | val 3.304849 | tok/s 2631053.42 | norm 0.3917 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24697 | train 2.965638 | val 3.304849 | tok/s 2627117.56 | norm 0.3817 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24698 | train 3.173375 | val 3.304849 | tok/s 2626466.61 | norm 0.4788 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24699 | train 2.962157 | val 3.304849 | tok/s 2613355.82 | norm 0.4162 | dt 200.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24700 | train 3.035829 | val 3.304849 | tok/s 2624120.81 | norm 0.3616 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24701 | train 3.024547 | val 3.304849 | tok/s 2627439.56 | norm 0.3670 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24702 | train 2.930316 | val 3.304849 | tok/s 2628115.18 | norm 0.3072 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24703 | train 2.927236 | val 3.304849 | tok/s 2632905.02 | norm 0.3549 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24704 | train 2.906650 | val 3.304849 | tok/s 2616724.45 | norm 0.3905 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24705 | train 2.941936 | val 3.304849 | tok/s 2613592.99 | norm 0.3581 | dt 200.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24706 | train 2.935246 | val 3.304849 | tok/s 2632404.20 | norm 0.3595 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24707 | train 2.874763 | val 3.304849 | tok/s 2631614.44 | norm 0.3922 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24708 | train 2.996155 | val 3.304849 | tok/s 2630292.96 | norm 0.3938 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24709 | train 2.884361 | val 3.304849 | tok/s 2628790.37 | norm 0.3621 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24710 | train 2.875175 | val 3.304849 | tok/s 2627993.64 | norm 0.3476 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24711 | train 2.978702 | val 3.304849 | tok/s 2634271.76 | norm 0.3593 | dt 199.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24712 | train 2.979908 | val 3.304849 | tok/s 2625222.46 | norm 0.3410 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24713 | train 3.061116 | val 3.304849 | tok/s 2628807.44 | norm 0.3679 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24714 | train 2.918560 | val 3.304849 | tok/s 2623482.39 | norm 0.3558 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24715 | train 2.894938 | val 3.304849 | tok/s 2625958.29 | norm 0.3663 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24716 | train 2.817894 | val 3.304849 | tok/s 2628991.70 | norm 0.3678 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24717 | train 2.971062 | val 3.304849 | tok/s 2631174.70 | norm 0.4317 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24718 | train 2.972281 | val 3.304849 | tok/s 2626588.22 | norm 0.3530 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24719 | train 2.925902 | val 3.304849 | tok/s 2626256.36 | norm 0.3493 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24720 | train 2.883643 | val 3.304849 | tok/s 2622929.31 | norm 0.3577 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24721 | train 3.132827 | val 3.304849 | tok/s 2633651.80 | norm 0.3943 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24722 | train 2.932978 | val 3.304849 | tok/s 2626600.13 | norm 0.3891 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24723 | train 2.902143 | val 3.304849 | tok/s 2626190.23 | norm 0.3832 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24724 | train 2.950922 | val 3.304849 | tok/s 2625079.23 | norm 0.3986 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24725 | train 2.885645 | val 3.304849 | tok/s 2627354.62 | norm 0.3933 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24726 | train 2.858366 | val 3.304849 | tok/s 2635028.76 | norm 0.3855 | dt 198.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24727 | train 2.990564 | val 3.304849 | tok/s 2630607.06 | norm 0.3704 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24728 | train 2.997535 | val 3.304849 | tok/s 2629045.86 | norm 0.3512 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24729 | train 2.894293 | val 3.304849 | tok/s 2612645.46 | norm 0.3594 | dt 200.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24730 | train 2.922948 | val 3.304849 | tok/s 2626074.83 | norm 0.3597 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24731 | train 2.953269 | val 3.304849 | tok/s 2634297.11 | norm 0.3626 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24732 | train 2.891185 | val 3.304849 | tok/s 2629749.03 | norm 0.3647 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24733 | train 2.991377 | val 3.304849 | tok/s 2631403.59 | norm 0.3465 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24734 | train 2.885216 | val 3.304849 | tok/s 2624500.27 | norm 0.3325 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24735 | train 2.810404 | val 3.304849 | tok/s 2627042.56 | norm 0.3391 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24736 | train 2.822554 | val 3.304849 | tok/s 2630132.32 | norm 0.3691 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24737 | train 2.874555 | val 3.304849 | tok/s 2620981.49 | norm 0.3651 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24738 | train 2.975289 | val 3.304849 | tok/s 2630292.60 | norm 0.3706 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24739 | train 3.088579 | val 3.304849 | tok/s 2626015.82 | norm 0.3946 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24740 | train 2.965325 | val 3.304849 | tok/s 2631094.06 | norm 0.3662 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24741 | train 3.104111 | val 3.304849 | tok/s 2629699.16 | norm 0.3528 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24742 | train 2.975777 | val 3.304849 | tok/s 2626195.45 | norm 0.3966 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24743 | train 3.023391 | val 3.304849 | tok/s 2632396.44 | norm 0.4345 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24744 | train 2.899803 | val 3.304849 | tok/s 2626638.88 | norm 0.3339 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24745 | train 3.069590 | val 3.304849 | tok/s 2624796.52 | norm 0.3281 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24746 | train 2.891607 | val 3.304849 | tok/s 2621564.16 | norm 0.3719 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24747 | train 2.997631 | val 3.304849 | tok/s 2624164.53 | norm 0.3587 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24748 | train 2.950474 | val 3.304849 | tok/s 2629849.92 | norm 0.3755 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24749 | train 2.967044 | val 3.304849 | tok/s 2627768.10 | norm 0.3567 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 24750 | val_loss 3.299933 step 24750 | train 2.887577 | val 3.299933 | tok/s 2614875.08 | norm 0.3741 | dt 200.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24751 | train 2.990496 | val 3.299933 | tok/s 2629401.62 | norm 0.3772 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24752 | train 2.861757 | val 3.299933 | tok/s 2629293.25 | norm 0.3582 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24753 | train 3.042555 | val 3.299933 | tok/s 2622570.73 | norm 0.3594 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24754 | train 2.998749 | val 3.299933 | tok/s 2627793.91 | norm 0.3685 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24755 | train 2.860656 | val 3.299933 | tok/s 2627676.54 | norm 0.3479 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24756 | train 3.062604 | val 3.299933 | tok/s 2625550.18 | norm 0.3579 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24757 | train 2.882824 | val 3.299933 | tok/s 2622588.65 | norm 0.3448 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24758 | train 2.846999 | val 3.299933 | tok/s 2626782.43 | norm 0.3591 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24759 | train 3.404038 | val 3.299933 | tok/s 2624596.42 | norm 1.8519 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24760 | train 2.976316 | val 3.299933 | tok/s 2627094.30 | norm 0.3519 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24761 | train 2.964166 | val 3.299933 | tok/s 2624558.82 | norm 0.4379 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24762 | train 2.970800 | val 3.299933 | tok/s 2600153.94 | norm 0.3781 | dt 201.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24763 | train 2.946254 | val 3.299933 | tok/s 2624869.43 | norm 0.3529 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24764 | train 2.855326 | val 3.299933 | tok/s 2631513.41 | norm 0.3640 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24765 | train 2.902050 | val 3.299933 | tok/s 2628041.10 | norm 0.3757 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24766 | train 2.904273 | val 3.299933 | tok/s 2630501.29 | norm 0.3502 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24767 | train 3.032424 | val 3.299933 | tok/s 2626034.06 | norm 0.3566 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24768 | train 2.933050 | val 3.299933 | tok/s 2624162.19 | norm 0.3960 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24769 | train 2.844468 | val 3.299933 | tok/s 2628553.38 | norm 0.3834 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24770 | train 2.981944 | val 3.299933 | tok/s 2626461.98 | norm 0.3741 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24771 | train 2.820431 | val 3.299933 | tok/s 2626782.60 | norm 0.3831 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24772 | train 2.983882 | val 3.299933 | tok/s 2619648.70 | norm 0.3657 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24773 | train 2.802887 | val 3.299933 | tok/s 2621924.19 | norm 0.3621 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24774 | train 2.942577 | val 3.299933 | tok/s 2629917.82 | norm 0.3748 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24775 | train 2.926397 | val 3.299933 | tok/s 2629149.57 | norm 0.3963 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24776 | train 2.928249 | val 3.299933 | tok/s 2629660.69 | norm 0.3763 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24777 | train 2.916741 | val 3.299933 | tok/s 2624750.94 | norm 0.3466 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24778 | train 2.831303 | val 3.299933 | tok/s 2623646.04 | norm 0.3706 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24779 | train 2.921971 | val 3.299933 | tok/s 2629761.62 | norm 0.4093 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24780 | train 2.825351 | val 3.299933 | tok/s 2627844.11 | norm 0.3404 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24781 | train 2.935806 | val 3.299933 | tok/s 2630599.87 | norm 0.3404 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24782 | train 2.842973 | val 3.299933 | tok/s 2623873.98 | norm 0.3516 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24783 | train 2.912654 | val 3.299933 | tok/s 2624096.95 | norm 0.3726 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24784 | train 2.941097 | val 3.299933 | tok/s 2628202.65 | norm 0.3745 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24785 | train 2.873896 | val 3.299933 | tok/s 2626188.69 | norm 0.3678 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24786 | train 2.981807 | val 3.299933 | tok/s 2630794.69 | norm 0.3410 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24787 | train 2.929631 | val 3.299933 | tok/s 2621023.96 | norm 0.3253 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24788 | train 2.944890 | val 3.299933 | tok/s 2626390.98 | norm 0.3406 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24789 | train 3.004401 | val 3.299933 | tok/s 2631351.68 | norm 0.3429 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24790 | train 2.892192 | val 3.299933 | tok/s 2627141.54 | norm 0.3586 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24791 | train 3.040446 | val 3.299933 | tok/s 2629614.30 | norm 0.3662 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24792 | train 2.923698 | val 3.299933 | tok/s 2625765.42 | norm 0.3671 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24793 | train 2.934124 | val 3.299933 | tok/s 2627705.83 | norm 0.3572 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24794 | train 2.992423 | val 3.299933 | tok/s 2628970.11 | norm 0.3579 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24795 | train 2.977477 | val 3.299933 | tok/s 2628159.57 | norm 0.3743 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24796 | train 2.959990 | val 3.299933 | tok/s 2622975.83 | norm 0.3786 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24797 | train 3.025539 | val 3.299933 | tok/s 2625706.50 | norm 0.3615 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24798 | train 3.328063 | val 3.299933 | tok/s 2627813.34 | norm 0.4237 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24799 | train 2.827049 | val 3.299933 | tok/s 2626397.67 | norm 0.3340 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24800 | train 2.962881 | val 3.299933 | tok/s 2623620.36 | norm 0.3204 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24801 | train 2.941803 | val 3.299933 | tok/s 2628212.61 | norm 0.3543 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24802 | train 2.904065 | val 3.299933 | tok/s 2624725.62 | norm 0.3879 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24803 | train 2.935091 | val 3.299933 | tok/s 2628674.49 | norm 0.3449 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24804 | train 3.049333 | val 3.299933 | tok/s 2630152.42 | norm 0.3561 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24805 | train 2.920558 | val 3.299933 | tok/s 2625586.88 | norm 0.3953 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24806 | train 2.992276 | val 3.299933 | tok/s 2488621.51 | norm 0.3830 | dt 210.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24807 | train 2.860833 | val 3.299933 | tok/s 2628249.02 | norm 0.4163 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24808 | train 2.768004 | val 3.299933 | tok/s 2627482.32 | norm 0.3944 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24809 | train 2.894270 | val 3.299933 | tok/s 2631854.02 | norm 0.3172 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24810 | train 2.811762 | val 3.299933 | tok/s 2629117.77 | norm 0.3446 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24811 | train 2.907136 | val 3.299933 | tok/s 2631880.46 | norm 0.3702 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24812 | train 2.920684 | val 3.299933 | tok/s 2630633.83 | norm 0.3913 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24813 | train 2.909381 | val 3.299933 | tok/s 2615897.18 | norm 0.3634 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24814 | train 2.979486 | val 3.299933 | tok/s 2611438.35 | norm 0.3679 | dt 200.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24815 | train 2.968638 | val 3.299933 | tok/s 2623227.21 | norm 0.3701 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24816 | train 2.870414 | val 3.299933 | tok/s 2629950.34 | norm 0.3552 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24817 | train 2.921321 | val 3.299933 | tok/s 2626229.49 | norm 0.3847 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24818 | train 2.766356 | val 3.299933 | tok/s 2627088.27 | norm 0.3714 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24819 | train 3.047134 | val 3.299933 | tok/s 2620314.56 | norm 0.3304 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24820 | train 2.907053 | val 3.299933 | tok/s 2622971.78 | norm 0.3354 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24821 | train 2.908276 | val 3.299933 | tok/s 2627893.41 | norm 0.3382 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24822 | train 2.919739 | val 3.299933 | tok/s 2625252.18 | norm 0.3430 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24823 | train 2.902329 | val 3.299933 | tok/s 2629668.67 | norm 0.3497 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24824 | train 2.946406 | val 3.299933 | tok/s 2619858.10 | norm 0.3492 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24825 | train 3.106665 | val 3.299933 | tok/s 2626855.54 | norm 0.3679 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24826 | train 2.966936 | val 3.299933 | tok/s 2630175.27 | norm 0.3803 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24827 | train 3.176512 | val 3.299933 | tok/s 2626665.16 | norm 0.3771 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24828 | train 3.017450 | val 3.299933 | tok/s 2626551.64 | norm 0.4099 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24829 | train 2.876528 | val 3.299933 | tok/s 2621048.55 | norm 0.3976 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24830 | train 2.853727 | val 3.299933 | tok/s 2625815.11 | norm 0.3651 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24831 | train 2.983732 | val 3.299933 | tok/s 2633350.91 | norm 0.3696 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24832 | train 2.899155 | val 3.299933 | tok/s 2630731.94 | norm 0.3526 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24833 | train 3.027212 | val 3.299933 | tok/s 2628010.46 | norm 0.3384 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24834 | train 2.841128 | val 3.299933 | tok/s 2624006.58 | norm 0.3838 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24835 | train 2.924280 | val 3.299933 | tok/s 2622133.27 | norm 0.3587 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24836 | train 3.014799 | val 3.299933 | tok/s 2628257.94 | norm 0.3878 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24837 | train 2.965154 | val 3.299933 | tok/s 2629169.40 | norm 0.3925 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24838 | train 2.836407 | val 3.299933 | tok/s 2630129.88 | norm 0.4004 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24839 | train 2.871576 | val 3.299933 | tok/s 2623806.94 | norm 0.3745 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24840 | train 2.883340 | val 3.299933 | tok/s 2622747.61 | norm 0.3456 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24841 | train 2.782991 | val 3.299933 | tok/s 2630730.07 | norm 0.3346 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24842 | train 2.978907 | val 3.299933 | tok/s 2629503.65 | norm 0.3786 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24843 | train 2.950681 | val 3.299933 | tok/s 2627677.28 | norm 0.3590 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24844 | train 2.876256 | val 3.299933 | tok/s 2623348.05 | norm 0.3236 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24845 | train 2.843694 | val 3.299933 | tok/s 2625206.91 | norm 0.3448 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24846 | train 2.935726 | val 3.299933 | tok/s 2629948.20 | norm 0.3464 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24847 | train 3.104940 | val 3.299933 | tok/s 2626387.09 | norm 0.3499 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24848 | train 2.905876 | val 3.299933 | tok/s 2629801.84 | norm 0.3587 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24849 | train 3.056407 | val 3.299933 | tok/s 2625495.31 | norm 0.3670 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24850 | train 3.073102 | val 3.299933 | tok/s 2626659.90 | norm 0.3741 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24851 | train 2.958953 | val 3.299933 | tok/s 2630705.23 | norm 0.3906 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24852 | train 2.949225 | val 3.299933 | tok/s 2626333.59 | norm 0.4294 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24853 | train 2.943139 | val 3.299933 | tok/s 2628980.96 | norm 0.3919 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24854 | train 2.790708 | val 3.299933 | tok/s 2513081.24 | norm 0.3565 | dt 208.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24855 | train 2.944052 | val 3.299933 | tok/s 2618542.22 | norm 0.3802 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24856 | train 3.022317 | val 3.299933 | tok/s 2630421.02 | norm 0.3424 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24857 | train 3.046044 | val 3.299933 | tok/s 2621210.40 | norm 0.3457 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24858 | train 3.004309 | val 3.299933 | tok/s 2628709.22 | norm 0.3521 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24859 | train 2.990450 | val 3.299933 | tok/s 2624789.20 | norm 0.3244 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24860 | train 2.937881 | val 3.299933 | tok/s 2629682.02 | norm 0.3340 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24861 | train 2.852435 | val 3.299933 | tok/s 2538310.46 | norm 0.3446 | dt 206.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24862 | train 2.901903 | val 3.299933 | tok/s 2578383.97 | norm 0.3395 | dt 203.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24863 | train 2.892931 | val 3.299933 | tok/s 2632052.87 | norm 0.3615 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24864 | train 2.819086 | val 3.299933 | tok/s 2624259.31 | norm 0.3795 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24865 | train 2.859072 | val 3.299933 | tok/s 2627960.19 | norm 0.3600 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24866 | train 2.937744 | val 3.299933 | tok/s 2612882.31 | norm 0.3533 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24867 | train 2.941093 | val 3.299933 | tok/s 2621023.94 | norm 0.3249 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24868 | train 2.830771 | val 3.299933 | tok/s 2629612.34 | norm 0.3187 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24869 | train 2.947776 | val 3.299933 | tok/s 2628057.42 | norm 0.3594 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24870 | train 2.932938 | val 3.299933 | tok/s 2623951.75 | norm 0.3317 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24871 | train 2.962141 | val 3.299933 | tok/s 2631240.07 | norm 0.3397 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24872 | train 2.975852 | val 3.299933 | tok/s 2625141.70 | norm 0.3730 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24873 | train 2.954106 | val 3.299933 | tok/s 2628168.28 | norm 0.3411 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24874 | train 2.914124 | val 3.299933 | tok/s 2623647.29 | norm 0.3427 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24875 | train 2.844520 | val 3.299933 | tok/s 2627253.46 | norm 0.3427 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24876 | train 2.959574 | val 3.299933 | tok/s 2621330.00 | norm 0.3461 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24877 | train 3.023480 | val 3.299933 | tok/s 2608080.56 | norm 0.3626 | dt 201.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24878 | train 2.956041 | val 3.299933 | tok/s 2633028.22 | norm 0.3532 | dt 199.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24879 | train 2.992964 | val 3.299933 | tok/s 2628906.92 | norm 0.3827 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24880 | train 2.998575 | val 3.299933 | tok/s 2628498.82 | norm 0.3691 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24881 | train 2.901385 | val 3.299933 | tok/s 2627235.38 | norm 0.3445 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24882 | train 2.834763 | val 3.299933 | tok/s 2627933.37 | norm 0.3884 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24883 | train 2.877251 | val 3.299933 | tok/s 2629816.73 | norm 0.3436 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24884 | train 2.937631 | val 3.299933 | tok/s 2628228.74 | norm 0.3510 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24885 | train 2.898727 | val 3.299933 | tok/s 2634315.93 | norm 0.3836 | dt 199.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24886 | train 3.011775 | val 3.299933 | tok/s 2620053.35 | norm 0.3531 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24887 | train 3.002472 | val 3.299933 | tok/s 2625517.68 | norm 0.3524 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24888 | train 2.995253 | val 3.299933 | tok/s 2633452.48 | norm 0.3931 | dt 199.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24889 | train 3.077303 | val 3.299933 | tok/s 2630725.33 | norm 0.5141 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24890 | train 2.900593 | val 3.299933 | tok/s 2630327.19 | norm 0.3708 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24891 | train 2.884636 | val 3.299933 | tok/s 2624809.06 | norm 0.3429 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24892 | train 3.239611 | val 3.299933 | tok/s 2626590.97 | norm 0.4753 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24893 | train 2.929584 | val 3.299933 | tok/s 2631821.14 | norm 0.3558 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24894 | train 2.882427 | val 3.299933 | tok/s 2628667.98 | norm 0.3779 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24895 | train 3.114643 | val 3.299933 | tok/s 2629167.74 | norm 0.4274 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24896 | train 2.907720 | val 3.299933 | tok/s 2624186.84 | norm 0.3786 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24897 | train 3.062791 | val 3.299933 | tok/s 2626882.16 | norm 0.3572 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24898 | train 3.005558 | val 3.299933 | tok/s 2600908.91 | norm 0.3499 | dt 201.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24899 | train 2.887213 | val 3.299933 | tok/s 2623345.48 | norm 0.3627 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24900 | train 2.976588 | val 3.299933 | tok/s 2596628.05 | norm 0.3664 | dt 201.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24901 | train 2.976791 | val 3.299933 | tok/s 2626723.83 | norm 0.3685 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24902 | train 2.842521 | val 3.299933 | tok/s 2618517.42 | norm 0.3661 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24903 | train 2.886035 | val 3.299933 | tok/s 2628995.98 | norm 0.3457 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24904 | train 3.047095 | val 3.299933 | tok/s 2626024.86 | norm 0.3512 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24905 | train 2.878300 | val 3.299933 | tok/s 2628478.68 | norm 0.3424 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24906 | train 2.787711 | val 3.299933 | tok/s 2480879.67 | norm 0.3170 | dt 211.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24907 | train 2.954088 | val 3.299933 | tok/s 2626197.96 | norm 0.3578 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24908 | train 3.051959 | val 3.299933 | tok/s 2627445.57 | norm 0.4087 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24909 | train 2.991596 | val 3.299933 | tok/s 2621158.04 | norm 0.3732 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24910 | train 2.853346 | val 3.299933 | tok/s 2629837.78 | norm 0.4416 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24911 | train 2.963821 | val 3.299933 | tok/s 2627446.68 | norm 0.3632 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24912 | train 2.973038 | val 3.299933 | tok/s 2630097.48 | norm 0.3430 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24913 | train 3.020365 | val 3.299933 | tok/s 2592405.40 | norm 0.3613 | dt 202.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24914 | train 2.846939 | val 3.299933 | tok/s 2590407.24 | norm 0.3827 | dt 202.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24915 | train 2.906080 | val 3.299933 | tok/s 2627229.14 | norm 0.3224 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24916 | train 2.945176 | val 3.299933 | tok/s 2628087.38 | norm 0.3476 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24917 | train 2.849938 | val 3.299933 | tok/s 2631686.68 | norm 0.3594 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24918 | train 3.008554 | val 3.299933 | tok/s 2626951.61 | norm 0.3569 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24919 | train 3.035633 | val 3.299933 | tok/s 2622196.45 | norm 0.3618 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24920 | train 2.930050 | val 3.299933 | tok/s 2627701.73 | norm 0.3951 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24921 | train 2.849163 | val 3.299933 | tok/s 2624161.77 | norm 0.3859 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24922 | train 2.795146 | val 3.299933 | tok/s 2627446.68 | norm 0.3838 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24923 | train 2.910354 | val 3.299933 | tok/s 2622177.28 | norm 0.3514 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24924 | train 2.843468 | val 3.299933 | tok/s 2624828.15 | norm 0.3336 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24925 | train 2.907826 | val 3.299933 | tok/s 2630398.86 | norm 0.3300 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24926 | train 2.932592 | val 3.299933 | tok/s 2626964.86 | norm 0.3427 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24927 | train 2.912524 | val 3.299933 | tok/s 2627253.13 | norm 0.3360 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24928 | train 3.033516 | val 3.299933 | tok/s 2622799.97 | norm 0.3579 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24929 | train 2.913021 | val 3.299933 | tok/s 2624640.51 | norm 0.3525 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24930 | train 2.936843 | val 3.299933 | tok/s 2631223.18 | norm 0.3338 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24931 | train 3.381401 | val 3.299933 | tok/s 2628261.01 | norm 0.4798 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24932 | train 3.008433 | val 3.299933 | tok/s 2629281.36 | norm 0.3532 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24933 | train 2.950938 | val 3.299933 | tok/s 2621258.66 | norm 0.3005 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24934 | train 3.037014 | val 3.299933 | tok/s 2624537.98 | norm 0.3576 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24935 | train 2.894216 | val 3.299933 | tok/s 2631800.95 | norm 0.3433 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24936 | train 2.823147 | val 3.299933 | tok/s 2629961.28 | norm 0.3458 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24937 | train 2.913847 | val 3.299933 | tok/s 2631288.19 | norm 0.3572 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24938 | train 2.979792 | val 3.299933 | tok/s 2627999.83 | norm 0.3822 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24939 | train 3.017711 | val 3.299933 | tok/s 2624306.32 | norm 0.4263 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24940 | train 2.892964 | val 3.299933 | tok/s 2625728.96 | norm 0.3293 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24941 | train 2.995808 | val 3.299933 | tok/s 2624801.11 | norm 0.3495 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24942 | train 2.887603 | val 3.299933 | tok/s 2631246.15 | norm 0.3481 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24943 | train 2.970579 | val 3.299933 | tok/s 2624618.82 | norm 0.3173 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24944 | train 3.001879 | val 3.299933 | tok/s 2622729.96 | norm 0.3291 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24945 | train 2.934124 | val 3.299933 | tok/s 2628089.03 | norm 0.3752 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24946 | train 3.021078 | val 3.299933 | tok/s 2623556.70 | norm 0.3643 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24947 | train 2.916771 | val 3.299933 | tok/s 2628996.85 | norm 0.3288 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24948 | train 2.863423 | val 3.299933 | tok/s 2627405.76 | norm 0.3570 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24949 | train 2.903222 | val 3.299933 | tok/s 2627091.18 | norm 0.4113 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24950 | train 3.072133 | val 3.299933 | tok/s 2624568.41 | norm 0.3585 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24951 | train 2.849539 | val 3.299933 | tok/s 2618117.01 | norm 0.4611 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24952 | train 2.861364 | val 3.299933 | tok/s 2626013.70 | norm 0.3492 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24953 | train 2.838929 | val 3.299933 | tok/s 2625655.86 | norm 0.3419 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24954 | train 3.088202 | val 3.299933 | tok/s 2624498.21 | norm 0.4980 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24955 | train 2.850352 | val 3.299933 | tok/s 2629831.56 | norm 0.4069 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24956 | train 3.043569 | val 3.299933 | tok/s 2621113.17 | norm 0.3407 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24957 | train 2.898108 | val 3.299933 | tok/s 2626962.26 | norm 0.3538 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24958 | train 2.895049 | val 3.299933 | tok/s 2624706.63 | norm 0.3875 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24959 | train 2.859416 | val 3.299933 | tok/s 2621689.36 | norm 0.3811 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24960 | train 2.966640 | val 3.299933 | tok/s 2625878.84 | norm 0.3706 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24961 | train 2.842643 | val 3.299933 | tok/s 2617315.44 | norm 0.3329 | dt 200.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24962 | train 2.886641 | val 3.299933 | tok/s 2626365.09 | norm 0.3284 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24963 | train 2.919738 | val 3.299933 | tok/s 2625490.65 | norm 0.3381 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24964 | train 3.015702 | val 3.299933 | tok/s 2632180.70 | norm 0.3306 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24965 | train 3.052930 | val 3.299933 | tok/s 2619127.10 | norm 0.3500 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24966 | train 2.921112 | val 3.299933 | tok/s 2621794.86 | norm 0.3522 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24967 | train 2.968239 | val 3.299933 | tok/s 2631165.93 | norm 0.3663 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24968 | train 2.958570 | val 3.299933 | tok/s 2625179.11 | norm 0.3377 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24969 | train 2.943183 | val 3.299933 | tok/s 2630275.77 | norm 0.3344 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24970 | train 3.017677 | val 3.299933 | tok/s 2617659.17 | norm 0.3487 | dt 200.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24971 | train 2.930498 | val 3.299933 | tok/s 2594106.77 | norm 0.4478 | dt 202.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24972 | train 2.980738 | val 3.299933 | tok/s 2629020.73 | norm 0.3300 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24973 | train 2.929157 | val 3.299933 | tok/s 2624813.73 | norm 0.3304 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24974 | train 3.021574 | val 3.299933 | tok/s 2627038.49 | norm 0.3569 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24975 | train 2.890460 | val 3.299933 | tok/s 2616814.88 | norm 0.3663 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24976 | train 2.905010 | val 3.299933 | tok/s 2622778.42 | norm 0.3508 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24977 | train 2.927189 | val 3.299933 | tok/s 2631974.40 | norm 0.3785 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24978 | train 2.886291 | val 3.299933 | tok/s 2622424.64 | norm 0.3664 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24979 | train 3.016218 | val 3.299933 | tok/s 2628790.45 | norm 0.3272 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24980 | train 2.909107 | val 3.299933 | tok/s 2622745.04 | norm 0.3395 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24981 | train 2.856431 | val 3.299933 | tok/s 2624699.08 | norm 0.3602 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24982 | train 2.878348 | val 3.299933 | tok/s 2629331.43 | norm 0.3642 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24983 | train 3.062852 | val 3.299933 | tok/s 2627631.69 | norm 0.3628 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24984 | train 2.952778 | val 3.299933 | tok/s 2626837.23 | norm 0.3465 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24985 | train 3.060961 | val 3.299933 | tok/s 2625318.15 | norm 0.3633 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24986 | train 3.030424 | val 3.299933 | tok/s 2621676.04 | norm 0.3715 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24987 | train 2.796084 | val 3.299933 | tok/s 2632308.41 | norm 0.4125 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24988 | train 2.922653 | val 3.299933 | tok/s 2625185.97 | norm 0.3954 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24989 | train 3.204270 | val 3.299933 | tok/s 2629410.57 | norm 0.3614 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24990 | train 2.939670 | val 3.299933 | tok/s 2622018.30 | norm 0.3496 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24991 | train 2.934009 | val 3.299933 | tok/s 2623320.42 | norm 0.3572 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24992 | train 3.127289 | val 3.299933 | tok/s 2631287.00 | norm 0.3733 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24993 | train 2.899646 | val 3.299933 | tok/s 2629353.06 | norm 0.3674 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24994 | train 3.015632 | val 3.299933 | tok/s 2627323.64 | norm 0.3867 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24995 | train 2.848033 | val 3.299933 | tok/s 2624774.26 | norm 0.3850 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24996 | train 2.886108 | val 3.299933 | tok/s 2627815.47 | norm 0.4092 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24997 | train 3.071962 | val 3.299933 | tok/s 2631772.56 | norm 0.4079 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24998 | train 2.872169 | val 3.299933 | tok/s 2627094.94 | norm 0.3576 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 24999 | train 2.831902 | val 3.299933 | tok/s 2625612.32 | norm 0.3553 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 25000 | val_loss 3.299106 step 25000 | train 2.886413 | val 3.299106 | tok/s 2627510.19 | norm 0.3962 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25001 | train 2.897501 | val 3.299106 | tok/s 2616257.68 | norm 0.3653 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25002 | train 2.918269 | val 3.299106 | tok/s 2623250.38 | norm 0.3920 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25003 | train 3.260725 | val 3.299106 | tok/s 2629566.90 | norm 0.4501 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25004 | train 3.137760 | val 3.299106 | tok/s 2624385.90 | norm 0.3659 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25005 | train 2.909114 | val 3.299106 | tok/s 2625863.19 | norm 0.3282 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25006 | train 2.992455 | val 3.299106 | tok/s 2623267.00 | norm 0.3468 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25007 | train 3.242882 | val 3.299106 | tok/s 2623855.04 | norm 0.4043 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25008 | train 2.989309 | val 3.299106 | tok/s 2622729.88 | norm 0.3609 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25009 | train 2.998637 | val 3.299106 | tok/s 2626653.87 | norm 0.3515 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25010 | train 2.937692 | val 3.299106 | tok/s 2630301.35 | norm 0.3411 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25011 | train 2.916516 | val 3.299106 | tok/s 2627443.12 | norm 0.3437 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25012 | train 3.098176 | val 3.299106 | tok/s 2628097.72 | norm 0.3838 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25013 | train 2.862037 | val 3.299106 | tok/s 2630370.90 | norm 0.3536 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25014 | train 2.945735 | val 3.299106 | tok/s 2563136.19 | norm 0.3307 | dt 204.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25015 | train 2.890049 | val 3.299106 | tok/s 2464563.87 | norm 0.3194 | dt 212.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25016 | train 2.982092 | val 3.299106 | tok/s 2627753.24 | norm 0.3422 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25017 | train 2.920740 | val 3.299106 | tok/s 2628082.00 | norm 0.3360 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25018 | train 2.870865 | val 3.299106 | tok/s 2621464.85 | norm 0.3389 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25019 | train 2.940430 | val 3.299106 | tok/s 2621394.03 | norm 0.4481 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25020 | train 2.991532 | val 3.299106 | tok/s 2632058.78 | norm 0.3586 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25021 | train 2.914172 | val 3.299106 | tok/s 2629629.34 | norm 0.3521 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25022 | train 2.922990 | val 3.299106 | tok/s 2627711.40 | norm 0.3761 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25023 | train 2.904348 | val 3.299106 | tok/s 2613585.67 | norm 0.3699 | dt 200.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25024 | train 2.967379 | val 3.299106 | tok/s 2620643.53 | norm 0.3649 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25025 | train 2.966358 | val 3.299106 | tok/s 2626501.99 | norm 0.3645 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25026 | train 2.854084 | val 3.299106 | tok/s 2626033.76 | norm 0.3572 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25027 | train 2.893650 | val 3.299106 | tok/s 2627091.95 | norm 0.3263 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25028 | train 2.981606 | val 3.299106 | tok/s 2620811.90 | norm 0.4198 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25029 | train 2.823180 | val 3.299106 | tok/s 2597432.89 | norm 0.3983 | dt 201.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25030 | train 2.853443 | val 3.299106 | tok/s 2631090.54 | norm 0.3699 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25031 | train 2.894681 | val 3.299106 | tok/s 2626132.10 | norm 0.3753 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25032 | train 2.884655 | val 3.299106 | tok/s 2625795.42 | norm 0.3590 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25033 | train 2.956096 | val 3.299106 | tok/s 2626279.76 | norm 0.3750 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25034 | train 2.977862 | val 3.299106 | tok/s 2624404.02 | norm 0.8038 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25035 | train 2.852927 | val 3.299106 | tok/s 2629385.75 | norm 0.4233 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25036 | train 2.858819 | val 3.299106 | tok/s 2630035.97 | norm 0.3651 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25037 | train 2.956594 | val 3.299106 | tok/s 2626439.98 | norm 0.3667 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25038 | train 3.016243 | val 3.299106 | tok/s 2620827.71 | norm 0.3195 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25039 | train 3.016261 | val 3.299106 | tok/s 2623784.29 | norm 0.3290 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25040 | train 2.923505 | val 3.299106 | tok/s 2628543.86 | norm 0.3402 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25041 | train 2.959321 | val 3.299106 | tok/s 2627696.26 | norm 0.3854 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25042 | train 2.974259 | val 3.299106 | tok/s 2626608.44 | norm 0.3642 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25043 | train 2.869954 | val 3.299106 | tok/s 2622936.74 | norm 0.3242 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25044 | train 2.985734 | val 3.299106 | tok/s 2623579.33 | norm 0.3732 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25045 | train 2.847389 | val 3.299106 | tok/s 2629901.35 | norm 0.3924 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25046 | train 2.855398 | val 3.299106 | tok/s 2626294.76 | norm 0.3845 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25047 | train 2.955743 | val 3.299106 | tok/s 2627210.80 | norm 0.4039 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25048 | train 2.969056 | val 3.299106 | tok/s 2619998.29 | norm 0.3930 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25049 | train 2.885675 | val 3.299106 | tok/s 2623069.81 | norm 0.3421 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25050 | train 2.936296 | val 3.299106 | tok/s 2630024.28 | norm 0.3417 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25051 | train 3.009612 | val 3.299106 | tok/s 2622790.57 | norm 0.3661 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25052 | train 2.949874 | val 3.299106 | tok/s 2626011.48 | norm 0.3677 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25053 | train 2.984594 | val 3.299106 | tok/s 2624009.39 | norm 0.3429 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25054 | train 2.948640 | val 3.299106 | tok/s 2625684.50 | norm 0.3578 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25055 | train 2.952224 | val 3.299106 | tok/s 2631271.87 | norm 0.3547 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25056 | train 2.914797 | val 3.299106 | tok/s 2625322.54 | norm 0.3263 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25057 | train 3.059700 | val 3.299106 | tok/s 2629278.37 | norm 0.3434 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25058 | train 2.822532 | val 3.299106 | tok/s 2625133.20 | norm 0.3622 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25059 | train 2.921178 | val 3.299106 | tok/s 2625972.66 | norm 0.3600 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25060 | train 2.971228 | val 3.299106 | tok/s 2628772.66 | norm 0.3474 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25061 | train 2.916024 | val 3.299106 | tok/s 2622308.97 | norm 0.3430 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25062 | train 2.853842 | val 3.299106 | tok/s 2622529.86 | norm 0.3400 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25063 | train 2.855335 | val 3.299106 | tok/s 2624956.50 | norm 0.3423 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25064 | train 2.999630 | val 3.299106 | tok/s 2621152.54 | norm 0.3535 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25065 | train 2.944025 | val 3.299106 | tok/s 2629720.44 | norm 0.3627 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25066 | train 2.928080 | val 3.299106 | tok/s 2624503.40 | norm 0.3878 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25067 | train 3.065290 | val 3.299106 | tok/s 2625756.61 | norm 0.3923 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25068 | train 2.906286 | val 3.299106 | tok/s 2627488.98 | norm 0.3621 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25069 | train 2.998733 | val 3.299106 | tok/s 2624323.02 | norm 0.3650 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25070 | train 2.872431 | val 3.299106 | tok/s 2627759.96 | norm 0.3841 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25071 | train 2.801094 | val 3.299106 | tok/s 2619011.91 | norm 0.3695 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25072 | train 2.975864 | val 3.299106 | tok/s 2626789.60 | norm 0.3720 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25073 | train 3.069275 | val 3.299106 | tok/s 2624759.82 | norm 0.4201 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25074 | train 2.967857 | val 3.299106 | tok/s 2623436.51 | norm 0.3764 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25075 | train 2.917310 | val 3.299106 | tok/s 2630365.95 | norm 0.3778 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25076 | train 2.928763 | val 3.299106 | tok/s 2623246.96 | norm 0.3892 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25077 | train 2.971309 | val 3.299106 | tok/s 2627837.31 | norm 0.4055 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25078 | train 3.144350 | val 3.299106 | tok/s 2625698.34 | norm 0.4028 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25079 | train 2.922074 | val 3.299106 | tok/s 2628331.39 | norm 0.3870 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25080 | train 3.055084 | val 3.299106 | tok/s 2623975.71 | norm 0.3569 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25081 | train 2.919777 | val 3.299106 | tok/s 2621757.53 | norm 0.3528 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25082 | train 2.796984 | val 3.299106 | tok/s 2562326.25 | norm 0.3671 | dt 204.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25083 | train 2.959519 | val 3.299106 | tok/s 2624115.18 | norm 0.3315 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25084 | train 2.902225 | val 3.299106 | tok/s 2628393.65 | norm 0.3454 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25085 | train 2.859582 | val 3.299106 | tok/s 2620368.48 | norm 0.3821 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25086 | train 2.933598 | val 3.299106 | tok/s 2628021.08 | norm 0.4338 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25087 | train 3.102344 | val 3.299106 | tok/s 2632332.86 | norm 0.3552 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25088 | train 2.930988 | val 3.299106 | tok/s 2628203.29 | norm 0.3696 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25089 | train 3.076425 | val 3.299106 | tok/s 2627714.71 | norm 0.3825 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25090 | train 2.983439 | val 3.299106 | tok/s 2619307.62 | norm 0.3797 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25091 | train 2.831776 | val 3.299106 | tok/s 2622018.43 | norm 0.3875 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25092 | train 2.880568 | val 3.299106 | tok/s 2630492.35 | norm 0.3749 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25093 | train 2.899709 | val 3.299106 | tok/s 2625608.94 | norm 0.3548 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25094 | train 2.849188 | val 3.299106 | tok/s 2628140.14 | norm 0.3370 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25095 | train 3.061164 | val 3.299106 | tok/s 2620462.74 | norm 0.3540 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25096 | train 2.989559 | val 3.299106 | tok/s 2621519.77 | norm 0.3968 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25097 | train 2.867399 | val 3.299106 | tok/s 2626595.52 | norm 0.3665 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25098 | train 2.774676 | val 3.299106 | tok/s 2626992.82 | norm 0.3374 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25099 | train 2.996607 | val 3.299106 | tok/s 2626429.48 | norm 0.3630 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25100 | train 2.947117 | val 3.299106 | tok/s 2618451.36 | norm 0.3773 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25101 | train 3.062395 | val 3.299106 | tok/s 2623798.39 | norm 0.3968 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25102 | train 2.891774 | val 3.299106 | tok/s 2630157.09 | norm 0.3445 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25103 | train 2.937853 | val 3.299106 | tok/s 2626796.16 | norm 0.4168 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25104 | train 3.138924 | val 3.299106 | tok/s 2628664.02 | norm 0.4135 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25105 | train 2.983719 | val 3.299106 | tok/s 2623732.95 | norm 0.3758 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25106 | train 2.944092 | val 3.299106 | tok/s 2624972.46 | norm 0.3699 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25107 | train 2.961749 | val 3.299106 | tok/s 2627447.62 | norm 0.3431 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25108 | train 2.987872 | val 3.299106 | tok/s 2629173.98 | norm 0.3720 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25109 | train 2.922464 | val 3.299106 | tok/s 2629726.21 | norm 0.3723 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25110 | train 3.013278 | val 3.299106 | tok/s 2625513.54 | norm 0.3498 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25111 | train 2.994434 | val 3.299106 | tok/s 2623327.55 | norm 0.3343 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25112 | train 2.864268 | val 3.299106 | tok/s 2630962.10 | norm 0.3375 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25113 | train 2.904884 | val 3.299106 | tok/s 2628180.76 | norm 0.3468 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25114 | train 3.057628 | val 3.299106 | tok/s 2628212.50 | norm 0.4016 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25115 | train 3.042263 | val 3.299106 | tok/s 2622792.64 | norm 0.3821 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25116 | train 2.950108 | val 3.299106 | tok/s 2624488.07 | norm 0.3154 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25117 | train 2.995034 | val 3.299106 | tok/s 2630214.78 | norm 0.3480 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25118 | train 2.926813 | val 3.299106 | tok/s 2621695.88 | norm 0.3337 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25119 | train 2.986682 | val 3.299106 | tok/s 2628347.17 | norm 0.3414 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25120 | train 2.851146 | val 3.299106 | tok/s 2627119.37 | norm 0.3435 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25121 | train 2.987749 | val 3.299106 | tok/s 2626582.49 | norm 0.3582 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25122 | train 2.857548 | val 3.299106 | tok/s 2623777.10 | norm 0.3744 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25123 | train 2.964356 | val 3.299106 | tok/s 2565082.01 | norm 0.3696 | dt 204.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25124 | train 2.917114 | val 3.299106 | tok/s 2583534.05 | norm 0.3571 | dt 202.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25125 | train 2.846048 | val 3.299106 | tok/s 2622940.94 | norm 0.3289 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25126 | train 2.988742 | val 3.299106 | tok/s 2624358.29 | norm 0.3672 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25127 | train 2.929427 | val 3.299106 | tok/s 2508238.84 | norm 0.3719 | dt 209.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25128 | train 2.957713 | val 3.299106 | tok/s 2618864.45 | norm 0.3434 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25129 | train 2.953419 | val 3.299106 | tok/s 2624991.65 | norm 0.3613 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25130 | train 2.865878 | val 3.299106 | tok/s 2552455.45 | norm 0.3781 | dt 205.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25131 | train 2.909679 | val 3.299106 | tok/s 2580659.63 | norm 0.3491 | dt 203.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25132 | train 2.949411 | val 3.299106 | tok/s 2617011.83 | norm 0.3553 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25133 | train 2.927222 | val 3.299106 | tok/s 2607533.36 | norm 0.3589 | dt 201.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25134 | train 2.859869 | val 3.299106 | tok/s 2630774.83 | norm 0.3744 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25135 | train 2.962914 | val 3.299106 | tok/s 2626360.83 | norm 0.3567 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25136 | train 2.872995 | val 3.299106 | tok/s 2626834.07 | norm 0.3665 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25137 | train 3.049387 | val 3.299106 | tok/s 2619723.54 | norm 0.3508 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25138 | train 2.903131 | val 3.299106 | tok/s 2620867.38 | norm 0.3443 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25139 | train 3.016515 | val 3.299106 | tok/s 2627831.90 | norm 0.3273 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25140 | train 3.019864 | val 3.299106 | tok/s 2625469.25 | norm 0.3441 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25141 | train 2.806848 | val 3.299106 | tok/s 2630372.61 | norm 0.3595 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25142 | train 2.954190 | val 3.299106 | tok/s 2620790.28 | norm 0.3329 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25143 | train 2.908706 | val 3.299106 | tok/s 2623172.97 | norm 0.3516 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25144 | train 3.029804 | val 3.299106 | tok/s 2629287.49 | norm 0.3512 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25145 | train 2.912704 | val 3.299106 | tok/s 2628086.38 | norm 0.3534 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25146 | train 3.001118 | val 3.299106 | tok/s 2626721.62 | norm 0.3681 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25147 | train 2.847078 | val 3.299106 | tok/s 2621266.58 | norm 0.3853 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25148 | train 2.878179 | val 3.299106 | tok/s 2625583.26 | norm 0.3762 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25149 | train 2.841303 | val 3.299106 | tok/s 2631445.97 | norm 0.3584 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25150 | train 2.916155 | val 3.299106 | tok/s 2624539.45 | norm 0.3864 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25151 | train 2.988966 | val 3.299106 | tok/s 2626971.59 | norm 0.3814 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25152 | train 3.012791 | val 3.299106 | tok/s 2626029.93 | norm 0.3220 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25153 | train 2.783138 | val 3.299106 | tok/s 2627590.52 | norm 0.3576 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25154 | train 2.854958 | val 3.299106 | tok/s 2630927.35 | norm 0.3643 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25155 | train 3.012331 | val 3.299106 | tok/s 2627719.71 | norm 0.3365 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25156 | train 3.135574 | val 3.299106 | tok/s 2631874.67 | norm 0.4216 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25157 | train 2.763386 | val 3.299106 | tok/s 2625203.56 | norm 0.3937 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25158 | train 2.942639 | val 3.299106 | tok/s 2626030.35 | norm 0.3882 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25159 | train 3.040401 | val 3.299106 | tok/s 2630482.15 | norm 0.3810 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25160 | train 2.980115 | val 3.299106 | tok/s 2626438.19 | norm 0.3664 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25161 | train 2.872018 | val 3.299106 | tok/s 2629767.53 | norm 0.3667 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25162 | train 2.837906 | val 3.299106 | tok/s 2627356.00 | norm 0.3929 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25163 | train 3.043949 | val 3.299106 | tok/s 2627797.58 | norm 0.3682 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25164 | train 2.977193 | val 3.299106 | tok/s 2627580.98 | norm 0.3864 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25165 | train 2.871533 | val 3.299106 | tok/s 2627245.29 | norm 0.3966 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25166 | train 2.868054 | val 3.299106 | tok/s 2631264.51 | norm 0.4149 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25167 | train 2.915282 | val 3.299106 | tok/s 2625798.21 | norm 0.3888 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25168 | train 2.969440 | val 3.299106 | tok/s 2628331.95 | norm 0.3709 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25169 | train 2.878603 | val 3.299106 | tok/s 2631576.27 | norm 0.3725 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25170 | train 2.901763 | val 3.299106 | tok/s 2626258.09 | norm 0.3950 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25171 | train 2.954823 | val 3.299106 | tok/s 2630263.10 | norm 0.4056 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25172 | train 2.976124 | val 3.299106 | tok/s 2609402.08 | norm 0.4028 | dt 200.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25173 | train 2.953033 | val 3.299106 | tok/s 2530189.94 | norm 0.3962 | dt 207.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25174 | train 2.880705 | val 3.299106 | tok/s 2627866.87 | norm 0.3943 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25175 | train 2.763724 | val 3.299106 | tok/s 2618884.40 | norm 0.4035 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25176 | train 2.965407 | val 3.299106 | tok/s 2627773.76 | norm 0.3645 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25177 | train 2.929096 | val 3.299106 | tok/s 2620066.76 | norm 0.3595 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25178 | train 2.883617 | val 3.299106 | tok/s 2618986.98 | norm 0.3680 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25179 | train 2.828747 | val 3.299106 | tok/s 2466154.58 | norm 0.3718 | dt 212.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25180 | train 2.877875 | val 3.299106 | tok/s 2550419.48 | norm 0.3973 | dt 205.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25181 | train 3.012670 | val 3.299106 | tok/s 2629235.20 | norm 0.3753 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25182 | train 2.890550 | val 3.299106 | tok/s 2625668.84 | norm 0.3574 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25183 | train 3.162376 | val 3.299106 | tok/s 2630445.84 | norm 0.3705 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25184 | train 2.952372 | val 3.299106 | tok/s 2617493.55 | norm 0.3496 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25185 | train 2.903457 | val 3.299106 | tok/s 2622799.72 | norm 0.3384 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25186 | train 2.925522 | val 3.299106 | tok/s 2627661.76 | norm 0.3420 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25187 | train 2.934175 | val 3.299106 | tok/s 2616901.56 | norm 0.3162 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25188 | train 2.960833 | val 3.299106 | tok/s 2630736.23 | norm 0.3618 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25189 | train 2.933848 | val 3.299106 | tok/s 2620623.72 | norm 0.3844 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25190 | train 2.848698 | val 3.299106 | tok/s 2622581.45 | norm 0.3902 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25191 | train 2.897089 | val 3.299106 | tok/s 2628255.95 | norm 0.3919 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25192 | train 2.945300 | val 3.299106 | tok/s 2628397.53 | norm 0.3717 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25193 | train 2.868954 | val 3.299106 | tok/s 2629111.89 | norm 0.3674 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25194 | train 2.989710 | val 3.299106 | tok/s 2621031.04 | norm 0.3761 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25195 | train 2.899476 | val 3.299106 | tok/s 2626071.90 | norm 0.3385 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25196 | train 2.976959 | val 3.299106 | tok/s 2628344.92 | norm 0.3615 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25197 | train 2.994146 | val 3.299106 | tok/s 2625250.84 | norm 0.3754 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25198 | train 2.905068 | val 3.299106 | tok/s 2630454.73 | norm 0.3745 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25199 | train 3.017325 | val 3.299106 | tok/s 2618194.15 | norm 0.3703 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25200 | train 2.958108 | val 3.299106 | tok/s 2625114.54 | norm 0.3588 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25201 | train 2.930925 | val 3.299106 | tok/s 2629223.84 | norm 0.3464 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25202 | train 2.960896 | val 3.299106 | tok/s 2625853.35 | norm 0.3477 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25203 | train 2.825670 | val 3.299106 | tok/s 2627801.83 | norm 0.3747 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25204 | train 2.950699 | val 3.299106 | tok/s 2625821.34 | norm 0.3641 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25205 | train 2.920053 | val 3.299106 | tok/s 2627205.35 | norm 0.3569 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25206 | train 2.985404 | val 3.299106 | tok/s 2631880.63 | norm 0.3802 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25207 | train 2.942623 | val 3.299106 | tok/s 2626696.65 | norm 0.3743 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25208 | train 2.864686 | val 3.299106 | tok/s 2628884.80 | norm 0.3579 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25209 | train 2.913572 | val 3.299106 | tok/s 2624772.69 | norm 0.3459 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25210 | train 2.926295 | val 3.299106 | tok/s 2622569.03 | norm 0.3606 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25211 | train 2.899693 | val 3.299106 | tok/s 2622264.53 | norm 0.3634 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25212 | train 2.976803 | val 3.299106 | tok/s 2620984.84 | norm 0.3879 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25213 | train 2.928749 | val 3.299106 | tok/s 2623592.47 | norm 0.3833 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25214 | train 2.980569 | val 3.299106 | tok/s 2621501.00 | norm 0.3816 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25215 | train 3.054316 | val 3.299106 | tok/s 2627401.77 | norm 0.3659 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25216 | train 2.776416 | val 3.299106 | tok/s 2629305.30 | norm 0.3652 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25217 | train 2.912618 | val 3.299106 | tok/s 2618688.01 | norm 0.3575 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25218 | train 3.061566 | val 3.299106 | tok/s 2621570.38 | norm 0.4309 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25219 | train 2.910425 | val 3.299106 | tok/s 2624222.66 | norm 0.4215 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25220 | train 2.986396 | val 3.299106 | tok/s 2628453.92 | norm 0.3644 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25221 | train 2.808081 | val 3.299106 | tok/s 2628968.34 | norm 0.3951 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25222 | train 3.012224 | val 3.299106 | tok/s 2478832.92 | norm 0.3749 | dt 211.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25223 | train 2.960260 | val 3.299106 | tok/s 2629131.62 | norm 0.3648 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25224 | train 2.991228 | val 3.299106 | tok/s 2622126.84 | norm 0.3818 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25225 | train 2.960635 | val 3.299106 | tok/s 2621098.25 | norm 0.3647 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25226 | train 3.261961 | val 3.299106 | tok/s 2621335.70 | norm 0.4739 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25227 | train 2.899557 | val 3.299106 | tok/s 2622695.21 | norm 0.3707 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25228 | train 3.003397 | val 3.299106 | tok/s 2629437.76 | norm 0.3419 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25229 | train 2.965010 | val 3.299106 | tok/s 2624915.15 | norm 0.3514 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25230 | train 2.926683 | val 3.299106 | tok/s 2629302.27 | norm 0.3410 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25231 | train 2.827281 | val 3.299106 | tok/s 2623329.53 | norm 0.3183 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25232 | train 3.015527 | val 3.299106 | tok/s 2621869.92 | norm 0.3337 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25233 | train 2.877090 | val 3.299106 | tok/s 2632340.72 | norm 0.3692 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25234 | train 3.059756 | val 3.299106 | tok/s 2624827.31 | norm 0.3668 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25235 | train 3.039616 | val 3.299106 | tok/s 2632124.17 | norm 0.3596 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25236 | train 2.875347 | val 3.299106 | tok/s 2620636.42 | norm 0.3537 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25237 | train 2.857221 | val 3.299106 | tok/s 2624694.21 | norm 0.3500 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25238 | train 2.972803 | val 3.299106 | tok/s 2626639.50 | norm 0.3436 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25239 | train 2.991346 | val 3.299106 | tok/s 2627085.59 | norm 0.3375 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25240 | train 2.845927 | val 3.299106 | tok/s 2629660.86 | norm 0.5471 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25241 | train 3.007867 | val 3.299106 | tok/s 2622754.87 | norm 0.3625 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25242 | train 2.928834 | val 3.299106 | tok/s 2628235.50 | norm 0.3487 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25243 | train 2.846377 | val 3.299106 | tok/s 2631859.65 | norm 0.3588 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25244 | train 3.023226 | val 3.299106 | tok/s 2628565.16 | norm 0.3318 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25245 | train 2.850124 | val 3.299106 | tok/s 2631428.46 | norm 0.4327 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25246 | train 2.957915 | val 3.299106 | tok/s 2620436.60 | norm 0.3711 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25247 | train 2.865845 | val 3.299106 | tok/s 2625769.30 | norm 0.3648 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25248 | train 2.923778 | val 3.299106 | tok/s 2631151.69 | norm 0.3997 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25249 | train 2.961672 | val 3.299106 | tok/s 2627601.05 | norm 0.3776 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 25250 | val_loss 3.300225 step 25250 | train 2.971560 | val 3.300225 | tok/s 2607414.13 | norm 0.3783 | dt 201.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25251 | train 2.925654 | val 3.300225 | tok/s 2622333.10 | norm 0.3897 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25252 | train 2.821610 | val 3.300225 | tok/s 2624833.88 | norm 0.3507 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25253 | train 2.853736 | val 3.300225 | tok/s 2623309.96 | norm 0.3734 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25254 | train 2.935056 | val 3.300225 | tok/s 2630439.68 | norm 0.3688 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25255 | train 2.848886 | val 3.300225 | tok/s 2622972.37 | norm 0.3439 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25256 | train 2.874402 | val 3.300225 | tok/s 2630774.24 | norm 0.3604 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25257 | train 2.907503 | val 3.300225 | tok/s 2622652.40 | norm 0.3664 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25258 | train 2.808821 | val 3.300225 | tok/s 2621084.79 | norm 0.3719 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25259 | train 2.939976 | val 3.300225 | tok/s 2485850.42 | norm 0.3549 | dt 210.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25260 | train 2.875043 | val 3.300225 | tok/s 2531506.60 | norm 0.3232 | dt 207.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25261 | train 2.967681 | val 3.300225 | tok/s 2621174.75 | norm 0.3267 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25262 | train 2.975668 | val 3.300225 | tok/s 2623834.20 | norm 0.3010 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25263 | train 2.917897 | val 3.300225 | tok/s 2628691.13 | norm 0.2966 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25264 | train 2.808932 | val 3.300225 | tok/s 2508437.86 | norm 0.3182 | dt 209.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25265 | train 2.928528 | val 3.300225 | tok/s 2601044.06 | norm 0.3551 | dt 201.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25266 | train 3.008678 | val 3.300225 | tok/s 2631151.66 | norm 0.3439 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25267 | train 2.882513 | val 3.300225 | tok/s 2620640.78 | norm 0.3517 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25268 | train 2.882140 | val 3.300225 | tok/s 2628496.13 | norm 0.3598 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25269 | train 2.994327 | val 3.300225 | tok/s 2625710.30 | norm 0.3725 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25270 | train 3.079113 | val 3.300225 | tok/s 2474844.45 | norm 0.3684 | dt 211.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25271 | train 2.974183 | val 3.300225 | tok/s 2627592.89 | norm 0.3658 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25272 | train 2.907185 | val 3.300225 | tok/s 2623926.03 | norm 0.3367 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25273 | train 2.898077 | val 3.300225 | tok/s 2628221.39 | norm 0.3402 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25274 | train 2.954013 | val 3.300225 | tok/s 2620488.59 | norm 0.3508 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25275 | train 2.921451 | val 3.300225 | tok/s 2622239.05 | norm 0.3861 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25276 | train 2.870103 | val 3.300225 | tok/s 2632259.18 | norm 0.3709 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25277 | train 2.901983 | val 3.300225 | tok/s 2626329.69 | norm 0.3578 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25278 | train 2.908127 | val 3.300225 | tok/s 2603577.69 | norm 0.3966 | dt 201.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25279 | train 2.863555 | val 3.300225 | tok/s 2622131.55 | norm 0.3553 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25280 | train 2.988217 | val 3.300225 | tok/s 2624081.09 | norm 0.4234 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25281 | train 3.053916 | val 3.300225 | tok/s 2629831.58 | norm 0.3556 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25282 | train 2.955372 | val 3.300225 | tok/s 2628596.84 | norm 0.3470 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25283 | train 2.974828 | val 3.300225 | tok/s 2627966.45 | norm 0.3800 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25284 | train 3.005056 | val 3.300225 | tok/s 2624298.40 | norm 0.3654 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25285 | train 2.950317 | val 3.300225 | tok/s 2626396.69 | norm 0.3370 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25286 | train 3.062578 | val 3.300225 | tok/s 2630171.39 | norm 0.3537 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25287 | train 2.993922 | val 3.300225 | tok/s 2625523.98 | norm 0.3842 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25288 | train 3.023044 | val 3.300225 | tok/s 2630925.79 | norm 0.3805 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25289 | train 2.932782 | val 3.300225 | tok/s 2626799.42 | norm 0.3666 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25290 | train 2.874117 | val 3.300225 | tok/s 2625691.72 | norm 0.3653 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25291 | train 2.910732 | val 3.300225 | tok/s 2631993.97 | norm 0.3717 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25292 | train 3.000896 | val 3.300225 | tok/s 2628590.16 | norm 0.3575 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25293 | train 2.907852 | val 3.300225 | tok/s 2629597.25 | norm 0.3573 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25294 | train 2.949486 | val 3.300225 | tok/s 2626714.63 | norm 0.3248 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25295 | train 3.013211 | val 3.300225 | tok/s 2624960.56 | norm 0.3841 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25296 | train 3.029614 | val 3.300225 | tok/s 2630348.12 | norm 0.3429 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25297 | train 3.078522 | val 3.300225 | tok/s 2619265.47 | norm 0.3533 | dt 200.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25298 | train 2.931541 | val 3.300225 | tok/s 2621282.43 | norm 0.4031 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25299 | train 3.026409 | val 3.300225 | tok/s 2628353.59 | norm 0.4034 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25300 | train 3.006840 | val 3.300225 | tok/s 2626429.66 | norm 0.3604 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25301 | train 2.903528 | val 3.300225 | tok/s 2627584.32 | norm 0.3480 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25302 | train 2.909614 | val 3.300225 | tok/s 2618618.60 | norm 0.3824 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25303 | train 2.873115 | val 3.300225 | tok/s 2627604.10 | norm 0.4893 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25304 | train 2.931599 | val 3.300225 | tok/s 2554356.51 | norm 0.3788 | dt 205.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25305 | train 2.987512 | val 3.300225 | tok/s 2475908.31 | norm 0.3150 | dt 211.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25306 | train 3.045478 | val 3.300225 | tok/s 2621943.94 | norm 0.3719 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25307 | train 2.922139 | val 3.300225 | tok/s 2622149.23 | norm 0.3466 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25308 | train 3.002671 | val 3.300225 | tok/s 2629186.27 | norm 0.3414 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25309 | train 2.970415 | val 3.300225 | tok/s 2625900.50 | norm 0.3944 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25310 | train 2.898862 | val 3.300225 | tok/s 2628409.94 | norm 0.3942 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25311 | train 2.933394 | val 3.300225 | tok/s 2620487.77 | norm 0.3736 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25312 | train 2.950984 | val 3.300225 | tok/s 2624743.52 | norm 0.3956 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25313 | train 2.847738 | val 3.300225 | tok/s 2629558.74 | norm 0.4048 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25314 | train 2.915126 | val 3.300225 | tok/s 2625235.03 | norm 0.3844 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25315 | train 2.831783 | val 3.300225 | tok/s 2630829.50 | norm 0.3686 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25316 | train 2.923751 | val 3.300225 | tok/s 2623112.19 | norm 0.3969 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25317 | train 2.983537 | val 3.300225 | tok/s 2622976.67 | norm 0.3897 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25318 | train 2.965754 | val 3.300225 | tok/s 2631409.71 | norm 0.3729 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25319 | train 3.119235 | val 3.300225 | tok/s 2629879.10 | norm 0.4085 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25320 | train 2.972920 | val 3.300225 | tok/s 2628353.11 | norm 0.4001 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25321 | train 3.029193 | val 3.300225 | tok/s 2623288.85 | norm 0.3549 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25322 | train 2.956765 | val 3.300225 | tok/s 2623437.31 | norm 0.3662 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25323 | train 2.964824 | val 3.300225 | tok/s 2629612.15 | norm 0.3568 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25324 | train 2.892070 | val 3.300225 | tok/s 2630134.65 | norm 0.3442 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25325 | train 2.965223 | val 3.300225 | tok/s 2629253.25 | norm 0.3575 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25326 | train 3.099930 | val 3.300225 | tok/s 2624232.87 | norm 0.4221 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25327 | train 2.961627 | val 3.300225 | tok/s 2622878.74 | norm 0.3200 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25328 | train 2.854600 | val 3.300225 | tok/s 2628525.95 | norm 0.3301 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25329 | train 2.957818 | val 3.300225 | tok/s 2628673.14 | norm 0.4228 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25330 | train 3.013556 | val 3.300225 | tok/s 2630369.10 | norm 0.3383 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25331 | train 2.978228 | val 3.300225 | tok/s 2624435.86 | norm 0.3452 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25332 | train 2.877239 | val 3.300225 | tok/s 2625050.60 | norm 0.3736 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25333 | train 2.973135 | val 3.300225 | tok/s 2628061.53 | norm 0.3644 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25334 | train 3.055299 | val 3.300225 | tok/s 2625465.03 | norm 0.3638 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25335 | train 3.080361 | val 3.300225 | tok/s 2630269.93 | norm 0.4529 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25336 | train 2.895133 | val 3.300225 | tok/s 2627753.30 | norm 0.3497 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25337 | train 2.904266 | val 3.300225 | tok/s 2626843.88 | norm 0.3496 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25338 | train 2.881560 | val 3.300225 | tok/s 2631984.08 | norm 0.3881 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25339 | train 2.830007 | val 3.300225 | tok/s 2627635.14 | norm 0.3419 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25340 | train 2.854990 | val 3.300225 | tok/s 2626923.99 | norm 0.3361 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25341 | train 2.931821 | val 3.300225 | tok/s 2626289.54 | norm 0.3806 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25342 | train 2.879126 | val 3.300225 | tok/s 2623987.46 | norm 0.3833 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25343 | train 3.039459 | val 3.300225 | tok/s 2631813.91 | norm 0.4535 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25344 | train 3.004805 | val 3.300225 | tok/s 2625979.00 | norm 0.3616 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25345 | train 2.859801 | val 3.300225 | tok/s 2628033.30 | norm 0.3719 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25346 | train 2.826858 | val 3.300225 | tok/s 2621744.87 | norm 0.4044 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25347 | train 2.946522 | val 3.300225 | tok/s 2625867.51 | norm 0.3677 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25348 | train 2.919374 | val 3.300225 | tok/s 2629658.83 | norm 0.3787 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25349 | train 2.892545 | val 3.300225 | tok/s 2624576.18 | norm 0.3567 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25350 | train 2.881846 | val 3.300225 | tok/s 2626771.66 | norm 0.3531 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25351 | train 3.003132 | val 3.300225 | tok/s 2626853.37 | norm 0.3694 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25352 | train 2.934588 | val 3.300225 | tok/s 2628834.48 | norm 0.3810 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25353 | train 2.898149 | val 3.300225 | tok/s 2623440.86 | norm 0.3733 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25354 | train 3.001881 | val 3.300225 | tok/s 2621439.74 | norm 0.3773 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25355 | train 3.024086 | val 3.300225 | tok/s 2529906.64 | norm 0.3691 | dt 207.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25356 | train 2.868557 | val 3.300225 | tok/s 2613111.09 | norm 0.3858 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25357 | train 2.979167 | val 3.300225 | tok/s 2628562.77 | norm 0.3811 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25358 | train 2.970159 | val 3.300225 | tok/s 2623210.09 | norm 0.3729 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25359 | train 3.003976 | val 3.300225 | tok/s 2625566.52 | norm 0.3739 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25360 | train 2.851727 | val 3.300225 | tok/s 2628842.72 | norm 0.3432 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25361 | train 2.908168 | val 3.300225 | tok/s 2628839.54 | norm 0.3631 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25362 | train 3.010537 | val 3.300225 | tok/s 2627541.08 | norm 0.3795 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25363 | train 2.900197 | val 3.300225 | tok/s 2614284.74 | norm 0.3505 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25364 | train 2.931297 | val 3.300225 | tok/s 2622498.60 | norm 0.3439 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25365 | train 2.927995 | val 3.300225 | tok/s 2631036.67 | norm 0.3240 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25366 | train 2.982737 | val 3.300225 | tok/s 2627534.12 | norm 0.3425 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25367 | train 2.869470 | val 3.300225 | tok/s 2627721.38 | norm 0.3389 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25368 | train 3.003740 | val 3.300225 | tok/s 2621282.15 | norm 0.3835 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25369 | train 2.991480 | val 3.300225 | tok/s 2623774.45 | norm 0.3955 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25370 | train 2.926989 | val 3.300225 | tok/s 2629549.62 | norm 0.3937 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25371 | train 2.916950 | val 3.300225 | tok/s 2623785.85 | norm 0.4085 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25372 | train 2.993518 | val 3.300225 | tok/s 2630517.87 | norm 0.3954 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25373 | train 2.886941 | val 3.300225 | tok/s 2609793.44 | norm 0.3718 | dt 200.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25374 | train 3.055052 | val 3.300225 | tok/s 2602292.37 | norm 0.3937 | dt 201.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25375 | train 2.824241 | val 3.300225 | tok/s 2631584.53 | norm 0.3338 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25376 | train 2.938802 | val 3.300225 | tok/s 2627495.93 | norm 0.3361 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25377 | train 2.987723 | val 3.300225 | tok/s 2629065.57 | norm 0.3597 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25378 | train 3.003379 | val 3.300225 | tok/s 2627005.68 | norm 0.3690 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25379 | train 2.911286 | val 3.300225 | tok/s 2625543.61 | norm 0.3524 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25380 | train 2.800109 | val 3.300225 | tok/s 2629401.79 | norm 0.3157 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25381 | train 2.819144 | val 3.300225 | tok/s 2627900.88 | norm 0.3268 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25382 | train 2.897315 | val 3.300225 | tok/s 2631125.09 | norm 0.3793 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25383 | train 3.006094 | val 3.300225 | tok/s 2622623.67 | norm 0.3534 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25384 | train 2.929264 | val 3.300225 | tok/s 2606843.07 | norm 0.3514 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25385 | train 2.981542 | val 3.300225 | tok/s 2497527.10 | norm 0.3959 | dt 209.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25386 | train 3.020418 | val 3.300225 | tok/s 2617774.93 | norm 0.3549 | dt 200.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25387 | train 2.846202 | val 3.300225 | tok/s 2617446.13 | norm 0.3356 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25388 | train 2.849581 | val 3.300225 | tok/s 2623081.14 | norm 0.3918 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25389 | train 2.886993 | val 3.300225 | tok/s 2629514.94 | norm 0.4301 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25390 | train 2.886979 | val 3.300225 | tok/s 2625129.88 | norm 0.3674 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25391 | train 2.957249 | val 3.300225 | tok/s 2625905.93 | norm 0.3458 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25392 | train 2.956501 | val 3.300225 | tok/s 2625680.38 | norm 0.3572 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25393 | train 2.838236 | val 3.300225 | tok/s 2625158.88 | norm 0.3509 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25394 | train 2.888559 | val 3.300225 | tok/s 2630620.44 | norm 0.3607 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25395 | train 2.809456 | val 3.300225 | tok/s 2628516.06 | norm 0.3561 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25396 | train 2.873540 | val 3.300225 | tok/s 2626244.69 | norm 0.3842 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25397 | train 2.969139 | val 3.300225 | tok/s 2622912.75 | norm 0.3942 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25398 | train 2.990050 | val 3.300225 | tok/s 2627560.09 | norm 0.3654 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25399 | train 2.860898 | val 3.300225 | tok/s 2628816.62 | norm 0.3350 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25400 | train 3.083456 | val 3.300225 | tok/s 2625628.98 | norm 0.3437 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25401 | train 2.909359 | val 3.300225 | tok/s 2624550.30 | norm 0.3557 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25402 | train 2.890999 | val 3.300225 | tok/s 2620958.73 | norm 0.3682 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25403 | train 2.978003 | val 3.300225 | tok/s 2622178.16 | norm 0.3821 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25404 | train 2.999979 | val 3.300225 | tok/s 2630185.14 | norm 0.3705 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25405 | train 2.886039 | val 3.300225 | tok/s 2623281.89 | norm 0.3797 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25406 | train 2.891689 | val 3.300225 | tok/s 2627688.80 | norm 0.4473 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25407 | train 2.851862 | val 3.300225 | tok/s 2623390.57 | norm 0.4431 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25408 | train 2.922529 | val 3.300225 | tok/s 2624728.77 | norm 0.3405 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25409 | train 2.971368 | val 3.300225 | tok/s 2627065.87 | norm 0.3098 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25410 | train 2.940899 | val 3.300225 | tok/s 2626455.70 | norm 0.3551 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25411 | train 2.910642 | val 3.300225 | tok/s 2628028.02 | norm 0.3710 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25412 | train 3.071776 | val 3.300225 | tok/s 2620464.94 | norm 0.4249 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25413 | train 2.895142 | val 3.300225 | tok/s 2622428.77 | norm 0.3818 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25414 | train 3.011058 | val 3.300225 | tok/s 2631124.12 | norm 0.4016 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25415 | train 2.901673 | val 3.300225 | tok/s 2628638.03 | norm 0.4333 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25416 | train 2.922122 | val 3.300225 | tok/s 2628606.05 | norm 0.4179 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25417 | train 2.831295 | val 3.300225 | tok/s 2626579.53 | norm 0.4040 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25418 | train 2.814997 | val 3.300225 | tok/s 2626366.37 | norm 0.3951 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25419 | train 2.877543 | val 3.300225 | tok/s 2630439.38 | norm 0.3455 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25420 | train 2.996592 | val 3.300225 | tok/s 2627881.08 | norm 0.3825 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25421 | train 2.891916 | val 3.300225 | tok/s 2629151.12 | norm 0.4060 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25422 | train 2.768279 | val 3.300225 | tok/s 2623565.27 | norm 0.3560 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25423 | train 2.973098 | val 3.300225 | tok/s 2626789.48 | norm 0.3675 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25424 | train 2.935052 | val 3.300225 | tok/s 2626509.06 | norm 0.3892 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25425 | train 2.898336 | val 3.300225 | tok/s 2517145.62 | norm 0.3750 | dt 208.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25426 | train 2.989486 | val 3.300225 | tok/s 2542194.11 | norm 0.3742 | dt 206.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25427 | train 2.874755 | val 3.300225 | tok/s 2627349.47 | norm 0.3747 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25428 | train 2.971055 | val 3.300225 | tok/s 2629633.03 | norm 0.3473 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25429 | train 3.105253 | val 3.300225 | tok/s 2622989.31 | norm 0.6580 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25430 | train 2.730713 | val 3.300225 | tok/s 2621111.31 | norm 0.5775 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25431 | train 2.962891 | val 3.300225 | tok/s 2629689.92 | norm 0.3139 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25432 | train 2.833015 | val 3.300225 | tok/s 2628371.88 | norm 0.3261 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25433 | train 2.892603 | val 3.300225 | tok/s 2630634.22 | norm 0.2987 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25434 | train 2.892900 | val 3.300225 | tok/s 2617911.92 | norm 0.3356 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25435 | train 3.039238 | val 3.300225 | tok/s 2624820.53 | norm 0.3550 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25436 | train 2.942006 | val 3.300225 | tok/s 2630657.34 | norm 0.3574 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25437 | train 3.059755 | val 3.300225 | tok/s 2631083.29 | norm 0.3458 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25438 | train 2.944466 | val 3.300225 | tok/s 2628597.63 | norm 0.3334 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25439 | train 2.907609 | val 3.300225 | tok/s 2626618.04 | norm 0.3792 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25440 | train 2.915663 | val 3.300225 | tok/s 2624069.78 | norm 0.4131 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25441 | train 2.936078 | val 3.300225 | tok/s 2625983.65 | norm 0.3399 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25442 | train 2.906385 | val 3.300225 | tok/s 2579713.55 | norm 0.3333 | dt 203.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25443 | train 2.861932 | val 3.300225 | tok/s 2581441.35 | norm 0.3715 | dt 203.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25444 | train 2.827479 | val 3.300225 | tok/s 2624439.17 | norm 0.3564 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25445 | train 3.075338 | val 3.300225 | tok/s 2628348.44 | norm 0.3637 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25446 | train 2.914776 | val 3.300225 | tok/s 2628685.35 | norm 0.3966 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25447 | train 2.874864 | val 3.300225 | tok/s 2626543.65 | norm 0.3742 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25448 | train 3.029525 | val 3.300225 | tok/s 2630402.68 | norm 0.3635 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25449 | train 2.955804 | val 3.300225 | tok/s 2623749.53 | norm 0.3300 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25450 | train 2.772467 | val 3.300225 | tok/s 2625988.84 | norm 0.3247 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25451 | train 2.919001 | val 3.300225 | tok/s 2627753.10 | norm 0.3198 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25452 | train 2.860165 | val 3.300225 | tok/s 2627617.45 | norm 0.3251 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25453 | train 3.036290 | val 3.300225 | tok/s 2626823.15 | norm 0.3537 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25454 | train 2.890090 | val 3.300225 | tok/s 2625720.99 | norm 0.3669 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25455 | train 3.031510 | val 3.300225 | tok/s 2628471.28 | norm 0.4200 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25456 | train 2.951947 | val 3.300225 | tok/s 2631809.17 | norm 0.3795 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25457 | train 2.807703 | val 3.300225 | tok/s 2624074.82 | norm 0.3636 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25458 | train 2.968925 | val 3.300225 | tok/s 2623560.01 | norm 0.3942 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25459 | train 2.892235 | val 3.300225 | tok/s 2626652.29 | norm 0.4412 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25460 | train 3.047163 | val 3.300225 | tok/s 2626960.81 | norm 0.4059 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25461 | train 3.040859 | val 3.300225 | tok/s 2631901.86 | norm 0.4078 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25462 | train 2.972374 | val 3.300225 | tok/s 2628099.04 | norm 0.3500 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25463 | train 2.951278 | val 3.300225 | tok/s 2629059.34 | norm 0.3280 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25464 | train 2.941442 | val 3.300225 | tok/s 2622042.30 | norm 0.3513 | dt 199.95 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25465 | train 3.044196 | val 3.300225 | tok/s 2625087.80 | norm 0.3576 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25466 | train 2.941535 | val 3.300225 | tok/s 2631957.99 | norm 0.3617 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25467 | train 2.913590 | val 3.300225 | tok/s 2626547.08 | norm 0.3645 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25468 | train 2.910450 | val 3.300225 | tok/s 2621670.81 | norm 0.3813 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25469 | train 2.918265 | val 3.300225 | tok/s 2622923.87 | norm 0.3394 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25470 | train 2.962443 | val 3.300225 | tok/s 2622000.61 | norm 0.3493 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25471 | train 2.781081 | val 3.300225 | tok/s 2627377.43 | norm 0.3512 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25472 | train 2.830583 | val 3.300225 | tok/s 2624051.81 | norm 0.3807 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25473 | train 2.886963 | val 3.300225 | tok/s 2630188.22 | norm 0.3839 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25474 | train 2.897467 | val 3.300225 | tok/s 2629145.38 | norm 0.4008 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25475 | train 3.042511 | val 3.300225 | tok/s 2629594.72 | norm 0.3557 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25476 | train 3.024046 | val 3.300225 | tok/s 2629497.70 | norm 0.3254 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25477 | train 2.807891 | val 3.300225 | tok/s 2625832.98 | norm 0.3373 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25478 | train 2.954864 | val 3.300225 | tok/s 2629461.74 | norm 0.3772 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25479 | train 2.965275 | val 3.300225 | tok/s 2626581.74 | norm 0.3412 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25480 | train 2.896367 | val 3.300225 | tok/s 2584221.78 | norm 0.3602 | dt 202.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25481 | train 2.954593 | val 3.300225 | tok/s 2433574.38 | norm 0.3636 | dt 215.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25482 | train 2.955168 | val 3.300225 | tok/s 2623424.56 | norm 0.3461 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25483 | train 2.853395 | val 3.300225 | tok/s 2628314.78 | norm 0.3544 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25484 | train 2.948499 | val 3.300225 | tok/s 2628220.51 | norm 0.3525 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25485 | train 3.029446 | val 3.300225 | tok/s 2630340.02 | norm 0.3646 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25486 | train 2.945288 | val 3.300225 | tok/s 2626756.28 | norm 0.3754 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25487 | train 2.990077 | val 3.300225 | tok/s 2572623.21 | norm 0.3563 | dt 203.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25488 | train 2.908422 | val 3.300225 | tok/s 2628761.93 | norm 0.3169 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25489 | train 2.908355 | val 3.300225 | tok/s 2627285.80 | norm 0.3410 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25490 | train 2.945436 | val 3.300225 | tok/s 2629263.83 | norm 0.4124 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25491 | train 3.049367 | val 3.300225 | tok/s 2623077.78 | norm 0.3950 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25492 | train 3.114019 | val 3.300225 | tok/s 2625057.55 | norm 0.3826 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25493 | train 2.957613 | val 3.300225 | tok/s 2629154.10 | norm 0.3639 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25494 | train 2.964254 | val 3.300225 | tok/s 2623793.88 | norm 0.3661 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25495 | train 2.854075 | val 3.300225 | tok/s 2628549.75 | norm 0.3996 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25496 | train 2.824814 | val 3.300225 | tok/s 2624496.33 | norm 0.3432 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25497 | train 3.023146 | val 3.300225 | tok/s 2625135.17 | norm 0.3268 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25498 | train 2.947393 | val 3.300225 | tok/s 2630472.71 | norm 0.3809 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25499 | train 2.952420 | val 3.300225 | tok/s 2625331.27 | norm 0.3532 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 25500 | val_loss 3.296453 step 25500 | train 2.915622 | val 3.296453 | tok/s 2611075.86 | norm 0.3245 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25501 | train 3.026758 | val 3.296453 | tok/s 2603833.83 | norm 0.3368 | dt 201.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25502 | train 3.018705 | val 3.296453 | tok/s 2606782.23 | norm 0.3868 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25503 | train 3.024388 | val 3.296453 | tok/s 2623475.54 | norm 0.3686 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25504 | train 2.850346 | val 3.296453 | tok/s 2598764.89 | norm 0.3804 | dt 201.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25505 | train 2.886634 | val 3.296453 | tok/s 2626839.76 | norm 0.4080 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25506 | train 2.881491 | val 3.296453 | tok/s 2627273.02 | norm 0.3904 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25507 | train 2.811710 | val 3.296453 | tok/s 2625769.38 | norm 0.3827 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25508 | train 2.976840 | val 3.296453 | tok/s 2623385.75 | norm 0.3582 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25509 | train 2.966912 | val 3.296453 | tok/s 2627890.09 | norm 0.3249 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25510 | train 2.968336 | val 3.296453 | tok/s 2623629.00 | norm 0.3618 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25511 | train 3.296596 | val 3.296453 | tok/s 2628314.17 | norm 0.5341 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25512 | train 2.892609 | val 3.296453 | tok/s 2625868.80 | norm 0.3618 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25513 | train 2.875432 | val 3.296453 | tok/s 2626764.89 | norm 0.3532 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25514 | train 2.845616 | val 3.296453 | tok/s 2627997.37 | norm 0.3699 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25515 | train 2.908821 | val 3.296453 | tok/s 2626729.66 | norm 0.3646 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25516 | train 2.893328 | val 3.296453 | tok/s 2629809.68 | norm 0.3645 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25517 | train 3.034720 | val 3.296453 | tok/s 2629368.24 | norm 0.3535 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25518 | train 3.002794 | val 3.296453 | tok/s 2629294.36 | norm 0.4904 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25519 | train 2.930512 | val 3.296453 | tok/s 2618284.85 | norm 0.3182 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25520 | train 2.830852 | val 3.296453 | tok/s 2618800.30 | norm 0.2990 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25521 | train 2.852815 | val 3.296453 | tok/s 2630569.63 | norm 0.3677 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25522 | train 2.923072 | val 3.296453 | tok/s 2627547.41 | norm 0.3737 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25523 | train 2.924670 | val 3.296453 | tok/s 2627963.37 | norm 0.3469 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25524 | train 2.994081 | val 3.296453 | tok/s 2619869.48 | norm 0.3334 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25525 | train 2.913867 | val 3.296453 | tok/s 2616755.58 | norm 0.3497 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25526 | train 2.946771 | val 3.296453 | tok/s 2629537.74 | norm 0.3803 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25527 | train 2.848705 | val 3.296453 | tok/s 2613816.33 | norm 0.3393 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25528 | train 2.963603 | val 3.296453 | tok/s 2631388.66 | norm 0.3378 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25529 | train 2.970352 | val 3.296453 | tok/s 2620687.98 | norm 0.3455 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25530 | train 2.814332 | val 3.296453 | tok/s 2620416.47 | norm 0.3323 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25531 | train 2.876644 | val 3.296453 | tok/s 2627373.11 | norm 0.3975 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25532 | train 2.849720 | val 3.296453 | tok/s 2627809.76 | norm 0.3688 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25533 | train 2.894729 | val 3.296453 | tok/s 2543060.83 | norm 0.3623 | dt 206.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25534 | train 2.894127 | val 3.296453 | tok/s 2620239.41 | norm 0.3872 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25535 | train 3.023540 | val 3.296453 | tok/s 2625395.18 | norm 0.3993 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25536 | train 2.985241 | val 3.296453 | tok/s 2628781.11 | norm 0.3451 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25537 | train 2.749257 | val 3.296453 | tok/s 2626607.22 | norm 0.3343 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25538 | train 2.905370 | val 3.296453 | tok/s 2624713.61 | norm 0.3480 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25539 | train 2.922706 | val 3.296453 | tok/s 2624686.26 | norm 0.3417 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25540 | train 2.992133 | val 3.296453 | tok/s 2617894.44 | norm 0.3501 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25541 | train 2.932746 | val 3.296453 | tok/s 2628987.67 | norm 0.3718 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25542 | train 2.945939 | val 3.296453 | tok/s 2619900.26 | norm 0.3792 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25543 | train 2.902915 | val 3.296453 | tok/s 2632690.30 | norm 0.3631 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25544 | train 2.891155 | val 3.296453 | tok/s 2624716.17 | norm 0.3519 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25545 | train 2.908313 | val 3.296453 | tok/s 2625181.79 | norm 0.3253 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25546 | train 3.000118 | val 3.296453 | tok/s 2627696.61 | norm 0.3585 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25547 | train 2.854020 | val 3.296453 | tok/s 2624399.50 | norm 0.3333 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25548 | train 2.827431 | val 3.296453 | tok/s 2631344.83 | norm 0.3393 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25549 | train 2.947699 | val 3.296453 | tok/s 2624761.75 | norm 0.3419 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25550 | train 2.926347 | val 3.296453 | tok/s 2624986.44 | norm 0.3475 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25551 | train 2.999498 | val 3.296453 | tok/s 2625536.95 | norm 0.3344 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25552 | train 2.985795 | val 3.296453 | tok/s 2616316.39 | norm 0.4103 | dt 200.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25553 | train 2.991391 | val 3.296453 | tok/s 2623830.55 | norm 0.3730 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25554 | train 2.983108 | val 3.296453 | tok/s 2620295.60 | norm 0.3668 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25555 | train 3.018127 | val 3.296453 | tok/s 2618440.68 | norm 0.3771 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25556 | train 2.884142 | val 3.296453 | tok/s 2628535.42 | norm 0.3709 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25557 | train 3.048071 | val 3.296453 | tok/s 2629745.64 | norm 0.3756 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25558 | train 3.018140 | val 3.296453 | tok/s 2629393.02 | norm 0.3743 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25559 | train 2.875836 | val 3.296453 | tok/s 2625001.41 | norm 0.3246 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25560 | train 2.828908 | val 3.296453 | tok/s 2623232.55 | norm 0.3276 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25561 | train 2.799302 | val 3.296453 | tok/s 2628449.52 | norm 0.3552 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25562 | train 2.848100 | val 3.296453 | tok/s 2628778.23 | norm 0.3512 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25563 | train 2.830961 | val 3.296453 | tok/s 2627179.40 | norm 0.3222 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25564 | train 2.978095 | val 3.296453 | tok/s 2619711.68 | norm 0.3353 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25565 | train 2.867865 | val 3.296453 | tok/s 2627052.50 | norm 0.3564 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25566 | train 2.905128 | val 3.296453 | tok/s 2617019.04 | norm 0.3835 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25567 | train 2.926229 | val 3.296453 | tok/s 2626190.99 | norm 0.3780 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25568 | train 2.886448 | val 3.296453 | tok/s 2629365.63 | norm 0.3614 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25569 | train 3.028116 | val 3.296453 | tok/s 2626977.25 | norm 0.3857 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25570 | train 2.910509 | val 3.296453 | tok/s 2627338.74 | norm 0.3682 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25571 | train 3.013268 | val 3.296453 | tok/s 2628441.47 | norm 0.3439 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25572 | train 2.873941 | val 3.296453 | tok/s 2616442.03 | norm 0.3385 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25573 | train 2.904258 | val 3.296453 | tok/s 2629402.41 | norm 0.3569 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25574 | train 3.078711 | val 3.296453 | tok/s 2630394.12 | norm 0.4122 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25575 | train 2.926192 | val 3.296453 | tok/s 2629893.93 | norm 0.3309 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25576 | train 2.894325 | val 3.296453 | tok/s 2626095.59 | norm 0.3542 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25577 | train 2.944431 | val 3.296453 | tok/s 2543009.12 | norm 0.3866 | dt 206.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25578 | train 3.041006 | val 3.296453 | tok/s 2633155.48 | norm 0.3688 | dt 199.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25579 | train 2.905274 | val 3.296453 | tok/s 2631430.74 | norm 0.4229 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25580 | train 2.980337 | val 3.296453 | tok/s 2630727.72 | norm 0.3662 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25581 | train 2.906242 | val 3.296453 | tok/s 2627341.48 | norm 0.3814 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25582 | train 2.774969 | val 3.296453 | tok/s 2624754.08 | norm 0.3725 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25583 | train 3.116338 | val 3.296453 | tok/s 2628080.42 | norm 0.3890 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25584 | train 3.046663 | val 3.296453 | tok/s 2480831.31 | norm 0.3532 | dt 211.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25585 | train 3.000654 | val 3.296453 | tok/s 2625340.88 | norm 0.3436 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25586 | train 2.871767 | val 3.296453 | tok/s 2621314.02 | norm 0.3561 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25587 | train 2.887841 | val 3.296453 | tok/s 2623423.96 | norm 0.3764 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25588 | train 3.002453 | val 3.296453 | tok/s 2630123.14 | norm 0.3783 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25589 | train 2.883213 | val 3.296453 | tok/s 2625238.30 | norm 0.3800 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25590 | train 2.961739 | val 3.296453 | tok/s 2628764.39 | norm 0.3894 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25591 | train 3.020024 | val 3.296453 | tok/s 2578569.26 | norm 0.3828 | dt 203.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25592 | train 2.840390 | val 3.296453 | tok/s 2626012.49 | norm 0.3775 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25593 | train 2.927253 | val 3.296453 | tok/s 2631192.33 | norm 0.3596 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25594 | train 2.951701 | val 3.296453 | tok/s 2630394.34 | norm 0.3652 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25595 | train 2.942433 | val 3.296453 | tok/s 2628951.60 | norm 0.3793 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25596 | train 2.979016 | val 3.296453 | tok/s 2623089.56 | norm 0.3851 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25597 | train 3.006530 | val 3.296453 | tok/s 2620358.58 | norm 0.4130 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25598 | train 2.903173 | val 3.296453 | tok/s 2620401.68 | norm 0.3938 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25599 | train 3.042909 | val 3.296453 | tok/s 2623533.79 | norm 0.3710 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25600 | train 2.895256 | val 3.296453 | tok/s 2630104.06 | norm 0.3739 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25601 | train 2.896248 | val 3.296453 | tok/s 2622237.76 | norm 0.3616 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25602 | train 2.887023 | val 3.296453 | tok/s 2626491.23 | norm 0.3740 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25603 | train 2.934580 | val 3.296453 | tok/s 2628354.27 | norm 0.4209 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25604 | train 2.892667 | val 3.296453 | tok/s 2630062.90 | norm 0.3761 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25605 | train 3.020870 | val 3.296453 | tok/s 2620315.22 | norm 0.3636 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25606 | train 2.966098 | val 3.296453 | tok/s 2618652.38 | norm 0.3533 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25607 | train 2.844940 | val 3.296453 | tok/s 2626222.13 | norm 0.3263 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25608 | train 2.829748 | val 3.296453 | tok/s 2632209.74 | norm 0.3216 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25609 | train 2.828655 | val 3.296453 | tok/s 2627848.96 | norm 0.5393 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25610 | train 2.915624 | val 3.296453 | tok/s 2632728.35 | norm 0.3413 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25611 | train 2.941519 | val 3.296453 | tok/s 2627851.91 | norm 0.3542 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25612 | train 3.026386 | val 3.296453 | tok/s 2625512.31 | norm 0.3834 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25613 | train 2.863537 | val 3.296453 | tok/s 2629184.25 | norm 0.3603 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25614 | train 2.879040 | val 3.296453 | tok/s 2624104.44 | norm 0.3577 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25615 | train 3.273359 | val 3.296453 | tok/s 2631673.22 | norm 0.6719 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25616 | train 3.053821 | val 3.296453 | tok/s 2625515.94 | norm 0.3331 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25617 | train 2.909466 | val 3.296453 | tok/s 2629653.28 | norm 0.3457 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25618 | train 2.986944 | val 3.296453 | tok/s 2628504.53 | norm 0.3216 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25619 | train 2.917768 | val 3.296453 | tok/s 2622855.50 | norm 0.3457 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25620 | train 2.866775 | val 3.296453 | tok/s 2632762.78 | norm 0.4202 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25621 | train 3.111869 | val 3.296453 | tok/s 2623681.00 | norm 0.3807 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25622 | train 2.987824 | val 3.296453 | tok/s 2631177.38 | norm 0.3568 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25623 | train 2.858788 | val 3.296453 | tok/s 2627580.97 | norm 0.3501 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25624 | train 3.030809 | val 3.296453 | tok/s 2620883.59 | norm 0.3465 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25625 | train 2.865396 | val 3.296453 | tok/s 2630502.24 | norm 0.3120 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25626 | train 2.998588 | val 3.296453 | tok/s 2616260.68 | norm 0.3590 | dt 200.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25627 | train 2.992469 | val 3.296453 | tok/s 2626729.23 | norm 0.3962 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25628 | train 3.013534 | val 3.296453 | tok/s 2628313.90 | norm 0.3502 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25629 | train 2.850448 | val 3.296453 | tok/s 2626390.94 | norm 0.3552 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25630 | train 2.847293 | val 3.296453 | tok/s 2628087.97 | norm 0.3774 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25631 | train 2.880788 | val 3.296453 | tok/s 2625329.89 | norm 0.3744 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25632 | train 3.023036 | val 3.296453 | tok/s 2624246.21 | norm 0.3556 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25633 | train 2.924843 | val 3.296453 | tok/s 2616608.24 | norm 0.3458 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25634 | train 2.876143 | val 3.296453 | tok/s 2620846.65 | norm 0.3390 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25635 | train 2.935905 | val 3.296453 | tok/s 2628764.62 | norm 0.3445 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25636 | train 2.899487 | val 3.296453 | tok/s 2625533.60 | norm 0.3721 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25637 | train 2.911726 | val 3.296453 | tok/s 2626264.57 | norm 0.4018 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25638 | train 3.099633 | val 3.296453 | tok/s 2514409.76 | norm 0.3724 | dt 208.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25639 | train 2.830799 | val 3.296453 | tok/s 2621490.94 | norm 0.3920 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25640 | train 2.888580 | val 3.296453 | tok/s 2625570.93 | norm 0.4056 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25641 | train 2.992807 | val 3.296453 | tok/s 2629040.25 | norm 0.3693 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25642 | train 2.883807 | val 3.296453 | tok/s 2628689.28 | norm 0.3587 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25643 | train 3.056436 | val 3.296453 | tok/s 2625492.04 | norm 0.3319 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25644 | train 2.940320 | val 3.296453 | tok/s 2625117.17 | norm 0.3069 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25645 | train 3.012109 | val 3.296453 | tok/s 2625114.94 | norm 0.3549 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25646 | train 2.858601 | val 3.296453 | tok/s 2625113.47 | norm 0.3935 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25647 | train 2.932105 | val 3.296453 | tok/s 2628358.43 | norm 0.3884 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25648 | train 2.925802 | val 3.296453 | tok/s 2624506.71 | norm 0.3514 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25649 | train 2.996749 | val 3.296453 | tok/s 2627638.49 | norm 0.3535 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25650 | train 2.976590 | val 3.296453 | tok/s 2630572.81 | norm 0.3531 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25651 | train 2.861668 | val 3.296453 | tok/s 2629700.35 | norm 0.3636 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25652 | train 2.874007 | val 3.296453 | tok/s 2630247.78 | norm 0.3553 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25653 | train 2.817406 | val 3.296453 | tok/s 2621097.93 | norm 0.3523 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25654 | train 2.940951 | val 3.296453 | tok/s 2626628.80 | norm 0.3542 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25655 | train 2.916574 | val 3.296453 | tok/s 2628809.82 | norm 0.3482 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25656 | train 2.925396 | val 3.296453 | tok/s 2625912.36 | norm 0.3117 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25657 | train 2.841820 | val 3.296453 | tok/s 2626225.31 | norm 0.3287 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25658 | train 2.740275 | val 3.296453 | tok/s 2625405.92 | norm 0.3758 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25659 | train 2.875446 | val 3.296453 | tok/s 2628322.38 | norm 0.3689 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25660 | train 2.898059 | val 3.296453 | tok/s 2628836.64 | norm 0.3512 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25661 | train 2.937768 | val 3.296453 | tok/s 2618711.81 | norm 0.3887 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25662 | train 2.917378 | val 3.296453 | tok/s 2628523.92 | norm 0.4145 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25663 | train 2.922875 | val 3.296453 | tok/s 2622804.45 | norm 0.3881 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25664 | train 3.015602 | val 3.296453 | tok/s 2626622.26 | norm 0.3769 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25665 | train 2.844650 | val 3.296453 | tok/s 2628132.04 | norm 0.3639 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25666 | train 3.023200 | val 3.296453 | tok/s 2619852.19 | norm 0.3868 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25667 | train 3.062353 | val 3.296453 | tok/s 2630181.36 | norm 0.4184 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25668 | train 2.900108 | val 3.296453 | tok/s 2626824.60 | norm 0.3341 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25669 | train 2.909096 | val 3.296453 | tok/s 2628083.41 | norm 0.3445 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25670 | train 2.950239 | val 3.296453 | tok/s 2619944.55 | norm 0.3262 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25671 | train 2.847818 | val 3.296453 | tok/s 2620708.19 | norm 0.3464 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25672 | train 2.961930 | val 3.296453 | tok/s 2630287.93 | norm 0.3279 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25673 | train 2.783760 | val 3.296453 | tok/s 2625165.32 | norm 0.3499 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25674 | train 2.902570 | val 3.296453 | tok/s 2629696.99 | norm 0.3929 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25675 | train 2.916039 | val 3.296453 | tok/s 2626360.78 | norm 0.4011 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25676 | train 2.979007 | val 3.296453 | tok/s 2595804.24 | norm 0.3839 | dt 201.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25677 | train 2.956398 | val 3.296453 | tok/s 2631003.24 | norm 0.3611 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25678 | train 2.896863 | val 3.296453 | tok/s 2626843.30 | norm 0.3879 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25679 | train 2.849449 | val 3.296453 | tok/s 2626807.21 | norm 0.3469 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25680 | train 2.780380 | val 3.296453 | tok/s 2617572.52 | norm 0.3679 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25681 | train 2.808218 | val 3.296453 | tok/s 2623656.90 | norm 0.3619 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25682 | train 2.941185 | val 3.296453 | tok/s 2626763.40 | norm 0.3486 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25683 | train 2.824989 | val 3.296453 | tok/s 2630205.29 | norm 0.3998 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25684 | train 2.851347 | val 3.296453 | tok/s 2629932.74 | norm 0.3685 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25685 | train 2.939291 | val 3.296453 | tok/s 2617420.82 | norm 0.3770 | dt 200.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25686 | train 3.363359 | val 3.296453 | tok/s 2623254.58 | norm 0.4390 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25687 | train 3.050494 | val 3.296453 | tok/s 2632687.13 | norm 0.3996 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25688 | train 2.929851 | val 3.296453 | tok/s 2626267.20 | norm 0.3432 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25689 | train 2.867637 | val 3.296453 | tok/s 2629964.21 | norm 0.3434 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25690 | train 3.053355 | val 3.296453 | tok/s 2624994.03 | norm 0.3637 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25691 | train 2.932099 | val 3.296453 | tok/s 2623799.77 | norm 0.3740 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25692 | train 3.059327 | val 3.296453 | tok/s 2631551.62 | norm 0.3660 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25693 | train 2.955157 | val 3.296453 | tok/s 2627252.13 | norm 0.3178 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25694 | train 2.904091 | val 3.296453 | tok/s 2627765.14 | norm 0.3387 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25695 | train 2.994938 | val 3.296453 | tok/s 2620730.02 | norm 0.3784 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25696 | train 2.986773 | val 3.296453 | tok/s 2626206.12 | norm 0.3722 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25697 | train 2.907458 | val 3.296453 | tok/s 2630245.59 | norm 0.3642 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25698 | train 3.001760 | val 3.296453 | tok/s 2623961.14 | norm 0.3878 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25699 | train 2.998772 | val 3.296453 | tok/s 2624184.07 | norm 0.3642 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25700 | train 2.816937 | val 3.296453 | tok/s 2620214.29 | norm 0.3453 | dt 200.09 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25701 | train 2.927111 | val 3.296453 | tok/s 2597532.85 | norm 0.3515 | dt 201.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25702 | train 2.925422 | val 3.296453 | tok/s 2628312.02 | norm 0.3451 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25703 | train 2.947047 | val 3.296453 | tok/s 2630049.38 | norm 0.3774 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25704 | train 2.794806 | val 3.296453 | tok/s 2627273.02 | norm 0.3642 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25705 | train 2.913151 | val 3.296453 | tok/s 2625875.62 | norm 0.3796 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25706 | train 3.009218 | val 3.296453 | tok/s 2624241.23 | norm 0.3861 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25707 | train 2.991519 | val 3.296453 | tok/s 2627230.01 | norm 0.3789 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25708 | train 2.996970 | val 3.296453 | tok/s 2627870.78 | norm 0.3922 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25709 | train 2.834874 | val 3.296453 | tok/s 2627853.49 | norm 0.3815 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25710 | train 3.000978 | val 3.296453 | tok/s 2627868.04 | norm 0.4114 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25711 | train 2.887357 | val 3.296453 | tok/s 2626141.29 | norm 0.3335 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25712 | train 2.904445 | val 3.296453 | tok/s 2631395.46 | norm 0.3639 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25713 | train 3.015772 | val 3.296453 | tok/s 2630181.91 | norm 0.3828 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25714 | train 3.001246 | val 3.296453 | tok/s 2628988.19 | norm 0.3713 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25715 | train 3.013628 | val 3.296453 | tok/s 2629604.08 | norm 0.3405 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25716 | train 2.843758 | val 3.296453 | tok/s 2627822.64 | norm 0.3274 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25717 | train 2.806550 | val 3.296453 | tok/s 2630317.60 | norm 0.3471 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25718 | train 3.081941 | val 3.296453 | tok/s 2618468.88 | norm 0.3862 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25719 | train 3.005236 | val 3.296453 | tok/s 2630183.55 | norm 0.3796 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25720 | train 2.957084 | val 3.296453 | tok/s 2627545.05 | norm 0.3876 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25721 | train 2.950837 | val 3.296453 | tok/s 2627950.46 | norm 0.3650 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25722 | train 3.030912 | val 3.296453 | tok/s 2629201.32 | norm 0.3544 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25723 | train 2.916580 | val 3.296453 | tok/s 2617879.17 | norm 0.3620 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25724 | train 2.933062 | val 3.296453 | tok/s 2629716.77 | norm 0.4096 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25725 | train 2.912206 | val 3.296453 | tok/s 2627375.16 | norm 0.4255 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25726 | train 2.902297 | val 3.296453 | tok/s 2631140.32 | norm 0.4126 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25727 | train 2.900193 | val 3.296453 | tok/s 2628148.98 | norm 0.4300 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25728 | train 2.948967 | val 3.296453 | tok/s 2622539.74 | norm 0.4151 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25729 | train 2.922940 | val 3.296453 | tok/s 2629755.48 | norm 0.3483 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25730 | train 3.004849 | val 3.296453 | tok/s 2627725.48 | norm 0.3300 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25731 | train 2.912487 | val 3.296453 | tok/s 2628492.31 | norm 0.3415 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25732 | train 2.934801 | val 3.296453 | tok/s 2630259.23 | norm 0.3628 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25733 | train 2.948514 | val 3.296453 | tok/s 2625818.47 | norm 0.3370 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25734 | train 2.903510 | val 3.296453 | tok/s 2630603.18 | norm 0.3301 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25735 | train 2.906414 | val 3.296453 | tok/s 2626391.35 | norm 0.3469 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25736 | train 2.890439 | val 3.296453 | tok/s 2629116.14 | norm 0.3794 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25737 | train 2.943303 | val 3.296453 | tok/s 2625721.83 | norm 0.3384 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25738 | train 2.940635 | val 3.296453 | tok/s 2625824.79 | norm 0.3360 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25739 | train 2.876734 | val 3.296453 | tok/s 2627616.26 | norm 0.3858 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25740 | train 2.977311 | val 3.296453 | tok/s 2625093.13 | norm 0.3394 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25741 | train 2.840489 | val 3.296453 | tok/s 2615911.09 | norm 0.3426 | dt 200.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25742 | train 2.890213 | val 3.296453 | tok/s 2619325.38 | norm 0.3417 | dt 200.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25743 | train 2.892827 | val 3.296453 | tok/s 2622605.94 | norm 0.3377 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25744 | train 2.862025 | val 3.296453 | tok/s 2628419.85 | norm 0.3018 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25745 | train 2.898959 | val 3.296453 | tok/s 2493745.52 | norm 0.3311 | dt 210.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25746 | train 3.149677 | val 3.296453 | tok/s 2509031.67 | norm 0.4520 | dt 208.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25747 | train 2.908458 | val 3.296453 | tok/s 2620762.55 | norm 0.3661 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25748 | train 2.925710 | val 3.296453 | tok/s 2623858.07 | norm 0.3168 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25749 | train 2.915337 | val 3.296453 | tok/s 2632867.06 | norm 0.3303 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 25750 | val_loss 3.291958 step 25750 | train 2.998516 | val 3.291958 | tok/s 2608395.53 | norm 0.3242 | dt 201.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25751 | train 2.929721 | val 3.291958 | tok/s 2485088.22 | norm 0.3130 | dt 210.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25752 | train 2.892437 | val 3.291958 | tok/s 2573514.22 | norm 0.3263 | dt 203.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25753 | train 2.984880 | val 3.291958 | tok/s 2624695.12 | norm 0.3386 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25754 | train 2.916786 | val 3.291958 | tok/s 2622308.14 | norm 0.4075 | dt 199.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25755 | train 2.998132 | val 3.291958 | tok/s 2628189.53 | norm 0.3706 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25756 | train 3.012569 | val 3.291958 | tok/s 2623789.05 | norm 0.3487 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25757 | train 2.854854 | val 3.291958 | tok/s 2627921.04 | norm 0.3570 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25758 | train 2.986670 | val 3.291958 | tok/s 2623269.09 | norm 0.3580 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25759 | train 2.899187 | val 3.291958 | tok/s 2628743.35 | norm 0.3469 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25760 | train 2.881681 | val 3.291958 | tok/s 2629519.66 | norm 0.3283 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25761 | train 2.960722 | val 3.291958 | tok/s 2625446.72 | norm 0.3429 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25762 | train 2.944801 | val 3.291958 | tok/s 2628230.69 | norm 0.3358 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25763 | train 2.829690 | val 3.291958 | tok/s 2624717.38 | norm 0.3343 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25764 | train 2.902609 | val 3.291958 | tok/s 2618688.09 | norm 0.3791 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25765 | train 2.934018 | val 3.291958 | tok/s 2630507.81 | norm 0.3902 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25766 | train 2.937875 | val 3.291958 | tok/s 2625188.94 | norm 0.3580 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25767 | train 2.954071 | val 3.291958 | tok/s 2625394.42 | norm 0.3886 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25768 | train 2.910868 | val 3.291958 | tok/s 2624828.39 | norm 0.3769 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25769 | train 2.950736 | val 3.291958 | tok/s 2624877.12 | norm 0.3883 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25770 | train 2.854513 | val 3.291958 | tok/s 2622797.13 | norm 0.3868 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25771 | train 2.930589 | val 3.291958 | tok/s 2619560.23 | norm 0.3650 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25772 | train 2.906427 | val 3.291958 | tok/s 2634446.99 | norm 0.3490 | dt 199.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25773 | train 3.002254 | val 3.291958 | tok/s 2631475.78 | norm 0.3680 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25774 | train 2.904974 | val 3.291958 | tok/s 2627847.09 | norm 0.3536 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25775 | train 2.911065 | val 3.291958 | tok/s 2624487.46 | norm 0.3556 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25776 | train 2.982920 | val 3.291958 | tok/s 2623219.92 | norm 0.3792 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25777 | train 2.900660 | val 3.291958 | tok/s 2632554.65 | norm 0.4080 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25778 | train 2.880229 | val 3.291958 | tok/s 2626834.06 | norm 0.3788 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25779 | train 2.983543 | val 3.291958 | tok/s 2626803.31 | norm 0.3463 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25780 | train 2.988536 | val 3.291958 | tok/s 2619037.96 | norm 0.3554 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25781 | train 2.916205 | val 3.291958 | tok/s 2624415.92 | norm 0.3802 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25782 | train 2.940764 | val 3.291958 | tok/s 2629353.86 | norm 0.3920 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25783 | train 2.978172 | val 3.291958 | tok/s 2631623.03 | norm 0.3603 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25784 | train 3.005941 | val 3.291958 | tok/s 2628979.51 | norm 0.3712 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25785 | train 2.816334 | val 3.291958 | tok/s 2624065.77 | norm 0.3816 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25786 | train 2.806026 | val 3.291958 | tok/s 2617989.67 | norm 0.3801 | dt 200.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25787 | train 3.302882 | val 3.291958 | tok/s 2532308.34 | norm 0.4863 | dt 207.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25788 | train 2.877442 | val 3.291958 | tok/s 2627709.28 | norm 0.3505 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25789 | train 2.880268 | val 3.291958 | tok/s 2629235.64 | norm 0.3068 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25790 | train 2.936073 | val 3.291958 | tok/s 2620152.94 | norm 0.3776 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25791 | train 3.049969 | val 3.291958 | tok/s 2623509.63 | norm 0.3992 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25792 | train 2.914252 | val 3.291958 | tok/s 2628967.63 | norm 0.3647 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25793 | train 3.116554 | val 3.291958 | tok/s 2490000.43 | norm 0.5294 | dt 210.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25794 | train 2.946228 | val 3.291958 | tok/s 2630777.05 | norm 0.3269 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25795 | train 2.833570 | val 3.291958 | tok/s 2627024.83 | norm 0.3978 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25796 | train 2.949793 | val 3.291958 | tok/s 2627151.89 | norm 0.3591 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25797 | train 2.918715 | val 3.291958 | tok/s 2628783.26 | norm 0.3131 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25798 | train 2.964769 | val 3.291958 | tok/s 2628000.58 | norm 0.3254 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25799 | train 2.919959 | val 3.291958 | tok/s 2625841.16 | norm 0.3712 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25800 | train 2.905245 | val 3.291958 | tok/s 2624344.90 | norm 0.3613 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25801 | train 3.035497 | val 3.291958 | tok/s 2624977.98 | norm 0.3868 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25802 | train 3.021075 | val 3.291958 | tok/s 2630637.87 | norm 0.3489 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25803 | train 2.974624 | val 3.291958 | tok/s 2627962.87 | norm 0.3350 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25804 | train 2.892296 | val 3.291958 | tok/s 2625105.03 | norm 0.3744 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25805 | train 2.918041 | val 3.291958 | tok/s 2625895.52 | norm 0.3657 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25806 | train 2.993511 | val 3.291958 | tok/s 2628087.59 | norm 0.3593 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25807 | train 2.785472 | val 3.291958 | tok/s 2630379.61 | norm 0.3652 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25808 | train 2.859358 | val 3.291958 | tok/s 2627351.25 | norm 0.3709 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25809 | train 3.043658 | val 3.291958 | tok/s 2629253.25 | norm 0.3764 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25810 | train 2.954050 | val 3.291958 | tok/s 2622884.63 | norm 0.3162 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25811 | train 3.001080 | val 3.291958 | tok/s 2622756.82 | norm 0.3177 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25812 | train 2.950778 | val 3.291958 | tok/s 2629013.33 | norm 0.3497 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25813 | train 3.003400 | val 3.291958 | tok/s 2625774.31 | norm 0.3404 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25814 | train 2.899307 | val 3.291958 | tok/s 2631887.33 | norm 0.3729 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25815 | train 2.776226 | val 3.291958 | tok/s 2623487.26 | norm 0.3598 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25816 | train 2.870790 | val 3.291958 | tok/s 2630092.35 | norm 0.3660 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25817 | train 2.757323 | val 3.291958 | tok/s 2623336.34 | norm 0.3737 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25818 | train 2.879498 | val 3.291958 | tok/s 2623094.08 | norm 0.3551 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25819 | train 3.017823 | val 3.291958 | tok/s 2630421.32 | norm 0.3726 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25820 | train 3.174997 | val 3.291958 | tok/s 2626614.54 | norm 0.4281 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25821 | train 3.033755 | val 3.291958 | tok/s 2625801.32 | norm 0.3515 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25822 | train 2.947010 | val 3.291958 | tok/s 2624235.39 | norm 0.3206 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25823 | train 2.900096 | val 3.291958 | tok/s 2593688.17 | norm 0.3539 | dt 202.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25824 | train 2.983566 | val 3.291958 | tok/s 2631624.85 | norm 0.3785 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25825 | train 2.946579 | val 3.291958 | tok/s 2627697.05 | norm 0.3780 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25826 | train 2.984135 | val 3.291958 | tok/s 2628995.23 | norm 0.3954 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25827 | train 2.827133 | val 3.291958 | tok/s 2615382.96 | norm 0.3739 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25828 | train 2.822960 | val 3.291958 | tok/s 2620719.91 | norm 0.3664 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25829 | train 2.924976 | val 3.291958 | tok/s 2627364.06 | norm 0.3745 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25830 | train 2.943475 | val 3.291958 | tok/s 2628745.94 | norm 0.3459 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25831 | train 2.860070 | val 3.291958 | tok/s 2627538.12 | norm 0.3523 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25832 | train 2.854370 | val 3.291958 | tok/s 2618433.20 | norm 0.3199 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25833 | train 2.963632 | val 3.291958 | tok/s 2621602.40 | norm 0.3236 | dt 199.99 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25834 | train 2.963605 | val 3.291958 | tok/s 2626957.57 | norm 0.3594 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25835 | train 2.844481 | val 3.291958 | tok/s 2628313.70 | norm 0.3578 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25836 | train 2.842320 | val 3.291958 | tok/s 2624744.38 | norm 0.4211 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25837 | train 2.914095 | val 3.291958 | tok/s 2616736.60 | norm 0.3094 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25838 | train 2.911696 | val 3.291958 | tok/s 2605617.26 | norm 0.3302 | dt 201.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25839 | train 2.881145 | val 3.291958 | tok/s 2628253.05 | norm 0.3461 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25840 | train 3.007569 | val 3.291958 | tok/s 2626709.45 | norm 0.3552 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25841 | train 2.893280 | val 3.291958 | tok/s 2628170.61 | norm 0.3361 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25842 | train 2.957259 | val 3.291958 | tok/s 2625436.18 | norm 0.3375 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25843 | train 2.978095 | val 3.291958 | tok/s 2607640.21 | norm 0.3788 | dt 201.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25844 | train 2.740999 | val 3.291958 | tok/s 2629747.29 | norm 0.3407 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25845 | train 2.947378 | val 3.291958 | tok/s 2487936.56 | norm 0.3718 | dt 210.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25846 | train 2.900223 | val 3.291958 | tok/s 2628134.92 | norm 0.4097 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25847 | train 2.859864 | val 3.291958 | tok/s 2612344.27 | norm 0.3677 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25848 | train 3.131731 | val 3.291958 | tok/s 2626903.97 | norm 0.3950 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25849 | train 3.012842 | val 3.291958 | tok/s 2633728.53 | norm 0.4014 | dt 199.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25850 | train 2.918028 | val 3.291958 | tok/s 2627931.44 | norm 0.3845 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25851 | train 2.972896 | val 3.291958 | tok/s 2630669.68 | norm 0.3663 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25852 | train 2.880529 | val 3.291958 | tok/s 2607385.54 | norm 0.3712 | dt 201.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25853 | train 2.959793 | val 3.291958 | tok/s 2629754.64 | norm 0.3729 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25854 | train 2.993929 | val 3.291958 | tok/s 2623725.92 | norm 0.3798 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25855 | train 2.920569 | val 3.291958 | tok/s 2621794.99 | norm 0.3951 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25856 | train 2.830079 | val 3.291958 | tok/s 2626427.77 | norm 0.3846 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25857 | train 2.884187 | val 3.291958 | tok/s 2621193.20 | norm 0.3896 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25858 | train 2.931724 | val 3.291958 | tok/s 2622702.42 | norm 0.3613 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25859 | train 2.904189 | val 3.291958 | tok/s 2629035.51 | norm 0.3349 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25860 | train 2.949977 | val 3.291958 | tok/s 2625399.88 | norm 0.3472 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25861 | train 2.878609 | val 3.291958 | tok/s 2630471.21 | norm 0.3450 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25862 | train 2.872886 | val 3.291958 | tok/s 2621287.85 | norm 0.3713 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25863 | train 2.901931 | val 3.291958 | tok/s 2628954.14 | norm 0.3587 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25864 | train 2.924994 | val 3.291958 | tok/s 2627221.00 | norm 0.3625 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25865 | train 2.939205 | val 3.291958 | tok/s 2621193.29 | norm 0.3662 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25866 | train 2.898848 | val 3.291958 | tok/s 2133117.83 | norm 0.3640 | dt 245.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25867 | train 2.862093 | val 3.291958 | tok/s 2493321.11 | norm 0.3377 | dt 210.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25868 | train 2.848448 | val 3.291958 | tok/s 2511732.76 | norm 0.3325 | dt 208.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25869 | train 2.854527 | val 3.291958 | tok/s 2627271.28 | norm 0.3753 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25870 | train 2.903929 | val 3.291958 | tok/s 2624393.66 | norm 0.3923 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25871 | train 2.921364 | val 3.291958 | tok/s 2627255.39 | norm 0.3935 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25872 | train 2.819558 | val 3.291958 | tok/s 2627021.25 | norm 0.3867 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25873 | train 2.894940 | val 3.291958 | tok/s 2627667.77 | norm 0.3783 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25874 | train 3.044809 | val 3.291958 | tok/s 2538578.77 | norm 0.3705 | dt 206.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25875 | train 2.904753 | val 3.291958 | tok/s 2624829.68 | norm 0.3796 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25876 | train 2.799360 | val 3.291958 | tok/s 2629197.21 | norm 0.3967 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25877 | train 2.921756 | val 3.291958 | tok/s 2628422.94 | norm 0.3986 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25878 | train 2.852128 | val 3.291958 | tok/s 2631655.03 | norm 0.3705 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25879 | train 2.883074 | val 3.291958 | tok/s 2627393.77 | norm 0.3609 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25880 | train 3.011915 | val 3.291958 | tok/s 2626695.23 | norm 0.3819 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25881 | train 3.051379 | val 3.291958 | tok/s 2505520.93 | norm 0.3764 | dt 209.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25882 | train 2.896347 | val 3.291958 | tok/s 2619896.79 | norm 0.3995 | dt 200.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25883 | train 2.952863 | val 3.291958 | tok/s 2628460.00 | norm 0.3891 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25884 | train 2.823603 | val 3.291958 | tok/s 2625470.62 | norm 0.3597 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25885 | train 2.985993 | val 3.291958 | tok/s 2627246.75 | norm 0.3997 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25886 | train 2.943711 | val 3.291958 | tok/s 2626100.25 | norm 0.3889 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25887 | train 2.894146 | val 3.291958 | tok/s 2624327.78 | norm 0.3518 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25888 | train 2.903687 | val 3.291958 | tok/s 2635080.87 | norm 0.3805 | dt 198.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25889 | train 2.913425 | val 3.291958 | tok/s 2630235.67 | norm 0.3585 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25890 | train 2.927686 | val 3.291958 | tok/s 2625119.16 | norm 0.3535 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25891 | train 2.901188 | val 3.291958 | tok/s 2626139.75 | norm 0.3602 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25892 | train 2.844025 | val 3.291958 | tok/s 2623618.44 | norm 0.3826 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25893 | train 3.070129 | val 3.291958 | tok/s 2630618.52 | norm 0.3494 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25894 | train 2.965505 | val 3.291958 | tok/s 2624514.71 | norm 0.3538 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25895 | train 2.870989 | val 3.291958 | tok/s 2628559.88 | norm 0.3392 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25896 | train 2.999419 | val 3.291958 | tok/s 2621718.45 | norm 0.3743 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25897 | train 2.969847 | val 3.291958 | tok/s 2625043.11 | norm 0.3521 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25898 | train 2.985291 | val 3.291958 | tok/s 2631600.52 | norm 0.3359 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25899 | train 2.953215 | val 3.291958 | tok/s 2625448.88 | norm 0.3653 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25900 | train 2.851618 | val 3.291958 | tok/s 2628050.19 | norm 0.3569 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25901 | train 2.959659 | val 3.291958 | tok/s 2617846.74 | norm 0.3574 | dt 200.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25902 | train 2.908486 | val 3.291958 | tok/s 2623671.33 | norm 0.3661 | dt 199.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25903 | train 3.013180 | val 3.291958 | tok/s 2625906.92 | norm 0.3665 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25904 | train 2.888621 | val 3.291958 | tok/s 2628113.70 | norm 0.3730 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25905 | train 2.912821 | val 3.291958 | tok/s 2629408.66 | norm 0.3764 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25906 | train 2.947762 | val 3.291958 | tok/s 2627523.95 | norm 0.3980 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25907 | train 2.872634 | val 3.291958 | tok/s 2624747.26 | norm 0.3830 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25908 | train 2.896835 | val 3.291958 | tok/s 2632516.17 | norm 0.3563 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25909 | train 2.858217 | val 3.291958 | tok/s 2626200.99 | norm 0.3591 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25910 | train 3.011489 | val 3.291958 | tok/s 2629794.11 | norm 0.3818 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25911 | train 3.030072 | val 3.291958 | tok/s 2482050.91 | norm 0.3718 | dt 211.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25912 | train 2.902923 | val 3.291958 | tok/s 2625910.95 | norm 0.3530 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25913 | train 2.939591 | val 3.291958 | tok/s 2631924.09 | norm 0.3603 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25914 | train 2.898877 | val 3.291958 | tok/s 2629015.35 | norm 0.3335 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25915 | train 2.972701 | val 3.291958 | tok/s 2628885.60 | norm 0.3528 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25916 | train 3.188412 | val 3.291958 | tok/s 2626283.63 | norm 0.3930 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25917 | train 2.867885 | val 3.291958 | tok/s 2627388.49 | norm 0.3914 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25918 | train 3.027188 | val 3.291958 | tok/s 2628084.64 | norm 0.3900 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25919 | train 2.999519 | val 3.291958 | tok/s 2625999.36 | norm 0.3795 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25920 | train 2.841688 | val 3.291958 | tok/s 2630330.99 | norm 0.3611 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25921 | train 2.830943 | val 3.291958 | tok/s 2623085.95 | norm 0.3562 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25922 | train 2.846712 | val 3.291958 | tok/s 2625758.41 | norm 0.3498 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25923 | train 2.873349 | val 3.291958 | tok/s 2626883.66 | norm 0.3779 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25924 | train 2.898814 | val 3.291958 | tok/s 2627345.63 | norm 0.3687 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25925 | train 2.969122 | val 3.291958 | tok/s 2629638.29 | norm 0.3476 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25926 | train 3.029529 | val 3.291958 | tok/s 2624293.71 | norm 0.3470 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25927 | train 2.867113 | val 3.291958 | tok/s 2628857.73 | norm 0.3671 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25928 | train 2.859473 | val 3.291958 | tok/s 2629259.56 | norm 0.3500 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25929 | train 2.873724 | val 3.291958 | tok/s 2625196.30 | norm 0.3417 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25930 | train 2.902678 | val 3.291958 | tok/s 2628059.17 | norm 0.4056 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25931 | train 2.985737 | val 3.291958 | tok/s 2627284.96 | norm 0.3777 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25932 | train 2.846855 | val 3.291958 | tok/s 2628910.27 | norm 0.3814 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25933 | train 2.905698 | val 3.291958 | tok/s 2630107.15 | norm 0.3869 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25934 | train 2.953687 | val 3.291958 | tok/s 2628070.81 | norm 0.3898 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25935 | train 2.924042 | val 3.291958 | tok/s 2630879.27 | norm 0.3607 | dt 199.28 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25936 | train 2.936785 | val 3.291958 | tok/s 2628034.69 | norm 0.3948 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25937 | train 2.899574 | val 3.291958 | tok/s 2626697.53 | norm 0.4414 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25938 | train 2.960251 | val 3.291958 | tok/s 2629385.64 | norm 0.3521 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25939 | train 2.925154 | val 3.291958 | tok/s 2621783.21 | norm 0.2824 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25940 | train 2.866901 | val 3.291958 | tok/s 2629030.97 | norm 0.3402 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25941 | train 2.846508 | val 3.291958 | tok/s 2627683.32 | norm 0.3708 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25942 | train 2.877666 | val 3.291958 | tok/s 2628921.67 | norm 0.3736 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25943 | train 2.944437 | val 3.291958 | tok/s 2622776.55 | norm 0.3617 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25944 | train 2.952540 | val 3.291958 | tok/s 2622574.11 | norm 0.3348 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25945 | train 2.877368 | val 3.291958 | tok/s 2632275.86 | norm 0.3689 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25946 | train 2.857902 | val 3.291958 | tok/s 2630022.65 | norm 0.3701 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25947 | train 2.997952 | val 3.291958 | tok/s 2628962.62 | norm 0.3574 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25948 | train 2.898186 | val 3.291958 | tok/s 2622726.21 | norm 0.3724 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25949 | train 2.921782 | val 3.291958 | tok/s 2624987.26 | norm 0.3434 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25950 | train 2.890944 | val 3.291958 | tok/s 2630099.27 | norm 0.3734 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25951 | train 3.007676 | val 3.291958 | tok/s 2605379.04 | norm 0.4494 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25952 | train 2.994410 | val 3.291958 | tok/s 2629373.12 | norm 0.3303 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25953 | train 2.967669 | val 3.291958 | tok/s 2586357.46 | norm 0.3289 | dt 202.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25954 | train 3.002871 | val 3.291958 | tok/s 2538285.32 | norm 0.3806 | dt 206.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25955 | train 2.846582 | val 3.291958 | tok/s 2629183.47 | norm 0.3811 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25956 | train 2.887458 | val 3.291958 | tok/s 2623314.87 | norm 0.3777 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25957 | train 2.962607 | val 3.291958 | tok/s 2626478.61 | norm 0.3846 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25958 | train 2.920170 | val 3.291958 | tok/s 2615543.14 | norm 0.3651 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25959 | train 2.819527 | val 3.291958 | tok/s 2624981.34 | norm 0.3613 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25960 | train 2.923097 | val 3.291958 | tok/s 2629347.16 | norm 0.3654 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25961 | train 2.932011 | val 3.291958 | tok/s 2627608.57 | norm 0.3571 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25962 | train 2.890974 | val 3.291958 | tok/s 2625498.34 | norm 0.3839 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25963 | train 2.905385 | val 3.291958 | tok/s 2620580.99 | norm 0.3726 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25964 | train 3.030010 | val 3.291958 | tok/s 2625708.42 | norm 0.3779 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25965 | train 2.848965 | val 3.291958 | tok/s 2632620.95 | norm 0.3493 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25966 | train 2.976877 | val 3.291958 | tok/s 2626426.80 | norm 0.3553 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25967 | train 2.939548 | val 3.291958 | tok/s 2629384.65 | norm 0.3767 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25968 | train 2.894701 | val 3.291958 | tok/s 2626617.19 | norm 0.3488 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25969 | train 2.980541 | val 3.291958 | tok/s 2627677.67 | norm 0.3303 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25970 | train 3.103929 | val 3.291958 | tok/s 2629935.84 | norm 0.4191 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25971 | train 3.017345 | val 3.291958 | tok/s 2624155.00 | norm 0.3507 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25972 | train 2.916237 | val 3.291958 | tok/s 2629966.79 | norm 0.3560 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25973 | train 3.054289 | val 3.291958 | tok/s 2628276.67 | norm 0.4175 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25974 | train 2.928435 | val 3.291958 | tok/s 2628072.98 | norm 0.3753 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25975 | train 3.016760 | val 3.291958 | tok/s 2629270.24 | norm 0.3739 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25976 | train 2.970648 | val 3.291958 | tok/s 2625150.66 | norm 0.4404 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25977 | train 2.958319 | val 3.291958 | tok/s 2630431.63 | norm 0.3609 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25978 | train 2.918602 | val 3.291958 | tok/s 2572185.63 | norm 0.3620 | dt 203.83 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25979 | train 2.960535 | val 3.291958 | tok/s 2594943.08 | norm 0.3371 | dt 202.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25980 | train 2.959330 | val 3.291958 | tok/s 2630825.18 | norm 0.3439 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25981 | train 2.935026 | val 3.291958 | tok/s 2621014.61 | norm 0.3616 | dt 200.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25982 | train 2.870033 | val 3.291958 | tok/s 2631021.01 | norm 0.3844 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25983 | train 2.888087 | val 3.291958 | tok/s 2621907.85 | norm 0.3433 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25984 | train 3.038896 | val 3.291958 | tok/s 2626293.81 | norm 0.3312 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25985 | train 2.982740 | val 3.291958 | tok/s 2626399.76 | norm 0.3317 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25986 | train 2.952952 | val 3.291958 | tok/s 2618186.17 | norm 0.3658 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25987 | train 2.883731 | val 3.291958 | tok/s 2613920.13 | norm 0.3792 | dt 200.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25988 | train 3.014357 | val 3.291958 | tok/s 2631120.76 | norm 0.3492 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25989 | train 2.875045 | val 3.291958 | tok/s 2630323.17 | norm 0.3679 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25990 | train 3.110513 | val 3.291958 | tok/s 2630046.70 | norm 0.3564 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25991 | train 2.990432 | val 3.291958 | tok/s 2625769.18 | norm 0.3611 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25992 | train 2.949526 | val 3.291958 | tok/s 2632746.22 | norm 0.3551 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25993 | train 2.849249 | val 3.291958 | tok/s 2632068.16 | norm 0.3979 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25994 | train 2.744257 | val 3.291958 | tok/s 2628701.21 | norm 0.4008 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25995 | train 2.941390 | val 3.291958 | tok/s 2619142.34 | norm 0.3857 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25996 | train 2.947558 | val 3.291958 | tok/s 2625221.71 | norm 0.4039 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25997 | train 2.937736 | val 3.291958 | tok/s 2629632.67 | norm 0.3768 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25998 | train 3.038284 | val 3.291958 | tok/s 2629744.30 | norm 0.3920 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 25999 | train 3.024522 | val 3.291958 | tok/s 2624591.48 | norm 0.3709 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 26000 | val_loss 3.298572 step 26000 | train 2.982946 | val 3.298572 | tok/s 2615589.05 | norm 0.4198 | dt 200.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26001 | train 2.916683 | val 3.298572 | tok/s 2623070.84 | norm 0.3901 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26002 | train 2.818300 | val 3.298572 | tok/s 2623446.32 | norm 0.3729 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26003 | train 2.848234 | val 3.298572 | tok/s 2627313.08 | norm 0.3501 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26004 | train 2.926069 | val 3.298572 | tok/s 2626351.58 | norm 0.3410 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26005 | train 2.931156 | val 3.298572 | tok/s 2629537.58 | norm 0.3328 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26006 | train 2.998642 | val 3.298572 | tok/s 2624811.02 | norm 0.3421 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26007 | train 2.844383 | val 3.298572 | tok/s 2619962.90 | norm 0.3551 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26008 | train 2.966257 | val 3.298572 | tok/s 2629039.75 | norm 0.3713 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26009 | train 2.771858 | val 3.298572 | tok/s 2628208.31 | norm 0.3784 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26010 | train 2.932882 | val 3.298572 | tok/s 2630390.98 | norm 0.3820 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26011 | train 2.860875 | val 3.298572 | tok/s 2620970.06 | norm 0.3754 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26012 | train 2.966540 | val 3.298572 | tok/s 2626357.03 | norm 0.3622 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26013 | train 2.935367 | val 3.298572 | tok/s 2625702.21 | norm 0.3753 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26014 | train 2.857906 | val 3.298572 | tok/s 2625564.96 | norm 0.3410 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26015 | train 2.949282 | val 3.298572 | tok/s 2627777.29 | norm 0.3522 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26016 | train 3.008714 | val 3.298572 | tok/s 2619602.60 | norm 0.3723 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26017 | train 2.842181 | val 3.298572 | tok/s 2625308.60 | norm 0.3547 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26018 | train 2.908280 | val 3.298572 | tok/s 2629918.60 | norm 0.3680 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26019 | train 2.904111 | val 3.298572 | tok/s 2625973.69 | norm 0.3727 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26020 | train 3.001550 | val 3.298572 | tok/s 2442121.46 | norm 0.3463 | dt 214.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26021 | train 2.983772 | val 3.298572 | tok/s 2613142.88 | norm 0.3352 | dt 200.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26022 | train 2.894172 | val 3.298572 | tok/s 2627967.82 | norm 0.3776 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26023 | train 3.049066 | val 3.298572 | tok/s 2629714.80 | norm 0.3831 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26024 | train 2.976187 | val 3.298572 | tok/s 2628587.83 | norm 0.3362 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26025 | train 3.001449 | val 3.298572 | tok/s 2630012.27 | norm 0.3750 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26026 | train 2.967409 | val 3.298572 | tok/s 2623454.71 | norm 0.3873 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26027 | train 2.844813 | val 3.298572 | tok/s 2626164.15 | norm 0.3755 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26028 | train 2.889638 | val 3.298572 | tok/s 2627286.40 | norm 0.4057 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26029 | train 2.940565 | val 3.298572 | tok/s 2626627.13 | norm 0.4127 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26030 | train 2.940095 | val 3.298572 | tok/s 2630090.29 | norm 0.3654 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26031 | train 2.911891 | val 3.298572 | tok/s 2623856.60 | norm 0.3269 | dt 199.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26032 | train 2.992195 | val 3.298572 | tok/s 2625505.58 | norm 0.3342 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26033 | train 2.941171 | val 3.298572 | tok/s 2630242.41 | norm 0.3640 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26034 | train 2.913744 | val 3.298572 | tok/s 2626894.50 | norm 0.3684 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26035 | train 2.877125 | val 3.298572 | tok/s 2629082.44 | norm 0.3584 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26036 | train 2.955252 | val 3.298572 | tok/s 2610213.48 | norm 0.3553 | dt 200.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26037 | train 3.001766 | val 3.298572 | tok/s 2628135.75 | norm 0.3598 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26038 | train 2.988242 | val 3.298572 | tok/s 2631421.69 | norm 0.3389 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26039 | train 2.989089 | val 3.298572 | tok/s 2627309.04 | norm 0.3528 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26040 | train 3.001586 | val 3.298572 | tok/s 2625322.67 | norm 0.3610 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26041 | train 2.870608 | val 3.298572 | tok/s 2627194.16 | norm 0.3458 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26042 | train 2.912507 | val 3.298572 | tok/s 2628033.72 | norm 0.3542 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26043 | train 2.909326 | val 3.298572 | tok/s 2629355.93 | norm 0.3296 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26044 | train 2.966886 | val 3.298572 | tok/s 2616766.56 | norm 0.3519 | dt 200.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26045 | train 2.925039 | val 3.298572 | tok/s 2631178.68 | norm 0.3521 | dt 199.26 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26046 | train 2.929945 | val 3.298572 | tok/s 2630552.15 | norm 0.3473 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26047 | train 2.917748 | val 3.298572 | tok/s 2628583.44 | norm 0.3779 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26048 | train 2.992335 | val 3.298572 | tok/s 2621713.27 | norm 0.3900 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26049 | train 2.907529 | val 3.298572 | tok/s 2619586.04 | norm 0.3642 | dt 200.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26050 | train 2.954746 | val 3.298572 | tok/s 2631896.50 | norm 0.3481 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26051 | train 3.023156 | val 3.298572 | tok/s 2628698.03 | norm 0.3800 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26052 | train 2.880107 | val 3.298572 | tok/s 2630318.15 | norm 0.3358 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26053 | train 3.099940 | val 3.298572 | tok/s 2620549.10 | norm 0.3774 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26054 | train 2.824937 | val 3.298572 | tok/s 2624206.90 | norm 0.3806 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26055 | train 2.916491 | val 3.298572 | tok/s 2623126.05 | norm 0.3560 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26056 | train 2.951540 | val 3.298572 | tok/s 2626311.39 | norm 0.3663 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26057 | train 2.929779 | val 3.298572 | tok/s 2624615.44 | norm 0.3594 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26058 | train 2.945300 | val 3.298572 | tok/s 2630563.24 | norm 0.3520 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26059 | train 2.929972 | val 3.298572 | tok/s 2626701.32 | norm 0.3399 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26060 | train 2.845516 | val 3.298572 | tok/s 2625119.99 | norm 0.3431 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26061 | train 2.984963 | val 3.298572 | tok/s 2627442.87 | norm 0.3600 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26062 | train 2.936630 | val 3.298572 | tok/s 2603532.47 | norm 0.3446 | dt 201.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26063 | train 3.111487 | val 3.298572 | tok/s 2621984.77 | norm 0.5674 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26064 | train 2.875282 | val 3.298572 | tok/s 2623552.04 | norm 0.3347 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26065 | train 2.935529 | val 3.298572 | tok/s 2629193.10 | norm 0.3212 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26066 | train 2.910613 | val 3.298572 | tok/s 2626850.38 | norm 0.3725 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26067 | train 2.959695 | val 3.298572 | tok/s 2632441.10 | norm 0.3362 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26068 | train 2.900991 | val 3.298572 | tok/s 2625701.95 | norm 0.3348 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26069 | train 2.846067 | val 3.298572 | tok/s 2626552.20 | norm 0.3599 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26070 | train 2.830523 | val 3.298572 | tok/s 2631586.54 | norm 0.4295 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26071 | train 2.874604 | val 3.298572 | tok/s 2628836.96 | norm 0.3530 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26072 | train 2.925086 | val 3.298572 | tok/s 2629145.68 | norm 0.3802 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26073 | train 2.920182 | val 3.298572 | tok/s 2625535.53 | norm 0.3825 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26074 | train 2.978885 | val 3.298572 | tok/s 2625976.36 | norm 0.3746 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26075 | train 3.039164 | val 3.298572 | tok/s 2629808.73 | norm 0.3891 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26076 | train 2.967932 | val 3.298572 | tok/s 2486311.87 | norm 0.3618 | dt 210.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26077 | train 2.964037 | val 3.298572 | tok/s 2629565.65 | norm 0.3782 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26078 | train 2.942289 | val 3.298572 | tok/s 2528946.26 | norm 0.3606 | dt 207.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26079 | train 2.984224 | val 3.298572 | tok/s 2628829.31 | norm 0.3614 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26080 | train 3.027277 | val 3.298572 | tok/s 2629666.17 | norm 0.5223 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26081 | train 3.036776 | val 3.298572 | tok/s 2631876.05 | norm 0.3926 | dt 199.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26082 | train 2.963506 | val 3.298572 | tok/s 2630609.37 | norm 0.7932 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26083 | train 2.770469 | val 3.298572 | tok/s 2627526.03 | norm 1.5450 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26084 | train 2.909347 | val 3.298572 | tok/s 2627492.96 | norm 0.4266 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26085 | train 2.922267 | val 3.298572 | tok/s 2614280.92 | norm 0.3486 | dt 200.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26086 | train 2.973595 | val 3.298572 | tok/s 2620840.72 | norm 0.3179 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26087 | train 2.920261 | val 3.298572 | tok/s 2633306.76 | norm 0.3490 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26088 | train 2.887100 | val 3.298572 | tok/s 2626114.80 | norm 0.3960 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26089 | train 2.987620 | val 3.298572 | tok/s 2629502.55 | norm 0.3306 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26090 | train 2.914468 | val 3.298572 | tok/s 2623563.30 | norm 0.3363 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26091 | train 2.987677 | val 3.298572 | tok/s 2621230.73 | norm 0.3468 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26092 | train 2.918174 | val 3.298572 | tok/s 2628881.38 | norm 0.3471 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26093 | train 3.035315 | val 3.298572 | tok/s 2633940.52 | norm 0.3785 | dt 199.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26094 | train 2.940379 | val 3.298572 | tok/s 2629780.45 | norm 0.3768 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26095 | train 2.973482 | val 3.298572 | tok/s 2627323.96 | norm 0.3387 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26096 | train 2.859133 | val 3.298572 | tok/s 2620501.18 | norm 0.3321 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26097 | train 3.016002 | val 3.298572 | tok/s 2632792.50 | norm 0.4217 | dt 199.14 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26098 | train 2.783104 | val 3.298572 | tok/s 2626156.25 | norm 0.3794 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26099 | train 2.887262 | val 3.298572 | tok/s 2629473.08 | norm 0.3411 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26100 | train 2.870491 | val 3.298572 | tok/s 2621255.11 | norm 0.3654 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26101 | train 2.925950 | val 3.298572 | tok/s 2627600.12 | norm 0.3794 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26102 | train 2.930119 | val 3.298572 | tok/s 2629626.92 | norm 0.3425 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26103 | train 2.816190 | val 3.298572 | tok/s 2626966.60 | norm 0.3330 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26104 | train 2.850617 | val 3.298572 | tok/s 2629905.42 | norm 0.3820 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26105 | train 2.900938 | val 3.298572 | tok/s 2626861.40 | norm 0.3650 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26106 | train 2.905242 | val 3.298572 | tok/s 2624740.45 | norm 0.3365 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26107 | train 2.889509 | val 3.298572 | tok/s 2628272.96 | norm 0.3522 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26108 | train 2.853570 | val 3.298572 | tok/s 2630506.26 | norm 0.3806 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26109 | train 2.891436 | val 3.298572 | tok/s 2630770.92 | norm 0.3313 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26110 | train 2.960548 | val 3.298572 | tok/s 2612877.81 | norm 0.3272 | dt 200.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26111 | train 2.805229 | val 3.298572 | tok/s 2624035.46 | norm 0.3404 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26112 | train 2.846222 | val 3.298572 | tok/s 2630186.59 | norm 0.3487 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26113 | train 2.869248 | val 3.298572 | tok/s 2627069.07 | norm 0.3439 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26114 | train 3.104174 | val 3.298572 | tok/s 2631320.90 | norm 0.3964 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26115 | train 2.898678 | val 3.298572 | tok/s 2623987.31 | norm 0.3473 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26116 | train 2.944269 | val 3.298572 | tok/s 2624718.56 | norm 0.3621 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26117 | train 2.960799 | val 3.298572 | tok/s 2633287.46 | norm 0.3999 | dt 199.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26118 | train 2.960933 | val 3.298572 | tok/s 2628535.32 | norm 0.3746 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26119 | train 2.879287 | val 3.298572 | tok/s 2629906.31 | norm 0.3631 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26120 | train 2.884014 | val 3.298572 | tok/s 2620822.51 | norm 0.3649 | dt 200.05 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26121 | train 2.918480 | val 3.298572 | tok/s 2625590.38 | norm 0.3632 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26122 | train 2.901479 | val 3.298572 | tok/s 2628744.03 | norm 0.3298 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26123 | train 2.955365 | val 3.298572 | tok/s 2628486.10 | norm 0.3099 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26124 | train 2.883842 | val 3.298572 | tok/s 2628295.51 | norm 0.3658 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26125 | train 3.067239 | val 3.298572 | tok/s 2622655.12 | norm 0.3790 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26126 | train 3.103240 | val 3.298572 | tok/s 2627475.88 | norm 0.3850 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26127 | train 2.977150 | val 3.298572 | tok/s 2628616.37 | norm 0.3704 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26128 | train 2.968024 | val 3.298572 | tok/s 2631968.48 | norm 0.3588 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26129 | train 2.959823 | val 3.298572 | tok/s 2630637.42 | norm 0.3627 | dt 199.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26130 | train 2.881882 | val 3.298572 | tok/s 2625436.07 | norm 0.3628 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26131 | train 2.898345 | val 3.298572 | tok/s 2628056.24 | norm 0.3727 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26132 | train 3.065006 | val 3.298572 | tok/s 2631570.06 | norm 0.3787 | dt 199.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26133 | train 2.863549 | val 3.298572 | tok/s 2629543.64 | norm 0.3616 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26134 | train 3.042685 | val 3.298572 | tok/s 2627609.55 | norm 0.3810 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26135 | train 2.991245 | val 3.298572 | tok/s 2626076.82 | norm 0.3552 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26136 | train 2.881479 | val 3.298572 | tok/s 2626156.43 | norm 0.3581 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26137 | train 2.954100 | val 3.298572 | tok/s 2610985.98 | norm 0.3587 | dt 200.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26138 | train 2.900917 | val 3.298572 | tok/s 2625811.56 | norm 0.3601 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26139 | train 2.899406 | val 3.298572 | tok/s 2613506.86 | norm 0.3784 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26140 | train 2.887501 | val 3.298572 | tok/s 2625665.38 | norm 0.3751 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26141 | train 2.987645 | val 3.298572 | tok/s 2626320.41 | norm 0.3678 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26142 | train 2.880182 | val 3.298572 | tok/s 2610765.92 | norm 0.3769 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26143 | train 2.821229 | val 3.298572 | tok/s 2616872.13 | norm 0.3547 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26144 | train 2.854389 | val 3.298572 | tok/s 2625700.04 | norm 0.3534 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26145 | train 2.927611 | val 3.298572 | tok/s 2625967.75 | norm 0.3771 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26146 | train 2.909140 | val 3.298572 | tok/s 2613520.37 | norm 0.3635 | dt 200.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26147 | train 2.956249 | val 3.298572 | tok/s 2603983.18 | norm 0.3459 | dt 201.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26148 | train 2.943996 | val 3.298572 | tok/s 2618848.87 | norm 0.3918 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26149 | train 3.108416 | val 3.298572 | tok/s 2623520.66 | norm 0.4652 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26150 | train 2.836140 | val 3.298572 | tok/s 2624716.09 | norm 0.3429 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26151 | train 2.861389 | val 3.298572 | tok/s 2624391.23 | norm 0.3488 | dt 199.78 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26152 | train 3.028085 | val 3.298572 | tok/s 2617516.40 | norm 0.3431 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26153 | train 2.931828 | val 3.298572 | tok/s 2620486.04 | norm 0.3260 | dt 200.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26154 | train 2.994517 | val 3.298572 | tok/s 2596916.38 | norm 0.3448 | dt 201.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26155 | train 2.942338 | val 3.298572 | tok/s 2615444.66 | norm 0.3482 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26156 | train 2.959020 | val 3.298572 | tok/s 2621721.10 | norm 0.3549 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26157 | train 2.885044 | val 3.298572 | tok/s 2609175.45 | norm 0.4113 | dt 200.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26158 | train 2.830439 | val 3.298572 | tok/s 2616897.01 | norm 0.3475 | dt 200.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26159 | train 2.918906 | val 3.298572 | tok/s 2626368.50 | norm 0.3503 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26160 | train 3.003153 | val 3.298572 | tok/s 2620854.63 | norm 0.3805 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26161 | train 2.930128 | val 3.298572 | tok/s 2624800.53 | norm 0.3797 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26162 | train 3.087048 | val 3.298572 | tok/s 2613994.77 | norm 0.3810 | dt 200.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26163 | train 2.950792 | val 3.298572 | tok/s 2611602.16 | norm 0.3408 | dt 200.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26164 | train 2.990317 | val 3.298572 | tok/s 2626050.36 | norm 0.3311 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26165 | train 3.028972 | val 3.298572 | tok/s 2624594.97 | norm 0.3593 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26166 | train 3.039973 | val 3.298572 | tok/s 2624404.25 | norm 0.3450 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26167 | train 3.094016 | val 3.298572 | tok/s 2617020.15 | norm 0.3606 | dt 200.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26168 | train 2.831834 | val 3.298572 | tok/s 2612341.92 | norm 0.3730 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26169 | train 3.031413 | val 3.298572 | tok/s 2627692.73 | norm 0.3913 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26170 | train 3.004107 | val 3.298572 | tok/s 2623388.85 | norm 0.3797 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26171 | train 2.892121 | val 3.298572 | tok/s 2626197.96 | norm 0.3457 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26172 | train 2.962188 | val 3.298572 | tok/s 2621783.69 | norm 0.3661 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26173 | train 2.967131 | val 3.298572 | tok/s 2614754.40 | norm 0.3663 | dt 200.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26174 | train 2.868506 | val 3.298572 | tok/s 2620709.50 | norm 0.3614 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26175 | train 2.880972 | val 3.298572 | tok/s 2617147.32 | norm 0.3306 | dt 200.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26176 | train 3.104496 | val 3.298572 | tok/s 2627229.56 | norm 0.3544 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26177 | train 2.971212 | val 3.298572 | tok/s 2619695.83 | norm 0.3697 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26178 | train 3.081225 | val 3.298572 | tok/s 2618848.92 | norm 0.3465 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26179 | train 3.145921 | val 3.298572 | tok/s 2629239.98 | norm 0.5903 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26180 | train 3.106311 | val 3.298572 | tok/s 2619001.15 | norm 0.3986 | dt 200.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26181 | train 3.033746 | val 3.298572 | tok/s 2629677.03 | norm 0.3182 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26182 | train 2.908342 | val 3.298572 | tok/s 2622476.10 | norm 0.3262 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26183 | train 2.975748 | val 3.298572 | tok/s 2624477.10 | norm 0.3316 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26184 | train 2.882150 | val 3.298572 | tok/s 2628412.15 | norm 0.3333 | dt 199.47 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26185 | train 3.022735 | val 3.298572 | tok/s 2620029.51 | norm 0.4331 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26186 | train 3.042092 | val 3.298572 | tok/s 2619766.53 | norm 0.3706 | dt 200.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26187 | train 2.849681 | val 3.298572 | tok/s 2617503.37 | norm 0.3378 | dt 200.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26188 | train 2.902086 | val 3.298572 | tok/s 2619993.63 | norm 0.3077 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26189 | train 2.866162 | val 3.298572 | tok/s 2624882.40 | norm 0.3409 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26190 | train 2.900676 | val 3.298572 | tok/s 2621808.06 | norm 0.3469 | dt 199.97 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26191 | train 2.820715 | val 3.298572 | tok/s 2626452.43 | norm 0.3460 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26192 | train 2.922155 | val 3.298572 | tok/s 2628220.44 | norm 0.3660 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26193 | train 2.913974 | val 3.298572 | tok/s 2620173.57 | norm 0.3832 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26194 | train 2.917351 | val 3.298572 | tok/s 2627348.85 | norm 0.3359 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26195 | train 3.063803 | val 3.298572 | tok/s 2618800.93 | norm 0.3680 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26196 | train 2.923668 | val 3.298572 | tok/s 2615830.06 | norm 0.3785 | dt 200.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26197 | train 2.971752 | val 3.298572 | tok/s 2622163.07 | norm 0.3316 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26198 | train 2.918206 | val 3.298572 | tok/s 2626555.09 | norm 0.3088 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26199 | train 2.937546 | val 3.298572 | tok/s 2607963.05 | norm 0.3485 | dt 201.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26200 | train 2.955336 | val 3.298572 | tok/s 2616560.30 | norm 0.3780 | dt 200.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26201 | train 2.940886 | val 3.298572 | tok/s 2578076.93 | norm 0.3450 | dt 203.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26202 | train 3.006881 | val 3.298572 | tok/s 2615482.35 | norm 0.3700 | dt 200.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26203 | train 2.932080 | val 3.298572 | tok/s 2614647.55 | norm 0.3409 | dt 200.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26204 | train 2.857152 | val 3.298572 | tok/s 2605442.57 | norm 0.3532 | dt 201.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26205 | train 2.949924 | val 3.298572 | tok/s 2620660.71 | norm 0.3957 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26206 | train 2.988880 | val 3.298572 | tok/s 2627485.58 | norm 0.3755 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26207 | train 2.998787 | val 3.298572 | tok/s 2620348.61 | norm 0.3625 | dt 200.08 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26208 | train 2.817463 | val 3.298572 | tok/s 2618510.24 | norm 0.3286 | dt 200.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26209 | train 2.853519 | val 3.298572 | tok/s 2613388.16 | norm 0.3233 | dt 200.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26210 | train 2.902359 | val 3.298572 | tok/s 2619494.42 | norm 0.3299 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26211 | train 2.835433 | val 3.298572 | tok/s 2626011.47 | norm 0.3238 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26212 | train 3.034947 | val 3.298572 | tok/s 2625849.44 | norm 0.3283 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26213 | train 2.947711 | val 3.298572 | tok/s 2624902.26 | norm 0.3406 | dt 199.74 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26214 | train 2.853270 | val 3.298572 | tok/s 2612545.80 | norm 0.3045 | dt 200.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26215 | train 2.898896 | val 3.298572 | tok/s 2620024.15 | norm 0.3184 | dt 200.11 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26216 | train 2.800081 | val 3.298572 | tok/s 2625278.59 | norm 0.3661 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26217 | train 2.960909 | val 3.298572 | tok/s 2619537.02 | norm 0.3637 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26218 | train 2.916115 | val 3.298572 | tok/s 2626985.97 | norm 0.3449 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26219 | train 2.957782 | val 3.298572 | tok/s 2614565.59 | norm 0.3716 | dt 200.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26220 | train 2.925468 | val 3.298572 | tok/s 2616475.39 | norm 0.3544 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26221 | train 2.983815 | val 3.298572 | tok/s 2627904.25 | norm 0.3484 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26222 | train 2.976260 | val 3.298572 | tok/s 2620694.24 | norm 0.3505 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26223 | train 2.922908 | val 3.298572 | tok/s 2625210.80 | norm 0.3277 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26224 | train 2.982016 | val 3.298572 | tok/s 2618855.01 | norm 0.3472 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26225 | train 2.950694 | val 3.298572 | tok/s 2612316.28 | norm 0.4026 | dt 200.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26226 | train 3.129368 | val 3.298572 | tok/s 2627187.39 | norm 0.3397 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26227 | train 2.977922 | val 3.298572 | tok/s 2624688.79 | norm 0.3412 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26228 | train 2.896011 | val 3.298572 | tok/s 2619471.83 | norm 0.3480 | dt 200.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26229 | train 3.004511 | val 3.298572 | tok/s 2618119.06 | norm 0.3368 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26230 | train 2.949813 | val 3.298572 | tok/s 2611179.33 | norm 0.3665 | dt 200.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26231 | train 2.938442 | val 3.298572 | tok/s 2630172.42 | norm 0.3912 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26232 | train 2.885030 | val 3.298572 | tok/s 2627943.37 | norm 0.3816 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26233 | train 2.859813 | val 3.298572 | tok/s 2632506.53 | norm 0.3724 | dt 199.16 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26234 | train 3.015901 | val 3.298572 | tok/s 2625398.22 | norm 0.3931 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26235 | train 2.969552 | val 3.298572 | tok/s 2625057.42 | norm 0.4004 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26236 | train 2.889471 | val 3.298572 | tok/s 2581758.50 | norm 0.3801 | dt 203.07 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26237 | train 2.877251 | val 3.298572 | tok/s 2540511.98 | norm 0.3857 | dt 206.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26238 | train 2.827703 | val 3.298572 | tok/s 2631400.06 | norm 0.3333 | dt 199.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26239 | train 3.133033 | val 3.298572 | tok/s 2626576.18 | norm 0.3818 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26240 | train 2.845904 | val 3.298572 | tok/s 2626353.83 | norm 0.3665 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26241 | train 2.884320 | val 3.298572 | tok/s 2632160.17 | norm 0.3347 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26242 | train 2.948515 | val 3.298572 | tok/s 2627047.83 | norm 0.3924 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26243 | train 2.843032 | val 3.298572 | tok/s 2488260.01 | norm 0.3482 | dt 210.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26244 | train 3.100431 | val 3.298572 | tok/s 2616460.41 | norm 0.3252 | dt 200.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26245 | train 3.032155 | val 3.298572 | tok/s 2504108.92 | norm 0.3973 | dt 209.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26246 | train 2.950896 | val 3.298572 | tok/s 2625444.37 | norm 0.4131 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26247 | train 2.970830 | val 3.298572 | tok/s 2621706.76 | norm 0.3275 | dt 199.98 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26248 | train 2.960979 | val 3.298572 | tok/s 2630195.51 | norm 0.3178 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26249 | train 2.897605 | val 3.298572 | tok/s 2621906.16 | norm 0.3639 | dt 199.96 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB [val] step 26250 | val_loss 3.300345 step 26250 | train 2.915826 | val 3.300345 | tok/s 2607139.49 | norm 0.3692 | dt 201.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26251 | train 2.928665 | val 3.300345 | tok/s 2627382.22 | norm 0.3623 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26252 | train 2.971573 | val 3.300345 | tok/s 2632406.01 | norm 0.3780 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26253 | train 2.861542 | val 3.300345 | tok/s 2627617.29 | norm 0.3625 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26254 | train 3.000266 | val 3.300345 | tok/s 2626668.99 | norm 0.3667 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26255 | train 2.980004 | val 3.300345 | tok/s 2627560.92 | norm 0.3805 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26256 | train 2.935264 | val 3.300345 | tok/s 2627628.74 | norm 0.3887 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26257 | train 2.987866 | val 3.300345 | tok/s 2627160.67 | norm 0.4009 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26258 | train 3.054079 | val 3.300345 | tok/s 2606868.85 | norm 0.3824 | dt 201.12 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26259 | train 2.905548 | val 3.300345 | tok/s 2635503.51 | norm 0.3907 | dt 198.93 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26260 | train 2.854235 | val 3.300345 | tok/s 2630502.73 | norm 0.3491 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26261 | train 2.858173 | val 3.300345 | tok/s 2627988.08 | norm 0.3559 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26262 | train 2.948962 | val 3.300345 | tok/s 2626108.43 | norm 0.3700 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26263 | train 3.016347 | val 3.300345 | tok/s 2623967.98 | norm 0.3469 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26264 | train 2.964358 | val 3.300345 | tok/s 2635473.20 | norm 0.3374 | dt 198.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26265 | train 3.014859 | val 3.300345 | tok/s 2625260.69 | norm 0.3422 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26266 | train 2.891692 | val 3.300345 | tok/s 2629355.42 | norm 0.3704 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26267 | train 2.868049 | val 3.300345 | tok/s 2627450.57 | norm 0.3584 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26268 | train 3.064960 | val 3.300345 | tok/s 2624134.11 | norm 0.4050 | dt 199.79 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26269 | train 2.856523 | val 3.300345 | tok/s 2598673.66 | norm 0.4037 | dt 201.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26270 | train 3.014596 | val 3.300345 | tok/s 2626029.56 | norm 0.3397 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26271 | train 2.882322 | val 3.300345 | tok/s 2630234.41 | norm 0.3868 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26272 | train 2.903290 | val 3.300345 | tok/s 2623404.67 | norm 0.3610 | dt 199.85 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26273 | train 2.930773 | val 3.300345 | tok/s 2625630.36 | norm 0.3504 | dt 199.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26274 | train 2.880060 | val 3.300345 | tok/s 2630434.60 | norm 0.3607 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26275 | train 3.024724 | val 3.300345 | tok/s 2627524.26 | norm 0.3495 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26276 | train 2.881091 | val 3.300345 | tok/s 2628306.52 | norm 0.3503 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26277 | train 3.010932 | val 3.300345 | tok/s 2628245.58 | norm 0.3815 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26278 | train 3.006689 | val 3.300345 | tok/s 2624464.01 | norm 0.3947 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26279 | train 3.047299 | val 3.300345 | tok/s 2632190.51 | norm 0.3551 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26280 | train 2.952358 | val 3.300345 | tok/s 2625461.89 | norm 0.3733 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26281 | train 2.911298 | val 3.300345 | tok/s 2630232.47 | norm 0.3809 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26282 | train 3.051959 | val 3.300345 | tok/s 2628164.11 | norm 0.3496 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26283 | train 3.146214 | val 3.300345 | tok/s 2625114.03 | norm 0.4895 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26284 | train 2.991828 | val 3.300345 | tok/s 2630288.94 | norm 0.3863 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26285 | train 2.951226 | val 3.300345 | tok/s 2628895.35 | norm 0.3614 | dt 199.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26286 | train 2.974159 | val 3.300345 | tok/s 2629129.88 | norm 0.3611 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26287 | train 2.983336 | val 3.300345 | tok/s 2621162.22 | norm 0.3644 | dt 200.02 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26288 | train 2.955959 | val 3.300345 | tok/s 2622783.57 | norm 0.3517 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26289 | train 2.923818 | val 3.300345 | tok/s 2628806.07 | norm 0.3356 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26290 | train 2.915176 | val 3.300345 | tok/s 2629707.34 | norm 0.3363 | dt 199.37 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26291 | train 2.855265 | val 3.300345 | tok/s 2631277.02 | norm 0.3674 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26292 | train 2.904817 | val 3.300345 | tok/s 2625136.86 | norm 0.3901 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26293 | train 2.967665 | val 3.300345 | tok/s 2622726.28 | norm 0.3594 | dt 199.90 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26294 | train 3.020452 | val 3.300345 | tok/s 2625241.10 | norm 0.3951 | dt 199.71 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26295 | train 2.956624 | val 3.300345 | tok/s 2621439.24 | norm 0.3716 | dt 200.00 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26296 | train 2.912600 | val 3.300345 | tok/s 2628638.21 | norm 0.3279 | dt 199.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26297 | train 2.833175 | val 3.300345 | tok/s 2589764.15 | norm 0.4006 | dt 202.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26298 | train 2.876655 | val 3.300345 | tok/s 2629999.39 | norm 0.3831 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26299 | train 2.884616 | val 3.300345 | tok/s 2618347.74 | norm 0.3678 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26300 | train 3.285012 | val 3.300345 | tok/s 2620954.46 | norm 0.4947 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26301 | train 2.904912 | val 3.300345 | tok/s 2630504.79 | norm 0.4124 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26302 | train 2.976349 | val 3.300345 | tok/s 2623046.34 | norm 0.4171 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26303 | train 2.765175 | val 3.300345 | tok/s 2623972.47 | norm 0.4168 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26304 | train 2.904145 | val 3.300345 | tok/s 2630402.73 | norm 0.3220 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26305 | train 2.946403 | val 3.300345 | tok/s 2623602.91 | norm 0.3376 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26306 | train 2.954207 | val 3.300345 | tok/s 2632149.07 | norm 0.3669 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26307 | train 2.910725 | val 3.300345 | tok/s 2623476.71 | norm 0.3525 | dt 199.84 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26308 | train 3.025771 | val 3.300345 | tok/s 2630238.28 | norm 0.3679 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26309 | train 2.892664 | val 3.300345 | tok/s 2587421.78 | norm 0.3593 | dt 202.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26310 | train 2.922740 | val 3.300345 | tok/s 2625362.65 | norm 0.3624 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26311 | train 2.984455 | val 3.300345 | tok/s 2591231.78 | norm 0.3548 | dt 202.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26312 | train 2.807882 | val 3.300345 | tok/s 2629646.55 | norm 0.3534 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26313 | train 2.932109 | val 3.300345 | tok/s 2589661.63 | norm 0.3424 | dt 202.45 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26314 | train 2.924110 | val 3.300345 | tok/s 2624099.72 | norm 0.3761 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26315 | train 2.988906 | val 3.300345 | tok/s 2603238.71 | norm 0.3737 | dt 201.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26316 | train 2.907622 | val 3.300345 | tok/s 2632170.33 | norm 0.3604 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26317 | train 2.983471 | val 3.300345 | tok/s 2628272.85 | norm 0.4150 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26318 | train 3.094514 | val 3.300345 | tok/s 2628196.99 | norm 0.6684 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26319 | train 2.926283 | val 3.300345 | tok/s 2623990.89 | norm 0.6569 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26320 | train 3.027696 | val 3.300345 | tok/s 2629338.47 | norm 0.3642 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26321 | train 2.879726 | val 3.300345 | tok/s 2629919.87 | norm 0.3190 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26322 | train 3.000148 | val 3.300345 | tok/s 2628539.08 | norm 0.3414 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26323 | train 2.919178 | val 3.300345 | tok/s 2632641.64 | norm 0.3751 | dt 199.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26324 | train 2.899165 | val 3.300345 | tok/s 2627395.32 | norm 0.3581 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26325 | train 2.970070 | val 3.300345 | tok/s 2478969.59 | norm 0.3395 | dt 211.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26326 | train 2.962114 | val 3.300345 | tok/s 2632376.46 | norm 0.4180 | dt 199.17 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26327 | train 3.141544 | val 3.300345 | tok/s 2483238.80 | norm 0.3636 | dt 211.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26328 | train 2.966882 | val 3.300345 | tok/s 2589921.39 | norm 0.3112 | dt 202.43 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26329 | train 2.863236 | val 3.300345 | tok/s 2626050.83 | norm 0.3779 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26330 | train 2.905998 | val 3.300345 | tok/s 2626545.37 | norm 0.3573 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26331 | train 2.951828 | val 3.300345 | tok/s 2626364.50 | norm 0.3532 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26332 | train 2.976902 | val 3.300345 | tok/s 2622661.67 | norm 0.3701 | dt 199.91 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26333 | train 2.968049 | val 3.300345 | tok/s 2629890.94 | norm 0.3546 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26334 | train 2.912396 | val 3.300345 | tok/s 2624467.99 | norm 0.3443 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26335 | train 2.964984 | val 3.300345 | tok/s 2629486.28 | norm 0.3365 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26336 | train 2.815504 | val 3.300345 | tok/s 2620191.53 | norm 0.3544 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26337 | train 3.005924 | val 3.300345 | tok/s 2623188.91 | norm 0.3717 | dt 199.87 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26338 | train 3.082062 | val 3.300345 | tok/s 2630362.60 | norm 0.3927 | dt 199.32 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26339 | train 2.924645 | val 3.300345 | tok/s 2627526.57 | norm 0.3879 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26340 | train 3.090836 | val 3.300345 | tok/s 2626304.94 | norm 0.4088 | dt 199.63 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26341 | train 2.978498 | val 3.300345 | tok/s 2625969.61 | norm 0.3732 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26342 | train 2.975852 | val 3.300345 | tok/s 2626204.01 | norm 0.3627 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26343 | train 2.893242 | val 3.300345 | tok/s 2632053.64 | norm 0.3176 | dt 199.19 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26344 | train 2.908743 | val 3.300345 | tok/s 2622844.49 | norm 0.3312 | dt 199.89 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26345 | train 2.931884 | val 3.300345 | tok/s 2627630.79 | norm 0.3447 | dt 199.53 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26346 | train 2.936683 | val 3.300345 | tok/s 2626127.70 | norm 0.3689 | dt 199.64 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26347 | train 2.946671 | val 3.300345 | tok/s 2618138.81 | norm 0.3564 | dt 200.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26348 | train 3.033357 | val 3.300345 | tok/s 2632858.34 | norm 0.3225 | dt 199.13 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26349 | train 2.910511 | val 3.300345 | tok/s 2630266.81 | norm 0.3396 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26350 | train 3.029018 | val 3.300345 | tok/s 2629518.99 | norm 0.3553 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26351 | train 2.912996 | val 3.300345 | tok/s 2574264.97 | norm 0.3601 | dt 203.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26352 | train 2.942747 | val 3.300345 | tok/s 2561355.53 | norm 0.3707 | dt 204.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26353 | train 2.897936 | val 3.300345 | tok/s 2629564.72 | norm 0.3467 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26354 | train 2.859502 | val 3.300345 | tok/s 2626837.22 | norm 0.3776 | dt 199.59 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26355 | train 2.961873 | val 3.300345 | tok/s 2622442.28 | norm 0.5630 | dt 199.92 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26356 | train 2.874167 | val 3.300345 | tok/s 2622230.72 | norm 0.3621 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26357 | train 2.915287 | val 3.300345 | tok/s 2607843.03 | norm 0.3244 | dt 201.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26358 | train 2.976735 | val 3.300345 | tok/s 2492993.60 | norm 0.3538 | dt 210.30 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26359 | train 3.021071 | val 3.300345 | tok/s 2620167.55 | norm 0.3476 | dt 200.10 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26360 | train 3.024831 | val 3.300345 | tok/s 2555593.65 | norm 0.3508 | dt 205.15 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26361 | train 2.821500 | val 3.300345 | tok/s 2624660.54 | norm 0.3960 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26362 | train 2.985300 | val 3.300345 | tok/s 2629595.92 | norm 0.4069 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26363 | train 2.948123 | val 3.300345 | tok/s 2620715.48 | norm 0.3606 | dt 200.06 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26364 | train 3.075053 | val 3.300345 | tok/s 2625022.89 | norm 0.3643 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26365 | train 2.983284 | val 3.300345 | tok/s 2629815.84 | norm 0.3815 | dt 199.36 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26366 | train 2.917856 | val 3.300345 | tok/s 2624114.58 | norm 0.3920 | dt 199.80 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26367 | train 2.889607 | val 3.300345 | tok/s 2628008.60 | norm 0.3493 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26368 | train 2.954830 | val 3.300345 | tok/s 2610782.82 | norm 0.3692 | dt 200.82 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26369 | train 2.814285 | val 3.300345 | tok/s 2625325.94 | norm 0.3578 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26370 | train 2.885823 | val 3.300345 | tok/s 2630769.43 | norm 0.3706 | dt 199.29 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26371 | train 3.010736 | val 3.300345 | tok/s 2628741.09 | norm 0.3655 | dt 199.44 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26372 | train 2.927118 | val 3.300345 | tok/s 2631947.93 | norm 0.3610 | dt 199.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26373 | train 2.915404 | val 3.300345 | tok/s 2623993.29 | norm 0.3605 | dt 199.81 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26374 | train 2.971311 | val 3.300345 | tok/s 2625972.76 | norm 0.3575 | dt 199.65 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26375 | train 2.883748 | val 3.300345 | tok/s 2629296.56 | norm 0.3543 | dt 199.40 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26376 | train 2.798152 | val 3.300345 | tok/s 2627876.75 | norm 0.3466 | dt 199.51 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26377 | train 2.918071 | val 3.300345 | tok/s 2627280.13 | norm 0.3314 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26378 | train 2.922889 | val 3.300345 | tok/s 2630144.11 | norm 0.3726 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26379 | train 2.923911 | val 3.300345 | tok/s 2627296.74 | norm 0.3488 | dt 199.55 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26380 | train 2.938695 | val 3.300345 | tok/s 2628165.05 | norm 0.3917 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26381 | train 2.942877 | val 3.300345 | tok/s 2629166.04 | norm 0.3645 | dt 199.41 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26382 | train 2.918149 | val 3.300345 | tok/s 2626945.45 | norm 0.3396 | dt 199.58 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26383 | train 2.897753 | val 3.300345 | tok/s 2622951.62 | norm 0.3125 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26384 | train 2.950020 | val 3.300345 | tok/s 2622263.88 | norm 0.3468 | dt 199.94 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26385 | train 2.933324 | val 3.300345 | tok/s 2629417.23 | norm 0.3554 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26386 | train 2.855524 | val 3.300345 | tok/s 2625166.84 | norm 0.3473 | dt 199.72 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26387 | train 2.915996 | val 3.300345 | tok/s 2630052.67 | norm 0.3363 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26388 | train 2.840862 | val 3.300345 | tok/s 2623008.10 | norm 0.3299 | dt 199.88 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26389 | train 2.925831 | val 3.300345 | tok/s 2627258.81 | norm 0.3757 | dt 199.56 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26390 | train 2.972576 | val 3.300345 | tok/s 2629657.34 | norm 0.3981 | dt 199.38 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26391 | train 2.862845 | val 3.300345 | tok/s 2628006.48 | norm 0.3510 | dt 199.50 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26392 | train 2.821592 | val 3.300345 | tok/s 2631106.75 | norm 0.3321 | dt 199.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26393 | train 3.137654 | val 3.300345 | tok/s 2630020.65 | norm 0.3761 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26394 | train 2.917023 | val 3.300345 | tok/s 2625398.84 | norm 0.4125 | dt 199.70 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26395 | train 3.054026 | val 3.300345 | tok/s 2630550.24 | norm 0.3419 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26396 | train 2.892616 | val 3.300345 | tok/s 2626690.78 | norm 0.3470 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26397 | train 3.007322 | val 3.300345 | tok/s 2626484.93 | norm 0.3444 | dt 199.62 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26398 | train 3.031266 | val 3.300345 | tok/s 2419657.45 | norm 0.3575 | dt 216.68 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26399 | train 2.886126 | val 3.300345 | tok/s 2595097.63 | norm 0.3497 | dt 202.03 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26400 | train 2.957910 | val 3.300345 | tok/s 2631766.63 | norm 0.3689 | dt 199.22 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26401 | train 2.937851 | val 3.300345 | tok/s 2628227.61 | norm 0.3832 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26402 | train 2.981824 | val 3.300345 | tok/s 2630562.16 | norm 0.3585 | dt 199.31 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26403 | train 2.890418 | val 3.300345 | tok/s 2625035.37 | norm 0.3824 | dt 199.73 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26404 | train 2.859155 | val 3.300345 | tok/s 2627529.58 | norm 0.3618 | dt 199.54 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26405 | train 2.902275 | val 3.300345 | tok/s 2624541.85 | norm 0.3742 | dt 199.76 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26406 | train 2.962420 | val 3.300345 | tok/s 2619145.80 | norm 0.3878 | dt 200.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26407 | train 2.828518 | val 3.300345 | tok/s 2566579.49 | norm 0.3606 | dt 204.27 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26408 | train 2.953460 | val 3.300345 | tok/s 2625889.69 | norm 0.3699 | dt 199.66 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26409 | train 2.998853 | val 3.300345 | tok/s 2628288.28 | norm 0.3409 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26410 | train 2.820771 | val 3.300345 | tok/s 2618385.69 | norm 0.3031 | dt 200.23 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26411 | train 2.888511 | val 3.300345 | tok/s 2621264.91 | norm 0.3060 | dt 200.01 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26412 | train 2.918908 | val 3.300345 | tok/s 2630313.65 | norm 0.3529 | dt 199.33 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26413 | train 2.874049 | val 3.300345 | tok/s 2628203.13 | norm 0.3703 | dt 199.49 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26414 | train 2.970923 | val 3.300345 | tok/s 2626734.09 | norm 0.3485 | dt 199.60 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26415 | train 2.960070 | val 3.300345 | tok/s 2624435.62 | norm 0.3573 | dt 199.77 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26416 | train 2.963759 | val 3.300345 | tok/s 2625566.43 | norm 0.3538 | dt 199.69 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26417 | train 2.964593 | val 3.300345 | tok/s 2632218.76 | norm 0.3580 | dt 199.18 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26418 | train 2.856117 | val 3.300345 | tok/s 2626525.15 | norm 0.3691 | dt 199.61 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26419 | train 3.117469 | val 3.300345 | tok/s 2628281.15 | norm 0.3703 | dt 199.48 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26420 | train 2.866149 | val 3.300345 | tok/s 2618856.07 | norm 0.3525 | dt 200.20 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26421 | train 2.895865 | val 3.300345 | tok/s 2623312.36 | norm 0.3956 | dt 199.86 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26422 | train 2.917024 | val 3.300345 | tok/s 2631366.99 | norm 0.3595 | dt 199.25 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26423 | train 3.020469 | val 3.300345 | tok/s 2629075.26 | norm 0.4032 | dt 199.42 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26424 | train 2.949997 | val 3.300345 | tok/s 2629995.82 | norm 0.3762 | dt 199.35 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26425 | train 3.007831 | val 3.300345 | tok/s 2618723.81 | norm 0.3618 | dt 200.21 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26426 | train 3.002506 | val 3.300345 | tok/s 2618314.61 | norm 0.3782 | dt 200.24 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26427 | train 2.792349 | val 3.300345 | tok/s 2627048.35 | norm 0.3716 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26428 | train 2.928289 | val 3.300345 | tok/s 2627120.43 | norm 0.4044 | dt 199.57 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26429 | train 2.881983 | val 3.300345 | tok/s 2630130.20 | norm 0.3937 | dt 199.34 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26430 | train 2.895210 | val 3.300345 | tok/s 2620914.10 | norm 0.3867 | dt 200.04 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26431 | train 3.016612 | val 3.300345 | tok/s 2625774.14 | norm 0.4030 | dt 199.67 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26432 | train 2.941926 | val 3.300345 | tok/s 2628541.90 | norm 0.3839 | dt 199.46 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26433 | train 2.942167 | val 3.300345 | tok/s 2627726.61 | norm 0.5569 | dt 199.52 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26434 | train 2.873488 | val 3.300345 | tok/s 2629514.96 | norm 0.3484 | dt 199.39 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26435 | train 2.944437 | val 3.300345 | tok/s 2624682.10 | norm 0.3600 | dt 199.75 ms | lr 1.0000e-03 | mlr 1.0000e-03 | mem 13.73/16.12 GiB step 26436 | train 2.982701 | val 3.300345 | tok/s 2626090.48 | norm 0.3567 | dt 199.65 ms | lr 9.9997e-04 | mlr 9.9997e-04 | mem 13.73/16.12 GiB step 26437 | train 2.929340 | val 3.300345 | tok/s 2630286.01 | norm 0.3562 | dt 199.33 ms | lr 9.9988e-04 | mlr 9.9988e-04 | mem 13.73/16.12 GiB step 26438 | train 2.946178 | val 3.300345 | tok/s 2627764.86 | norm 0.3374 | dt 199.52 ms | lr 9.9972e-04 | mlr 9.9972e-04 | mem 13.73/16.12 GiB step 26439 | train 2.852630 | val 3.300345 | tok/s 2628000.79 | norm 0.3565 | dt 199.50 ms | lr 9.9950e-04 | mlr 9.9950e-04 | mem 13.73/16.12 GiB step 26440 | train 2.900133 | val 3.300345 | tok/s 2621747.16 | norm 0.3380 | dt 199.98 ms | lr 9.9922e-04 | mlr 9.9922e-04 | mem 13.73/16.12 GiB step 26441 | train 2.967072 | val 3.300345 | tok/s 2626684.60 | norm 0.3239 | dt 199.60 ms | lr 9.9888e-04 | mlr 9.9888e-04 | mem 13.73/16.12 GiB step 26442 | train 2.933041 | val 3.300345 | tok/s 2632183.72 | norm 0.3622 | dt 199.18 ms | lr 9.9847e-04 | mlr 9.9847e-04 | mem 13.73/16.12 GiB step 26443 | train 2.987133 | val 3.300345 | tok/s 2628353.59 | norm 0.3436 | dt 199.47 ms | lr 9.9801e-04 | mlr 9.9801e-04 | mem 13.73/16.12 GiB step 26444 | train 2.997205 | val 3.300345 | tok/s 2630706.72 | norm 0.3513 | dt 199.30 ms | lr 9.9748e-04 | mlr 9.9748e-04 | mem 13.73/16.12 GiB step 26445 | train 2.803062 | val 3.300345 | tok/s 2626793.88 | norm 0.4073 | dt 199.59 ms | lr 9.9689e-04 | mlr 9.9689e-04 | mem 13.73/16.12 GiB step 26446 | train 3.023186 | val 3.300345 | tok/s 2626952.04 | norm 0.3730 | dt 199.58 ms | lr 9.9624e-04 | mlr 9.9624e-04 | mem 13.73/16.12 GiB step 26447 | train 2.943350 | val 3.300345 | tok/s 2621560.92 | norm 0.3548 | dt 199.99 ms | lr 9.9552e-04 | mlr 9.9552e-04 | mem 13.73/16.12 GiB step 26448 | train 2.862069 | val 3.300345 | tok/s 2622632.14 | norm 0.3776 | dt 199.91 ms | lr 9.9475e-04 | mlr 9.9475e-04 | mem 13.73/16.12 GiB step 26449 | train 2.864929 | val 3.300345 | tok/s 2628757.63 | norm 0.3210 | dt 199.44 ms | lr 9.9391e-04 | mlr 9.9391e-04 | mem 13.73/16.12 GiB step 26450 | train 2.858018 | val 3.300345 | tok/s 2619792.19 | norm 0.3322 | dt 200.13 ms | lr 9.9301e-04 | mlr 9.9301e-04 | mem 13.73/16.12 GiB step 26451 | train 2.941040 | val 3.300345 | tok/s 2499444.63 | norm 0.3696 | dt 209.76 ms | lr 9.9205e-04 | mlr 9.9205e-04 | mem 13.73/16.12 GiB step 26452 | train 2.967257 | val 3.300345 | tok/s 2611175.81 | norm 0.3453 | dt 200.79 ms | lr 9.9103e-04 | mlr 9.9103e-04 | mem 13.73/16.12 GiB step 26453 | train 2.854445 | val 3.300345 | tok/s 2616510.79 | norm 0.3422 | dt 200.38 ms | lr 9.8995e-04 | mlr 9.8995e-04 | mem 13.73/16.12 GiB step 26454 | train 2.840138 | val 3.300345 | tok/s 2626019.50 | norm 0.3656 | dt 199.65 ms | lr 9.8880e-04 | mlr 9.8880e-04 | mem 13.73/16.12 GiB step 26455 | train 2.990350 | val 3.300345 | tok/s 2624496.25 | norm 0.3671 | dt 199.77 ms | lr 9.8760e-04 | mlr 9.8760e-04 | mem 13.73/16.12 GiB step 26456 | train 2.914780 | val 3.300345 | tok/s 2623272.95 | norm 0.3761 | dt 199.86 ms | lr 9.8633e-04 | mlr 9.8633e-04 | mem 13.73/16.12 GiB step 26457 | train 2.954236 | val 3.300345 | tok/s 2620883.70 | norm 0.3714 | dt 200.04 ms | lr 9.8501e-04 | mlr 9.8501e-04 | mem 13.73/16.12 GiB step 26458 | train 2.900583 | val 3.300345 | tok/s 2587987.56 | norm 0.3585 | dt 202.59 ms | lr 9.8362e-04 | mlr 9.8362e-04 | mem 13.73/16.12 GiB step 26459 | train 2.875985 | val 3.300345 | tok/s 2633085.25 | norm 0.3438 | dt 199.12 ms | lr 9.8218e-04 | mlr 9.8218e-04 | mem 13.73/16.12 GiB step 26460 | train 2.909560 | val 3.300345 | tok/s 2631272.43 | norm 0.3677 | dt 199.25 ms | lr 9.8067e-04 | mlr 9.8067e-04 | mem 13.73/16.12 GiB step 26461 | train 2.849681 | val 3.300345 | tok/s 2629764.74 | norm 0.3775 | dt 199.37 ms | lr 9.7911e-04 | mlr 9.7911e-04 | mem 13.73/16.12 GiB step 26462 | train 2.858026 | val 3.300345 | tok/s 2624226.29 | norm 0.3583 | dt 199.79 ms | lr 9.7748e-04 | mlr 9.7748e-04 | mem 13.73/16.12 GiB step 26463 | train 2.974437 | val 3.300345 | tok/s 2627050.54 | norm 0.3632 | dt 199.57 ms | lr 9.7580e-04 | mlr 9.7580e-04 | mem 13.73/16.12 GiB step 26464 | train 3.041285 | val 3.300345 | tok/s 2632223.81 | norm 0.3644 | dt 199.18 ms | lr 9.7406e-04 | mlr 9.7406e-04 | mem 13.73/16.12 GiB step 26465 | train 2.855299 | val 3.300345 | tok/s 2627560.42 | norm 0.3395 | dt 199.53 ms | lr 9.7225e-04 | mlr 9.7225e-04 | mem 13.73/16.12 GiB step 26466 | train 3.103740 | val 3.300345 | tok/s 2629188.36 | norm 0.3468 | dt 199.41 ms | lr 9.7040e-04 | mlr 9.7040e-04 | mem 13.73/16.12 GiB step 26467 | train 2.916525 | val 3.300345 | tok/s 2621801.02 | norm 0.3167 | dt 199.97 ms | lr 9.6848e-04 | mlr 9.6848e-04 | mem 13.73/16.12 GiB step 26468 | train 2.983520 | val 3.300345 | tok/s 2625788.95 | norm 0.3471 | dt 199.67 ms | lr 9.6650e-04 | mlr 9.6650e-04 | mem 13.73/16.12 GiB step 26469 | train 2.937799 | val 3.300345 | tok/s 2627829.41 | norm 0.3599 | dt 199.51 ms | lr 9.6447e-04 | mlr 9.6447e-04 | mem 13.73/16.12 GiB step 26470 | train 2.882789 | val 3.300345 | tok/s 2625048.94 | norm 0.3371 | dt 199.73 ms | lr 9.6238e-04 | mlr 9.6238e-04 | mem 13.73/16.12 GiB step 26471 | train 2.878792 | val 3.300345 | tok/s 2629124.23 | norm 0.3566 | dt 199.42 ms | lr 9.6023e-04 | mlr 9.6023e-04 | mem 13.73/16.12 GiB step 26472 | train 2.809965 | val 3.300345 | tok/s 2623153.65 | norm 0.3482 | dt 199.87 ms | lr 9.5802e-04 | mlr 9.5802e-04 | mem 13.73/16.12 GiB step 26473 | train 2.910331 | val 3.300345 | tok/s 2622992.90 | norm 0.3379 | dt 199.88 ms | lr 9.5576e-04 | mlr 9.5576e-04 | mem 13.73/16.12 GiB step 26474 | train 2.926641 | val 3.300345 | tok/s 2631567.36 | norm 0.3360 | dt 199.23 ms | lr 9.5345e-04 | mlr 9.5345e-04 | mem 13.73/16.12 GiB step 26475 | train 2.763137 | val 3.300345 | tok/s 2627242.11 | norm 0.3599 | dt 199.56 ms | lr 9.5107e-04 | mlr 9.5107e-04 | mem 13.73/16.12 GiB step 26476 | train 2.744550 | val 3.300345 | tok/s 2625887.78 | norm 0.3534 | dt 199.66 ms | lr 9.4864e-04 | mlr 9.4864e-04 | mem 13.73/16.12 GiB step 26477 | train 2.840741 | val 3.300345 | tok/s 2622628.60 | norm 0.3327 | dt 199.91 ms | lr 9.4616e-04 | mlr 9.4616e-04 | mem 13.73/16.12 GiB step 26478 | train 2.848233 | val 3.300345 | tok/s 2623346.20 | norm 0.3367 | dt 199.85 ms | lr 9.4362e-04 | mlr 9.4362e-04 | mem 13.73/16.12 GiB step 26479 | train 2.941063 | val 3.300345 | tok/s 2627681.97 | norm 0.3409 | dt 199.52 ms | lr 9.4103e-04 | mlr 9.4103e-04 | mem 13.73/16.12 GiB step 26480 | train 2.864502 | val 3.300345 | tok/s 2626040.14 | norm 0.3303 | dt 199.65 ms | lr 9.3838e-04 | mlr 9.3838e-04 | mem 13.73/16.12 GiB step 26481 | train 2.817165 | val 3.300345 | tok/s 2626486.90 | norm 0.3568 | dt 199.62 ms | lr 9.3568e-04 | mlr 9.3568e-04 | mem 13.73/16.12 GiB step 26482 | train 2.784449 | val 3.300345 | tok/s 2623678.04 | norm 0.3465 | dt 199.83 ms | lr 9.3293e-04 | mlr 9.3293e-04 | mem 13.73/16.12 GiB step 26483 | train 3.200454 | val 3.300345 | tok/s 2623236.84 | norm 0.4390 | dt 199.86 ms | lr 9.3012e-04 | mlr 9.3012e-04 | mem 13.73/16.12 GiB step 26484 | train 2.869236 | val 3.300345 | tok/s 2632140.69 | norm 0.3134 | dt 199.19 ms | lr 9.2726e-04 | mlr 9.2726e-04 | mem 13.73/16.12 GiB step 26485 | train 3.085630 | val 3.300345 | tok/s 2625373.68 | norm 0.3182 | dt 199.70 ms | lr 9.2435e-04 | mlr 9.2435e-04 | mem 13.73/16.12 GiB step 26486 | train 3.067570 | val 3.300345 | tok/s 2627755.26 | norm 0.3068 | dt 199.52 ms | lr 9.2138e-04 | mlr 9.2138e-04 | mem 13.73/16.12 GiB step 26487 | train 3.044795 | val 3.300345 | tok/s 2623776.08 | norm 0.3475 | dt 199.82 ms | lr 9.1837e-04 | mlr 9.1837e-04 | mem 13.73/16.12 GiB step 26488 | train 2.879246 | val 3.300345 | tok/s 2627939.45 | norm 0.3482 | dt 199.51 ms | lr 9.1530e-04 | mlr 9.1530e-04 | mem 13.73/16.12 GiB step 26489 | train 2.874245 | val 3.300345 | tok/s 2627718.60 | norm 0.3365 | dt 199.52 ms | lr 9.1218e-04 | mlr 9.1218e-04 | mem 13.73/16.12 GiB step 26490 | train 3.045520 | val 3.300345 | tok/s 2626972.44 | norm 0.4315 | dt 199.58 ms | lr 9.0901e-04 | mlr 9.0901e-04 | mem 13.73/16.12 GiB step 26491 | train 2.934344 | val 3.300345 | tok/s 2629297.85 | norm 0.3333 | dt 199.40 ms | lr 9.0580e-04 | mlr 9.0580e-04 | mem 13.73/16.12 GiB step 26492 | train 2.857272 | val 3.300345 | tok/s 2627441.43 | norm 0.3046 | dt 199.54 ms | lr 9.0253e-04 | mlr 9.0253e-04 | mem 13.73/16.12 GiB step 26493 | train 2.813360 | val 3.300345 | tok/s 2610619.80 | norm 0.3210 | dt 200.83 ms | lr 8.9922e-04 | mlr 8.9922e-04 | mem 13.73/16.12 GiB step 26494 | train 2.903278 | val 3.300345 | tok/s 2631453.07 | norm 0.3148 | dt 199.24 ms | lr 8.9585e-04 | mlr 8.9585e-04 | mem 13.73/16.12 GiB step 26495 | train 2.914973 | val 3.300345 | tok/s 2621053.68 | norm 0.3027 | dt 200.03 ms | lr 8.9244e-04 | mlr 8.9244e-04 | mem 13.73/16.12 GiB step 26496 | train 2.861085 | val 3.300345 | tok/s 2631501.26 | norm 0.3143 | dt 199.24 ms | lr 8.8898e-04 | mlr 8.8898e-04 | mem 13.73/16.12 GiB step 26497 | train 2.871762 | val 3.300345 | tok/s 2625895.33 | norm 0.3127 | dt 199.66 ms | lr 8.8548e-04 | mlr 8.8548e-04 | mem 13.73/16.12 GiB step 26498 | train 3.038677 | val 3.300345 | tok/s 2626327.38 | norm 0.3477 | dt 199.63 ms | lr 8.8192e-04 | mlr 8.8192e-04 | mem 13.73/16.12 GiB step 26499 | train 3.043288 | val 3.300345 | tok/s 2629323.17 | norm 0.3046 | dt 199.40 ms | lr 8.7833e-04 | mlr 8.7833e-04 | mem 13.73/16.12 GiB [val] step 26500 | val_loss 3.289458 step 26500 | train 2.859131 | val 3.289458 | tok/s 2608378.52 | norm 0.2809 | dt 201.00 ms | lr 8.7468e-04 | mlr 8.7468e-04 | mem 13.73/16.12 GiB step 26501 | train 2.759243 | val 3.289458 | tok/s 2626735.09 | norm 0.2748 | dt 199.60 ms | lr 8.7099e-04 | mlr 8.7099e-04 | mem 13.73/16.12 GiB step 26502 | train 2.796804 | val 3.289458 | tok/s 2625855.94 | norm 0.2853 | dt 199.66 ms | lr 8.6726e-04 | mlr 8.6726e-04 | mem 13.73/16.12 GiB step 26503 | train 2.895804 | val 3.289458 | tok/s 2623801.92 | norm 0.3050 | dt 199.82 ms | lr 8.6348e-04 | mlr 8.6348e-04 | mem 13.73/16.12 GiB step 26504 | train 2.821807 | val 3.289458 | tok/s 2628384.52 | norm 0.3057 | dt 199.47 ms | lr 8.5966e-04 | mlr 8.5966e-04 | mem 13.73/16.12 GiB step 26505 | train 2.845270 | val 3.289458 | tok/s 2629593.36 | norm 0.2944 | dt 199.38 ms | lr 8.5580e-04 | mlr 8.5580e-04 | mem 13.73/16.12 GiB step 26506 | train 2.791733 | val 3.289458 | tok/s 2547419.96 | norm 0.2998 | dt 205.81 ms | lr 8.5189e-04 | mlr 8.5189e-04 | mem 13.73/16.12 GiB step 26507 | train 2.891274 | val 3.289458 | tok/s 2629338.31 | norm 0.2890 | dt 199.40 ms | lr 8.4795e-04 | mlr 8.4795e-04 | mem 13.73/16.12 GiB step 26508 | train 2.892620 | val 3.289458 | tok/s 2625905.92 | norm 0.3076 | dt 199.66 ms | lr 8.4396e-04 | mlr 8.4396e-04 | mem 13.73/16.12 GiB step 26509 | train 2.948014 | val 3.289458 | tok/s 2629895.44 | norm 0.3209 | dt 199.36 ms | lr 8.3993e-04 | mlr 8.3993e-04 | mem 13.73/16.12 GiB step 26510 | train 2.908050 | val 3.289458 | tok/s 2625946.57 | norm 0.2940 | dt 199.66 ms | lr 8.3586e-04 | mlr 8.3586e-04 | mem 13.73/16.12 GiB step 26511 | train 2.940177 | val 3.289458 | tok/s 2615461.50 | norm 0.2880 | dt 200.46 ms | lr 8.3175e-04 | mlr 8.3175e-04 | mem 13.73/16.12 GiB step 26512 | train 2.849379 | val 3.289458 | tok/s 2628454.89 | norm 0.2926 | dt 199.47 ms | lr 8.2760e-04 | mlr 8.2760e-04 | mem 13.73/16.12 GiB step 26513 | train 2.936800 | val 3.289458 | tok/s 2622185.38 | norm 0.3358 | dt 199.94 ms | lr 8.2342e-04 | mlr 8.2342e-04 | mem 13.73/16.12 GiB step 26514 | train 2.845877 | val 3.289458 | tok/s 2631674.24 | norm 0.2907 | dt 199.22 ms | lr 8.1919e-04 | mlr 8.1919e-04 | mem 13.73/16.12 GiB step 26515 | train 2.935906 | val 3.289458 | tok/s 2621833.73 | norm 0.2698 | dt 199.97 ms | lr 8.1493e-04 | mlr 8.1493e-04 | mem 13.73/16.12 GiB step 26516 | train 2.960000 | val 3.289458 | tok/s 2564605.35 | norm 0.2757 | dt 204.43 ms | lr 8.1063e-04 | mlr 8.1063e-04 | mem 13.73/16.12 GiB step 26517 | train 2.990458 | val 3.289458 | tok/s 2529923.92 | norm 0.2637 | dt 207.23 ms | lr 8.0630e-04 | mlr 8.0630e-04 | mem 13.73/16.12 GiB step 26518 | train 2.853104 | val 3.289458 | tok/s 2626386.58 | norm 0.2580 | dt 199.62 ms | lr 8.0193e-04 | mlr 8.0193e-04 | mem 13.73/16.12 GiB step 26519 | train 2.869704 | val 3.289458 | tok/s 2628851.78 | norm 0.2736 | dt 199.44 ms | lr 7.9752e-04 | mlr 7.9752e-04 | mem 13.73/16.12 GiB step 26520 | train 2.920492 | val 3.289458 | tok/s 2626169.64 | norm 0.2944 | dt 199.64 ms | lr 7.9309e-04 | mlr 7.9309e-04 | mem 13.73/16.12 GiB step 26521 | train 2.989647 | val 3.289458 | tok/s 2623604.00 | norm 0.2946 | dt 199.84 ms | lr 7.8861e-04 | mlr 7.8861e-04 | mem 13.73/16.12 GiB step 26522 | train 2.834404 | val 3.289458 | tok/s 2627520.12 | norm 0.2786 | dt 199.54 ms | lr 7.8411e-04 | mlr 7.8411e-04 | mem 13.73/16.12 GiB step 26523 | train 2.885770 | val 3.289458 | tok/s 2630007.63 | norm 0.2716 | dt 199.35 ms | lr 7.7957e-04 | mlr 7.7957e-04 | mem 13.73/16.12 GiB step 26524 | train 2.917666 | val 3.289458 | tok/s 2630146.36 | norm 0.3033 | dt 199.34 ms | lr 7.7500e-04 | mlr 7.7500e-04 | mem 13.73/16.12 GiB step 26525 | train 2.942044 | val 3.289458 | tok/s 2619823.54 | norm 0.2889 | dt 200.12 ms | lr 7.7040e-04 | mlr 7.7040e-04 | mem 13.73/16.12 GiB step 26526 | train 2.910398 | val 3.289458 | tok/s 2627683.68 | norm 0.2836 | dt 199.52 ms | lr 7.6577e-04 | mlr 7.6577e-04 | mem 13.73/16.12 GiB step 26527 | train 2.912456 | val 3.289458 | tok/s 2629490.51 | norm 0.3104 | dt 199.39 ms | lr 7.6111e-04 | mlr 7.6111e-04 | mem 13.73/16.12 GiB step 26528 | train 2.878145 | val 3.289458 | tok/s 2625547.05 | norm 0.2847 | dt 199.69 ms | lr 7.5642e-04 | mlr 7.5642e-04 | mem 13.73/16.12 GiB step 26529 | train 2.870641 | val 3.289458 | tok/s 2624636.22 | norm 0.2992 | dt 199.76 ms | lr 7.5170e-04 | mlr 7.5170e-04 | mem 13.73/16.12 GiB step 26530 | train 2.779775 | val 3.289458 | tok/s 2621063.65 | norm 0.3129 | dt 200.03 ms | lr 7.4695e-04 | mlr 7.4695e-04 | mem 13.73/16.12 GiB step 26531 | train 2.969983 | val 3.289458 | tok/s 2623702.27 | norm 0.2991 | dt 199.83 ms | lr 7.4218e-04 | mlr 7.4218e-04 | mem 13.73/16.12 GiB step 26532 | train 2.932211 | val 3.289458 | tok/s 2630058.20 | norm 0.2646 | dt 199.34 ms | lr 7.3737e-04 | mlr 7.3737e-04 | mem 13.73/16.12 GiB step 26533 | train 2.875925 | val 3.289458 | tok/s 2628990.36 | norm 0.2495 | dt 199.43 ms | lr 7.3255e-04 | mlr 7.3255e-04 | mem 13.73/16.12 GiB step 26534 | train 2.872297 | val 3.289458 | tok/s 2629830.70 | norm 0.2691 | dt 199.36 ms | lr 7.2770e-04 | mlr 7.2770e-04 | mem 13.73/16.12 GiB step 26535 | train 2.890844 | val 3.289458 | tok/s 2621097.67 | norm 0.2876 | dt 200.03 ms | lr 7.2282e-04 | mlr 7.2282e-04 | mem 13.73/16.12 GiB step 26536 | train 2.907475 | val 3.289458 | tok/s 2624186.45 | norm 0.2772 | dt 199.79 ms | lr 7.1792e-04 | mlr 7.1792e-04 | mem 13.73/16.12 GiB step 26537 | train 3.023177 | val 3.289458 | tok/s 2623380.71 | norm 0.2600 | dt 199.85 ms | lr 7.1299e-04 | mlr 7.1299e-04 | mem 13.73/16.12 GiB step 26538 | train 2.943886 | val 3.289458 | tok/s 2625583.63 | norm 0.2399 | dt 199.68 ms | lr 7.0805e-04 | mlr 7.0805e-04 | mem 13.73/16.12 GiB step 26539 | train 2.965067 | val 3.289458 | tok/s 2625973.44 | norm 0.2491 | dt 199.65 ms | lr 7.0308e-04 | mlr 7.0308e-04 | mem 13.73/16.12 GiB step 26540 | train 2.888156 | val 3.289458 | tok/s 2624045.97 | norm 0.3084 | dt 199.80 ms | lr 6.9809e-04 | mlr 6.9809e-04 | mem 13.73/16.12 GiB step 26541 | train 2.919208 | val 3.289458 | tok/s 2627421.63 | norm 0.2702 | dt 199.54 ms | lr 6.9308e-04 | mlr 6.9308e-04 | mem 13.73/16.12 GiB step 26542 | train 2.902462 | val 3.289458 | tok/s 2628883.99 | norm 0.2645 | dt 199.43 ms | lr 6.8805e-04 | mlr 6.8805e-04 | mem 13.73/16.12 GiB step 26543 | train 2.980889 | val 3.289458 | tok/s 2628012.15 | norm 0.2535 | dt 199.50 ms | lr 6.8300e-04 | mlr 6.8300e-04 | mem 13.73/16.12 GiB step 26544 | train 2.915236 | val 3.289458 | tok/s 2628509.57 | norm 0.2335 | dt 199.46 ms | lr 6.7793e-04 | mlr 6.7793e-04 | mem 13.73/16.12 GiB step 26545 | train 2.914557 | val 3.289458 | tok/s 2627373.87 | norm 0.2471 | dt 199.55 ms | lr 6.7285e-04 | mlr 6.7285e-04 | mem 13.73/16.12 GiB step 26546 | train 2.980907 | val 3.289458 | tok/s 2626966.94 | norm 0.2708 | dt 199.58 ms | lr 6.6775e-04 | mlr 6.6775e-04 | mem 13.73/16.12 GiB step 26547 | train 2.903224 | val 3.289458 | tok/s 2629681.20 | norm 0.2483 | dt 199.37 ms | lr 6.6263e-04 | mlr 6.6263e-04 | mem 13.73/16.12 GiB step 26548 | train 2.968033 | val 3.289458 | tok/s 2627967.99 | norm 0.2515 | dt 199.50 ms | lr 6.5749e-04 | mlr 6.5749e-04 | mem 13.73/16.12 GiB step 26549 | train 2.868164 | val 3.289458 | tok/s 2626198.13 | norm 0.2327 | dt 199.64 ms | lr 6.5235e-04 | mlr 6.5235e-04 | mem 13.73/16.12 GiB step 26550 | train 2.978411 | val 3.289458 | tok/s 2623169.93 | norm 0.2517 | dt 199.87 ms | lr 6.4718e-04 | mlr 6.4718e-04 | mem 13.73/16.12 GiB step 26551 | train 2.857445 | val 3.289458 | tok/s 2625490.74 | norm 0.2456 | dt 199.69 ms | lr 6.4201e-04 | mlr 6.4201e-04 | mem 13.73/16.12 GiB step 26552 | train 2.800891 | val 3.289458 | tok/s 2586311.18 | norm 0.2378 | dt 202.72 ms | lr 6.3682e-04 | mlr 6.3682e-04 | mem 13.73/16.12 GiB step 26553 | train 2.889100 | val 3.289458 | tok/s 2516665.42 | norm 0.2233 | dt 208.33 ms | lr 6.3162e-04 | mlr 6.3162e-04 | mem 13.73/16.12 GiB step 26554 | train 2.977222 | val 3.289458 | tok/s 2622324.52 | norm 0.2323 | dt 199.93 ms | lr 6.2640e-04 | mlr 6.2640e-04 | mem 13.73/16.12 GiB step 26555 | train 2.883501 | val 3.289458 | tok/s 2626408.35 | norm 0.2272 | dt 199.62 ms | lr 6.2118e-04 | mlr 6.2118e-04 | mem 13.73/16.12 GiB step 26556 | train 2.961716 | val 3.289458 | tok/s 2626728.95 | norm 0.2243 | dt 199.60 ms | lr 6.1595e-04 | mlr 6.1595e-04 | mem 13.73/16.12 GiB step 26557 | train 2.883223 | val 3.289458 | tok/s 2620137.65 | norm 0.2409 | dt 200.10 ms | lr 6.1070e-04 | mlr 6.1070e-04 | mem 13.73/16.12 GiB step 26558 | train 2.854953 | val 3.289458 | tok/s 2467585.80 | norm 0.2267 | dt 212.47 ms | lr 6.0545e-04 | mlr 6.0545e-04 | mem 13.73/16.12 GiB step 26559 | train 3.129373 | val 3.289458 | tok/s 2632341.69 | norm 0.4560 | dt 199.17 ms | lr 6.0020e-04 | mlr 6.0020e-04 | mem 13.73/16.12 GiB step 26560 | train 2.788901 | val 3.289458 | tok/s 2625164.20 | norm 0.2610 | dt 199.72 ms | lr 5.9493e-04 | mlr 5.9493e-04 | mem 13.73/16.12 GiB step 26561 | train 2.818650 | val 3.289458 | tok/s 2629420.38 | norm 0.2189 | dt 199.39 ms | lr 5.8966e-04 | mlr 5.8966e-04 | mem 13.73/16.12 GiB step 26562 | train 2.897565 | val 3.289458 | tok/s 2620190.22 | norm 0.2268 | dt 200.10 ms | lr 5.8438e-04 | mlr 5.8438e-04 | mem 13.73/16.12 GiB step 26563 | train 3.118796 | val 3.289458 | tok/s 2623768.58 | norm 0.4629 | dt 199.82 ms | lr 5.7910e-04 | mlr 5.7910e-04 | mem 13.73/16.12 GiB step 26564 | train 3.053594 | val 3.289458 | tok/s 2632150.40 | norm 0.2624 | dt 199.19 ms | lr 5.7382e-04 | mlr 5.7382e-04 | mem 13.73/16.12 GiB step 26565 | train 2.864825 | val 3.289458 | tok/s 2626031.26 | norm 0.2427 | dt 199.65 ms | lr 5.6853e-04 | mlr 5.6853e-04 | mem 13.73/16.12 GiB step 26566 | train 2.803156 | val 3.289458 | tok/s 2629099.25 | norm 0.2279 | dt 199.42 ms | lr 5.6324e-04 | mlr 5.6324e-04 | mem 13.73/16.12 GiB step 26567 | train 2.961317 | val 3.289458 | tok/s 2610539.93 | norm 0.2530 | dt 200.84 ms | lr 5.5794e-04 | mlr 5.5794e-04 | mem 13.73/16.12 GiB step 26568 | train 2.998026 | val 3.289458 | tok/s 2622646.93 | norm 0.2546 | dt 199.91 ms | lr 5.5265e-04 | mlr 5.5265e-04 | mem 13.73/16.12 GiB step 26569 | train 2.972151 | val 3.289458 | tok/s 2629886.55 | norm 0.2271 | dt 199.36 ms | lr 5.4735e-04 | mlr 5.4735e-04 | mem 13.73/16.12 GiB step 26570 | train 2.914037 | val 3.289458 | tok/s 2623695.04 | norm 0.2275 | dt 199.83 ms | lr 5.4206e-04 | mlr 5.4206e-04 | mem 13.73/16.12 GiB step 26571 | train 2.920666 | val 3.289458 | tok/s 2631070.49 | norm 0.2355 | dt 199.27 ms | lr 5.3676e-04 | mlr 5.3676e-04 | mem 13.73/16.12 GiB step 26572 | train 2.839555 | val 3.289458 | tok/s 2618635.45 | norm 0.2239 | dt 200.21 ms | lr 5.3147e-04 | mlr 5.3147e-04 | mem 13.73/16.12 GiB step 26573 | train 2.763272 | val 3.289458 | tok/s 2622434.37 | norm 0.2267 | dt 199.92 ms | lr 5.2618e-04 | mlr 5.2618e-04 | mem 13.73/16.12 GiB step 26574 | train 2.868431 | val 3.289458 | tok/s 2630026.25 | norm 0.2265 | dt 199.35 ms | lr 5.2090e-04 | mlr 5.2090e-04 | mem 13.73/16.12 GiB step 26575 | train 2.891928 | val 3.289458 | tok/s 2627754.60 | norm 0.2654 | dt 199.52 ms | lr 5.1562e-04 | mlr 5.1562e-04 | mem 13.73/16.12 GiB step 26576 | train 3.265763 | val 3.289458 | tok/s 2628476.22 | norm 0.3113 | dt 199.46 ms | lr 5.1034e-04 | mlr 5.1034e-04 | mem 13.73/16.12 GiB step 26577 | train 3.070178 | val 3.289458 | tok/s 2622786.98 | norm 0.3343 | dt 199.90 ms | lr 5.0507e-04 | mlr 5.0507e-04 | mem 13.73/16.12 GiB step 26578 | train 2.986869 | val 3.289458 | tok/s 2626034.42 | norm 0.2482 | dt 199.65 ms | lr 4.9980e-04 | mlr 4.9980e-04 | mem 13.73/16.12 GiB step 26579 | train 2.821450 | val 3.289458 | tok/s 2630600.91 | norm 0.2231 | dt 199.30 ms | lr 4.9455e-04 | mlr 4.9455e-04 | mem 13.73/16.12 GiB step 26580 | train 2.856515 | val 3.289458 | tok/s 2625116.62 | norm 0.2255 | dt 199.72 ms | lr 4.8930e-04 | mlr 4.8930e-04 | mem 13.73/16.12 GiB step 26581 | train 2.993975 | val 3.289458 | tok/s 2631089.51 | norm 0.2284 | dt 199.27 ms | lr 4.8405e-04 | mlr 4.8405e-04 | mem 13.73/16.12 GiB step 26582 | train 2.927657 | val 3.289458 | tok/s 2622189.79 | norm 0.2454 | dt 199.94 ms | lr 4.7882e-04 | mlr 4.7882e-04 | mem 13.73/16.12 GiB step 26583 | train 2.982417 | val 3.289458 | tok/s 2622770.98 | norm 0.2376 | dt 199.90 ms | lr 4.7360e-04 | mlr 4.7360e-04 | mem 13.73/16.12 GiB step 26584 | train 2.993814 | val 3.289458 | tok/s 2630694.96 | norm 0.2250 | dt 199.30 ms | lr 4.6838e-04 | mlr 4.6838e-04 | mem 13.73/16.12 GiB step 26585 | train 3.000117 | val 3.289458 | tok/s 2627496.58 | norm 0.2071 | dt 199.54 ms | lr 4.6318e-04 | mlr 4.6318e-04 | mem 13.73/16.12 GiB step 26586 | train 3.034226 | val 3.289458 | tok/s 2629916.54 | norm 0.2579 | dt 199.36 ms | lr 4.5799e-04 | mlr 4.5799e-04 | mem 13.73/16.12 GiB step 26587 | train 2.887729 | val 3.289458 | tok/s 2624233.64 | norm 0.2398 | dt 199.79 ms | lr 4.5282e-04 | mlr 4.5282e-04 | mem 13.73/16.12 GiB step 26588 | train 2.934825 | val 3.289458 | tok/s 2626609.55 | norm 0.2179 | dt 199.61 ms | lr 4.4765e-04 | mlr 4.4765e-04 | mem 13.73/16.12 GiB step 26589 | train 2.815868 | val 3.289458 | tok/s 2628239.11 | norm 0.2199 | dt 199.48 ms | lr 4.4251e-04 | mlr 4.4251e-04 | mem 13.73/16.12 GiB step 26590 | train 2.923623 | val 3.289458 | tok/s 2630079.72 | norm 0.2128 | dt 199.34 ms | lr 4.3737e-04 | mlr 4.3737e-04 | mem 13.73/16.12 GiB step 26591 | train 2.940316 | val 3.289458 | tok/s 2629281.89 | norm 0.2103 | dt 199.40 ms | lr 4.3225e-04 | mlr 4.3225e-04 | mem 13.73/16.12 GiB step 26592 | train 3.030907 | val 3.289458 | tok/s 2626392.96 | norm 0.3428 | dt 199.62 ms | lr 4.2715e-04 | mlr 4.2715e-04 | mem 13.73/16.12 GiB step 26593 | train 2.895724 | val 3.289458 | tok/s 2626621.87 | norm 0.2329 | dt 199.61 ms | lr 4.2207e-04 | mlr 4.2207e-04 | mem 13.73/16.12 GiB step 26594 | train 2.899248 | val 3.289458 | tok/s 2630927.77 | norm 0.2302 | dt 199.28 ms | lr 4.1700e-04 | mlr 4.1700e-04 | mem 13.73/16.12 GiB step 26595 | train 2.842007 | val 3.289458 | tok/s 2627610.11 | norm 0.2011 | dt 199.53 ms | lr 4.1195e-04 | mlr 4.1195e-04 | mem 13.73/16.12 GiB step 26596 | train 2.855732 | val 3.289458 | tok/s 2631668.44 | norm 0.2186 | dt 199.22 ms | lr 4.0692e-04 | mlr 4.0692e-04 | mem 13.73/16.12 GiB step 26597 | train 2.973279 | val 3.289458 | tok/s 2622604.62 | norm 0.2140 | dt 199.91 ms | lr 4.0191e-04 | mlr 4.0191e-04 | mem 13.73/16.12 GiB step 26598 | train 2.846132 | val 3.289458 | tok/s 2626596.99 | norm 0.1936 | dt 199.61 ms | lr 3.9692e-04 | mlr 3.9692e-04 | mem 13.73/16.12 GiB step 26599 | train 2.761690 | val 3.289458 | tok/s 2628913.01 | norm 0.1985 | dt 199.43 ms | lr 3.9195e-04 | mlr 3.9195e-04 | mem 13.73/16.12 GiB step 26600 | train 2.959696 | val 3.289458 | tok/s 2627666.87 | norm 0.2474 | dt 199.53 ms | lr 3.8701e-04 | mlr 3.8701e-04 | mem 13.73/16.12 GiB step 26601 | train 2.848764 | val 3.289458 | tok/s 2629946.29 | norm 0.2201 | dt 199.35 ms | lr 3.8208e-04 | mlr 3.8208e-04 | mem 13.73/16.12 GiB step 26602 | train 2.794801 | val 3.289458 | tok/s 2625732.34 | norm 0.1933 | dt 199.67 ms | lr 3.7718e-04 | mlr 3.7718e-04 | mem 13.73/16.12 GiB step 26603 | train 2.910420 | val 3.289458 | tok/s 2627719.67 | norm 0.2065 | dt 199.52 ms | lr 3.7230e-04 | mlr 3.7230e-04 | mem 13.73/16.12 GiB step 26604 | train 2.851387 | val 3.289458 | tok/s 2630829.35 | norm 0.2466 | dt 199.29 ms | lr 3.6745e-04 | mlr 3.6745e-04 | mem 13.73/16.12 GiB step 26605 | train 2.881493 | val 3.289458 | tok/s 2621980.64 | norm 0.2301 | dt 199.96 ms | lr 3.6263e-04 | mlr 3.6263e-04 | mem 13.73/16.12 GiB step 26606 | train 2.911453 | val 3.289458 | tok/s 2631797.37 | norm 0.2034 | dt 199.21 ms | lr 3.5782e-04 | mlr 3.5782e-04 | mem 13.73/16.12 GiB step 26607 | train 2.933882 | val 3.289458 | tok/s 2626945.88 | norm 0.2172 | dt 199.58 ms | lr 3.5305e-04 | mlr 3.5305e-04 | mem 13.73/16.12 GiB step 26608 | train 2.815906 | val 3.289458 | tok/s 2626928.49 | norm 0.2478 | dt 199.58 ms | lr 3.4830e-04 | mlr 3.4830e-04 | mem 13.73/16.12 GiB step 26609 | train 2.837195 | val 3.289458 | tok/s 2621718.43 | norm 0.2078 | dt 199.98 ms | lr 3.4358e-04 | mlr 3.4358e-04 | mem 13.73/16.12 GiB step 26610 | train 2.799098 | val 3.289458 | tok/s 2618089.03 | norm 0.1963 | dt 200.26 ms | lr 3.3889e-04 | mlr 3.3889e-04 | mem 13.73/16.12 GiB step 26611 | train 2.931643 | val 3.289458 | tok/s 2485609.72 | norm 0.1948 | dt 210.93 ms | lr 3.3423e-04 | mlr 3.3423e-04 | mem 13.73/16.12 GiB step 26612 | train 2.808698 | val 3.289458 | tok/s 2627096.84 | norm 0.1912 | dt 199.57 ms | lr 3.2960e-04 | mlr 3.2960e-04 | mem 13.73/16.12 GiB step 26613 | train 2.933851 | val 3.289458 | tok/s 2631050.93 | norm 0.2090 | dt 199.27 ms | lr 3.2500e-04 | mlr 3.2500e-04 | mem 13.73/16.12 GiB step 26614 | train 2.844724 | val 3.289458 | tok/s 2624940.53 | norm 0.1969 | dt 199.73 ms | lr 3.2043e-04 | mlr 3.2043e-04 | mem 13.73/16.12 GiB step 26615 | train 2.871821 | val 3.289458 | tok/s 2624177.37 | norm 0.1963 | dt 199.79 ms | lr 3.1589e-04 | mlr 3.1589e-04 | mem 13.73/16.12 GiB step 26616 | train 2.877568 | val 3.289458 | tok/s 2628973.97 | norm 0.1904 | dt 199.43 ms | lr 3.1139e-04 | mlr 3.1139e-04 | mem 13.73/16.12 GiB step 26617 | train 2.853347 | val 3.289458 | tok/s 2624389.73 | norm 0.1914 | dt 199.78 ms | lr 3.0691e-04 | mlr 3.0691e-04 | mem 13.73/16.12 GiB step 26618 | train 2.964597 | val 3.289458 | tok/s 2630658.15 | norm 0.2140 | dt 199.30 ms | lr 3.0248e-04 | mlr 3.0248e-04 | mem 13.73/16.12 GiB step 26619 | train 2.919396 | val 3.289458 | tok/s 2617755.44 | norm 0.2071 | dt 200.28 ms | lr 2.9807e-04 | mlr 2.9807e-04 | mem 13.73/16.12 GiB step 26620 | train 2.943287 | val 3.289458 | tok/s 2624145.82 | norm 0.2030 | dt 199.79 ms | lr 2.9370e-04 | mlr 2.9370e-04 | mem 13.73/16.12 GiB step 26621 | train 3.005761 | val 3.289458 | tok/s 2627609.21 | norm 0.1901 | dt 199.53 ms | lr 2.8937e-04 | mlr 2.8937e-04 | mem 13.73/16.12 GiB step 26622 | train 2.877430 | val 3.289458 | tok/s 2625171.50 | norm 0.2004 | dt 199.72 ms | lr 2.8507e-04 | mlr 2.8507e-04 | mem 13.73/16.12 GiB step 26623 | train 3.015507 | val 3.289458 | tok/s 2629375.96 | norm 0.1958 | dt 199.40 ms | lr 2.8081e-04 | mlr 2.8081e-04 | mem 13.73/16.12 GiB step 26624 | train 2.891142 | val 3.289458 | tok/s 2619503.40 | norm 0.2167 | dt 200.15 ms | lr 2.7658e-04 | mlr 2.7658e-04 | mem 13.73/16.12 GiB step 26625 | train 2.849949 | val 3.289458 | tok/s 2622485.26 | norm 0.2002 | dt 199.92 ms | lr 2.7240e-04 | mlr 2.7240e-04 | mem 13.73/16.12 GiB step 26626 | train 2.808276 | val 3.289458 | tok/s 2628547.66 | norm 0.1955 | dt 199.46 ms | lr 2.6825e-04 | mlr 2.6825e-04 | mem 13.73/16.12 GiB step 26627 | train 2.823945 | val 3.289458 | tok/s 2621684.45 | norm 0.1944 | dt 199.98 ms | lr 2.6414e-04 | mlr 2.6414e-04 | mem 13.73/16.12 GiB step 26628 | train 2.885769 | val 3.289458 | tok/s 2627946.57 | norm 0.1896 | dt 199.50 ms | lr 2.6007e-04 | mlr 2.6007e-04 | mem 13.73/16.12 GiB step 26629 | train 2.866428 | val 3.289458 | tok/s 2622403.45 | norm 0.1824 | dt 199.93 ms | lr 2.5604e-04 | mlr 2.5604e-04 | mem 13.73/16.12 GiB step 26630 | train 2.799767 | val 3.289458 | tok/s 2621288.84 | norm 0.1878 | dt 200.01 ms | lr 2.5205e-04 | mlr 2.5205e-04 | mem 13.73/16.12 GiB step 26631 | train 2.873062 | val 3.289458 | tok/s 2629253.59 | norm 0.1893 | dt 199.41 ms | lr 2.4811e-04 | mlr 2.4811e-04 | mem 13.73/16.12 GiB step 26632 | train 2.875777 | val 3.289458 | tok/s 2622003.27 | norm 0.1879 | dt 199.96 ms | lr 2.4420e-04 | mlr 2.4420e-04 | mem 13.73/16.12 GiB step 26633 | train 2.867437 | val 3.289458 | tok/s 2631629.42 | norm 0.1872 | dt 199.23 ms | lr 2.4034e-04 | mlr 2.4034e-04 | mem 13.73/16.12 GiB step 26634 | train 2.902358 | val 3.289458 | tok/s 2625680.38 | norm 0.1909 | dt 199.68 ms | lr 2.3652e-04 | mlr 2.3652e-04 | mem 13.73/16.12 GiB step 26635 | train 2.763932 | val 3.289458 | tok/s 2620795.09 | norm 0.1889 | dt 200.05 ms | lr 2.3274e-04 | mlr 2.3274e-04 | mem 13.73/16.12 GiB step 26636 | train 2.885804 | val 3.289458 | tok/s 2632196.89 | norm 0.1901 | dt 199.18 ms | lr 2.2901e-04 | mlr 2.2901e-04 | mem 13.73/16.12 GiB step 26637 | train 2.995324 | val 3.289458 | tok/s 2625700.09 | norm 0.1983 | dt 199.68 ms | lr 2.2532e-04 | mlr 2.2532e-04 | mem 13.73/16.12 GiB step 26638 | train 2.924528 | val 3.289458 | tok/s 2630031.60 | norm 0.1860 | dt 199.35 ms | lr 2.2167e-04 | mlr 2.2167e-04 | mem 13.73/16.12 GiB step 26639 | train 2.909957 | val 3.289458 | tok/s 2625838.87 | norm 0.1823 | dt 199.66 ms | lr 2.1808e-04 | mlr 2.1808e-04 | mem 13.73/16.12 GiB step 26640 | train 2.864637 | val 3.289458 | tok/s 2627122.14 | norm 0.1885 | dt 199.57 ms | lr 2.1452e-04 | mlr 2.1452e-04 | mem 13.73/16.12 GiB step 26641 | train 2.819041 | val 3.289458 | tok/s 2630341.50 | norm 0.1873 | dt 199.32 ms | lr 2.1102e-04 | mlr 2.1102e-04 | mem 13.73/16.12 GiB step 26642 | train 2.820002 | val 3.289458 | tok/s 2625498.05 | norm 0.1766 | dt 199.69 ms | lr 2.0756e-04 | mlr 2.0756e-04 | mem 13.73/16.12 GiB step 26643 | train 2.775635 | val 3.289458 | tok/s 2630517.55 | norm 0.2642 | dt 199.31 ms | lr 2.0415e-04 | mlr 2.0415e-04 | mem 13.73/16.12 GiB step 26644 | train 2.871961 | val 3.289458 | tok/s 2629042.92 | norm 0.1930 | dt 199.42 ms | lr 2.0078e-04 | mlr 2.0078e-04 | mem 13.73/16.12 GiB step 26645 | train 2.953038 | val 3.289458 | tok/s 2628749.46 | norm 0.2043 | dt 199.44 ms | lr 1.9747e-04 | mlr 1.9747e-04 | mem 13.73/16.12 GiB step 26646 | train 2.757096 | val 3.289458 | tok/s 2631633.01 | norm 0.1878 | dt 199.23 ms | lr 1.9420e-04 | mlr 1.9420e-04 | mem 13.73/16.12 GiB step 26647 | train 2.885858 | val 3.289458 | tok/s 2626794.60 | norm 0.2368 | dt 199.59 ms | lr 1.9099e-04 | mlr 1.9099e-04 | mem 13.73/16.12 GiB step 26648 | train 2.960401 | val 3.289458 | tok/s 2629558.76 | norm 0.1923 | dt 199.38 ms | lr 1.8782e-04 | mlr 1.8782e-04 | mem 13.73/16.12 GiB step 26649 | train 2.756675 | val 3.289458 | tok/s 2624199.03 | norm 0.1848 | dt 199.79 ms | lr 1.8470e-04 | mlr 1.8470e-04 | mem 13.73/16.12 GiB step 26650 | train 2.998996 | val 3.289458 | tok/s 2625295.56 | norm 0.1817 | dt 199.71 ms | lr 1.8163e-04 | mlr 1.8163e-04 | mem 13.73/16.12 GiB step 26651 | train 2.993791 | val 3.289458 | tok/s 2534635.45 | norm 0.2226 | dt 206.85 ms | lr 1.7862e-04 | mlr 1.7862e-04 | mem 13.73/16.12 GiB step 26652 | train 2.980755 | val 3.289458 | tok/s 2623831.10 | norm 0.2028 | dt 199.82 ms | lr 1.7565e-04 | mlr 1.7565e-04 | mem 13.73/16.12 GiB step 26653 | train 2.809751 | val 3.289458 | tok/s 2628456.59 | norm 0.1897 | dt 199.47 ms | lr 1.7274e-04 | mlr 1.7274e-04 | mem 13.73/16.12 GiB step 26654 | train 2.910653 | val 3.289458 | tok/s 2612136.71 | norm 0.2024 | dt 200.71 ms | lr 1.6988e-04 | mlr 1.6988e-04 | mem 13.73/16.12 GiB step 26655 | train 2.822073 | val 3.289458 | tok/s 2627782.23 | norm 0.1793 | dt 199.52 ms | lr 1.6707e-04 | mlr 1.6707e-04 | mem 13.73/16.12 GiB step 26656 | train 2.942868 | val 3.289458 | tok/s 2626086.43 | norm 0.1874 | dt 199.65 ms | lr 1.6432e-04 | mlr 1.6432e-04 | mem 13.73/16.12 GiB step 26657 | train 2.857567 | val 3.289458 | tok/s 2621730.00 | norm 0.1952 | dt 199.98 ms | lr 1.6162e-04 | mlr 1.6162e-04 | mem 13.73/16.12 GiB step 26658 | train 2.889300 | val 3.289458 | tok/s 2628084.02 | norm 0.6817 | dt 199.49 ms | lr 1.5897e-04 | mlr 1.5897e-04 | mem 13.73/16.12 GiB step 26659 | train 2.961637 | val 3.289458 | tok/s 2630573.30 | norm 0.2941 | dt 199.31 ms | lr 1.5638e-04 | mlr 1.5638e-04 | mem 13.73/16.12 GiB step 26660 | train 2.950912 | val 3.289458 | tok/s 2632293.30 | norm 0.2457 | dt 199.18 ms | lr 1.5384e-04 | mlr 1.5384e-04 | mem 13.73/16.12 GiB step 26661 | train 3.016022 | val 3.289458 | tok/s 2625067.59 | norm 0.2288 | dt 199.72 ms | lr 1.5136e-04 | mlr 1.5136e-04 | mem 13.73/16.12 GiB step 26662 | train 2.818035 | val 3.289458 | tok/s 2625790.40 | norm 0.5807 | dt 199.67 ms | lr 1.4893e-04 | mlr 1.4893e-04 | mem 13.73/16.12 GiB step 26663 | train 2.885356 | val 3.289458 | tok/s 2630688.32 | norm 0.1885 | dt 199.30 ms | lr 1.4655e-04 | mlr 1.4655e-04 | mem 13.73/16.12 GiB step 26664 | train 2.857085 | val 3.289458 | tok/s 2626651.89 | norm 0.2030 | dt 199.60 ms | lr 1.4424e-04 | mlr 1.4424e-04 | mem 13.73/16.12 GiB step 26665 | train 2.937424 | val 3.289458 | tok/s 2631519.93 | norm 0.2070 | dt 199.23 ms | lr 1.4198e-04 | mlr 1.4198e-04 | mem 13.73/16.12 GiB step 26666 | train 2.905941 | val 3.289458 | tok/s 2625147.93 | norm 0.2004 | dt 199.72 ms | lr 1.3977e-04 | mlr 1.3977e-04 | mem 13.73/16.12 GiB step 26667 | train 2.821992 | val 3.289458 | tok/s 2626144.88 | norm 0.1979 | dt 199.64 ms | lr 1.3762e-04 | mlr 1.3762e-04 | mem 13.73/16.12 GiB step 26668 | train 2.918719 | val 3.289458 | tok/s 2626928.30 | norm 0.1910 | dt 199.58 ms | lr 1.3553e-04 | mlr 1.3553e-04 | mem 13.73/16.12 GiB step 26669 | train 2.796809 | val 3.289458 | tok/s 2629517.89 | norm 0.1883 | dt 199.39 ms | lr 1.3350e-04 | mlr 1.3350e-04 | mem 13.73/16.12 GiB step 26670 | train 2.758191 | val 3.289458 | tok/s 2630677.97 | norm 0.1953 | dt 199.30 ms | lr 1.3152e-04 | mlr 1.3152e-04 | mem 13.73/16.12 GiB step 26671 | train 2.910337 | val 3.289458 | tok/s 2618609.61 | norm 0.1953 | dt 200.22 ms | lr 1.2960e-04 | mlr 1.2960e-04 | mem 13.73/16.12 GiB step 26672 | train 2.922848 | val 3.289458 | tok/s 2624130.53 | norm 0.1954 | dt 199.79 ms | lr 1.2775e-04 | mlr 1.2775e-04 | mem 13.73/16.12 GiB step 26673 | train 2.852049 | val 3.289458 | tok/s 2629973.51 | norm 0.1814 | dt 199.35 ms | lr 1.2594e-04 | mlr 1.2594e-04 | mem 13.73/16.12 GiB step 26674 | train 2.822684 | val 3.289458 | tok/s 2596849.90 | norm 0.1924 | dt 201.89 ms | lr 1.2420e-04 | mlr 1.2420e-04 | mem 13.73/16.12 GiB step 26675 | train 2.829536 | val 3.289458 | tok/s 2626071.50 | norm 0.1806 | dt 199.65 ms | lr 1.2252e-04 | mlr 1.2252e-04 | mem 13.73/16.12 GiB step 26676 | train 2.915682 | val 3.289458 | tok/s 2621173.14 | norm 0.2050 | dt 200.02 ms | lr 1.2089e-04 | mlr 1.2089e-04 | mem 13.73/16.12 GiB step 26677 | train 2.775991 | val 3.289458 | tok/s 2623029.32 | norm 0.1782 | dt 199.88 ms | lr 1.1933e-04 | mlr 1.1933e-04 | mem 13.73/16.12 GiB step 26678 | train 2.853154 | val 3.289458 | tok/s 2630289.00 | norm 0.1928 | dt 199.33 ms | lr 1.1782e-04 | mlr 1.1782e-04 | mem 13.73/16.12 GiB step 26679 | train 2.911838 | val 3.289458 | tok/s 2612967.43 | norm 0.1982 | dt 200.65 ms | lr 1.1638e-04 | mlr 1.1638e-04 | mem 13.73/16.12 GiB step 26680 | train 2.848967 | val 3.289458 | tok/s 2631409.49 | norm 0.1925 | dt 199.24 ms | lr 1.1499e-04 | mlr 1.1499e-04 | mem 13.73/16.12 GiB step 26681 | train 3.032769 | val 3.289458 | tok/s 2625002.61 | norm 0.2699 | dt 199.73 ms | lr 1.1367e-04 | mlr 1.1367e-04 | mem 13.73/16.12 GiB step 26682 | train 2.872963 | val 3.289458 | tok/s 2626728.68 | norm 0.1871 | dt 199.60 ms | lr 1.1240e-04 | mlr 1.1240e-04 | mem 13.73/16.12 GiB step 26683 | train 3.005019 | val 3.289458 | tok/s 2632693.33 | norm 0.2056 | dt 199.15 ms | lr 1.1120e-04 | mlr 1.1120e-04 | mem 13.73/16.12 GiB step 26684 | train 2.865924 | val 3.289458 | tok/s 2627547.68 | norm 0.1999 | dt 199.54 ms | lr 1.1005e-04 | mlr 1.1005e-04 | mem 13.73/16.12 GiB step 26685 | train 2.817148 | val 3.289458 | tok/s 2628040.60 | norm 0.1894 | dt 199.50 ms | lr 1.0897e-04 | mlr 1.0897e-04 | mem 13.73/16.12 GiB step 26686 | train 2.803329 | val 3.289458 | tok/s 2623188.25 | norm 0.1889 | dt 199.87 ms | lr 1.0795e-04 | mlr 1.0795e-04 | mem 13.73/16.12 GiB step 26687 | train 2.850069 | val 3.289458 | tok/s 2625224.06 | norm 0.1942 | dt 199.71 ms | lr 1.0699e-04 | mlr 1.0699e-04 | mem 13.73/16.12 GiB step 26688 | train 2.914777 | val 3.289458 | tok/s 2632221.01 | norm 0.1845 | dt 199.18 ms | lr 1.0609e-04 | mlr 1.0609e-04 | mem 13.73/16.12 GiB step 26689 | train 2.873407 | val 3.289458 | tok/s 2631938.18 | norm 0.1800 | dt 199.20 ms | lr 1.0525e-04 | mlr 1.0525e-04 | mem 13.73/16.12 GiB step 26690 | train 2.886518 | val 3.289458 | tok/s 2632178.22 | norm 0.1837 | dt 199.18 ms | lr 1.0448e-04 | mlr 1.0448e-04 | mem 13.73/16.12 GiB step 26691 | train 2.809777 | val 3.289458 | tok/s 2623864.60 | norm 0.1846 | dt 199.82 ms | lr 1.0376e-04 | mlr 1.0376e-04 | mem 13.73/16.12 GiB step 26692 | train 2.774429 | val 3.289458 | tok/s 2623821.67 | norm 0.1789 | dt 199.82 ms | lr 1.0311e-04 | mlr 1.0311e-04 | mem 13.73/16.12 GiB step 26693 | train 3.060335 | val 3.289458 | tok/s 2630992.46 | norm 0.1946 | dt 199.27 ms | lr 1.0252e-04 | mlr 1.0252e-04 | mem 13.73/16.12 GiB step 26694 | train 2.864460 | val 3.289458 | tok/s 2629471.27 | norm 0.1837 | dt 199.39 ms | lr 1.0199e-04 | mlr 1.0199e-04 | mem 13.73/16.12 GiB step 26695 | train 2.858253 | val 3.289458 | tok/s 2629512.83 | norm 0.1926 | dt 199.39 ms | lr 1.0153e-04 | mlr 1.0153e-04 | mem 13.73/16.12 GiB step 26696 | train 2.928054 | val 3.289458 | tok/s 2624519.24 | norm 0.1873 | dt 199.77 ms | lr 1.0112e-04 | mlr 1.0112e-04 | mem 13.73/16.12 GiB step 26697 | train 2.914061 | val 3.289458 | tok/s 2629665.33 | norm 0.1906 | dt 199.37 ms | lr 1.0078e-04 | mlr 1.0078e-04 | mem 13.73/16.12 GiB step 26698 | train 2.914749 | val 3.289458 | tok/s 2630937.71 | norm 0.1773 | dt 199.28 ms | lr 1.0050e-04 | mlr 1.0050e-04 | mem 13.73/16.12 GiB step 26699 | train 2.844621 | val 3.289458 | tok/s 2626687.56 | norm 0.1853 | dt 199.60 ms | lr 1.0028e-04 | mlr 1.0028e-04 | mem 13.73/16.12 GiB step 26700 | train 2.959374 | val 3.289458 | tok/s 2630985.60 | norm 0.2159 | dt 199.27 ms | lr 1.0012e-04 | mlr 1.0012e-04 | mem 13.73/16.12 GiB step 26701 | train 2.849717 | val 3.289458 | tok/s 2626979.85 | norm 0.1940 | dt 199.58 ms | lr 1.0003e-04 | mlr 1.0003e-04 | mem 13.73/16.12 GiB [val] step 26702 | val_loss 3.232783 step 26702 | train 2.819735 | val 3.232783 | tok/s 2612544.72 | norm 0.1691 | dt 200.68 ms | lr 1.0000e-04 | mlr 1.0000e-04 | mem 13.73/16.12 GiB ==================================================================== TARGET LOSS MILESTONES ==================================================================== target loss step hours tokens 2.690595 not reached - - 2.554042 not reached - - 2.505404 not reached - - 2.480678 not reached - - 2.450056 not reached - - 2.437589 not reached - - ==================================================================== ============================================================================== STEADY-STATE SUMMARY — DDP x8 optimizer=muon+adamw precision=bf16-params rms+rope=fused torch.optim.Muon lr=0.001 adjust_lr_fn=match_rms_adamw momentum=0.95 ns_steps=5 (26700 steps, first 3 excluded) ============================================================================== step time : 199.54 ms (mean 199.92) throughput : 2622426.2 tok/s tokens / step : 524288 (last stage) tokens consumed : 14,000,062,464 static (p+g+opt) : 0.93 GiB (per GPU) activations (peak) : 12.80 GiB peak mem allocated : 13.73 GiB peak mem reserved : 16.12 GiB ============================================================================== === Training complete ===