`rope_scaling`'s beta_fast field must be a float, got 32
`rope_scaling`'s beta_slow field must be a float, got 1
`rope_scaling`'s beta_fast field must be a float, got 32
`rope_scaling`'s beta_slow field must be a float, got 1
🦥 Unsloth: Will patch your computer to enable 2x faster free finetuning.
⚙️  Running in WANDB offline mode
🦥 Unsloth Zoo will now patch everything to make training faster!
==((====))==  Unsloth 2026.9.2: Fast Olmo3 patching. Transformers: 4.57.6.
   \\   /|    NVIDIA H100 80GB HBM3. Num GPUs = 1. Max memory: 79.179 GB. Platform: Linux.
O^O/ \_/ \    Torch: 2.11.0+cu128. CUDA: 9.0. CUDA Toolkit: 12.8. Triton: 3.6.0
\        /    Bfloat16 = TRUE. FA [Xformers = 0.0.35. FA2 = False]
 "-____-"     Free license: http://github.com/unslothai/unsloth
Unsloth: Fast downloading is enabled - ignore downloading bars which are red colored!
Unsloth: QLoRA and full finetuning all not selected. Switching to 16bit LoRA.

Loading checkpoint shards:   0%|          | 0/3 [00:00<?, ?it/s]
Loading checkpoint shards:  33%|███▎      | 1/3 [00:00<00:01,  1.13it/s]
Loading checkpoint shards:  67%|██████▋   | 2/3 [00:01<00:00,  1.15it/s]
Loading checkpoint shards: 100%|██████████| 3/3 [00:02<00:00,  1.20it/s]
Loading checkpoint shards: 100%|██████████| 3/3 [00:02<00:00,  1.19it/s]
ROPE_GUARD OK train_unsloth load allenai/Olmo-3-1025-7B: olmo3 theta=500000.0 scaling={'rope_type': 'yarn', 'factor': 8.0, 'original_max_position_embeddings': 8192.0, 'beta_fast': 32.0, 'beta_slow': 1.0, 'attention_factor': 1.2079441541679836}
Unsloth: Dropout = 0 is supported for fast patching. You are using dropout = 0.05.
Unsloth will patch all other layers, except LoRA matrices, causing a performance hit.
Unsloth: Explicit target_modules are constrained by the finetune_(vision|language|attention|mlp) filters; adapters attach only where both select.
Unsloth: Allowing gradients for `base_model.model.model.embed_tokens` since it's in `modules_to_save`.
Unsloth: Allowing gradients for `base_model.model.lm_head` since it's in `modules_to_save`.

Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 666 examples [00:00, 1082.16 examples/s]
Generating train split: 1308 examples [00:00, 1707.96 examples/s]
Generating train split: 1991 examples [00:01, 2054.25 examples/s]
Generating train split: 2648 examples [00:01, 1733.29 examples/s]
Generating train split: 3299 examples [00:01, 1995.85 examples/s]
Generating train split: 3983 examples [00:02, 1680.16 examples/s]
Generating train split: 4619 examples [00:02, 1895.30 examples/s]
Generating train split: 5267 examples [00:02, 2111.71 examples/s]
Generating train split: 5946 examples [00:03, 1816.35 examples/s]
Generating train split: 6655 examples [00:03, 2029.22 examples/s]
Generating train split: 7317 examples [00:04, 1712.43 examples/s]
Generating train split: 7996 examples [00:04, 1933.20 examples/s]
Generating train split: 8636 examples [00:04, 2050.00 examples/s]
Generating train split: 9250 examples [00:05, 1705.56 examples/s]
Generating train split: 9250 examples [00:05, 1810.59 examples/s]

Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 5000 examples [00:00, 8010.55 examples/s]
Generating train split: 5000 examples [00:00, 7809.35 examples/s]

tokenizing (num_proc=8):   0%|          | 0/14250 [00:00<?, ? examples/s]
tokenizing (num_proc=8):   0%|          | 7/14250 [00:00<20:37, 11.51 examples/s]
tokenizing (num_proc=8):   0%|          | 43/14250 [00:00<03:04, 77.13 examples/s]
tokenizing (num_proc=8):   1%|          | 115/14250 [00:00<01:06, 213.11 examples/s]
tokenizing (num_proc=8):   2%|▏         | 254/14250 [00:00<00:29, 472.41 examples/s]
tokenizing (num_proc=8):   3%|▎         | 426/14250 [00:01<00:17, 772.42 examples/s]
tokenizing (num_proc=8):   4%|▍         | 581/14250 [00:01<00:14, 971.30 examples/s]
tokenizing (num_proc=8):   5%|▌         | 767/14250 [00:01<00:11, 1185.65 examples/s]
tokenizing (num_proc=8):   7%|▋         | 957/14250 [00:01<00:09, 1363.39 examples/s]
tokenizing (num_proc=8):   8%|▊         | 1138/14250 [00:01<00:08, 1464.92 examples/s]
tokenizing (num_proc=8):   9%|▉         | 1298/14250 [00:01<00:08, 1462.99 examples/s]
tokenizing (num_proc=8):  10%|█         | 1485/14250 [00:01<00:08, 1536.51 examples/s]
tokenizing (num_proc=8):  12%|█▏        | 1654/14250 [00:01<00:08, 1510.57 examples/s]
tokenizing (num_proc=8):  13%|█▎        | 1811/14250 [00:01<00:08, 1505.85 examples/s]
tokenizing (num_proc=8):  14%|█▍        | 2009/14250 [00:01<00:07, 1623.43 examples/s]
tokenizing (num_proc=8):  15%|█▌        | 2184/14250 [00:02<00:07, 1642.48 examples/s]
tokenizing (num_proc=8):  17%|█▋        | 2370/14250 [00:02<00:06, 1701.44 examples/s]
tokenizing (num_proc=8):  18%|█▊        | 2544/14250 [00:02<00:07, 1585.09 examples/s]
tokenizing (num_proc=8):  19%|█▉        | 2737/14250 [00:02<00:07, 1627.67 examples/s]
tokenizing (num_proc=8):  20%|██        | 2917/14250 [00:02<00:06, 1654.15 examples/s]
tokenizing (num_proc=8):  22%|██▏       | 3100/14250 [00:02<00:06, 1665.20 examples/s]
tokenizing (num_proc=8):  23%|██▎       | 3268/14250 [00:02<00:06, 1602.08 examples/s]
tokenizing (num_proc=8):  24%|██▍       | 3469/14250 [00:02<00:06, 1690.23 examples/s]
tokenizing (num_proc=8):  26%|██▌       | 3650/14250 [00:02<00:06, 1685.48 examples/s]
tokenizing (num_proc=8):  27%|██▋       | 3826/14250 [00:03<00:06, 1647.17 examples/s]
tokenizing (num_proc=8):  28%|██▊       | 3997/14250 [00:03<00:06, 1577.13 examples/s]
tokenizing (num_proc=8):  29%|██▉       | 4157/14250 [00:03<00:06, 1558.27 examples/s]
tokenizing (num_proc=8):  30%|███       | 4320/14250 [00:03<00:06, 1520.91 examples/s]
tokenizing (num_proc=8):  31%|███▏      | 4481/14250 [00:03<00:06, 1544.44 examples/s]
tokenizing (num_proc=8):  33%|███▎      | 4636/14250 [00:03<00:06, 1535.47 examples/s]
tokenizing (num_proc=8):  34%|███▎      | 4793/14250 [00:03<00:06, 1544.38 examples/s]
tokenizing (num_proc=8):  35%|███▍      | 4970/14250 [00:03<00:05, 1578.49 examples/s]
tokenizing (num_proc=8):  36%|███▌      | 5134/14250 [00:03<00:05, 1551.05 examples/s]
tokenizing (num_proc=8):  37%|███▋      | 5342/14250 [00:04<00:05, 1660.28 examples/s]
tokenizing (num_proc=8):  39%|███▊      | 5509/14250 [00:04<00:05, 1583.78 examples/s]
tokenizing (num_proc=8):  40%|███▉      | 5669/14250 [00:04<00:05, 1527.58 examples/s]
tokenizing (num_proc=8):  41%|████      | 5853/14250 [00:04<00:05, 1592.46 examples/s]
tokenizing (num_proc=8):  42%|████▏     | 6025/14250 [00:04<00:05, 1539.09 examples/s]
tokenizing (num_proc=8):  43%|████▎     | 6188/14250 [00:04<00:05, 1521.44 examples/s]
tokenizing (num_proc=8):  45%|████▍     | 6348/14250 [00:04<00:05, 1494.84 examples/s]
tokenizing (num_proc=8):  46%|████▌     | 6501/14250 [00:04<00:05, 1491.33 examples/s]
tokenizing (num_proc=8):  47%|████▋     | 6667/14250 [00:04<00:05, 1501.41 examples/s]
tokenizing (num_proc=8):  48%|████▊     | 6824/14250 [00:05<00:04, 1499.45 examples/s]
tokenizing (num_proc=8):  49%|████▉     | 6979/14250 [00:05<00:05, 1448.80 examples/s]
tokenizing (num_proc=8):  50%|█████     | 7130/14250 [00:05<00:04, 1429.87 examples/s]
tokenizing (num_proc=8):  51%|█████     | 7300/14250 [00:05<00:04, 1486.08 examples/s]
tokenizing (num_proc=8):  53%|█████▎    | 7482/14250 [00:05<00:04, 1554.99 examples/s]
tokenizing (num_proc=8):  54%|█████▎    | 7640/14250 [00:05<00:04, 1410.70 examples/s]
tokenizing (num_proc=8):  55%|█████▍    | 7791/14250 [00:05<00:05, 1227.83 examples/s]
tokenizing (num_proc=8):  56%|█████▌    | 7925/14250 [00:06<00:06, 983.42 examples/s] 
tokenizing (num_proc=8):  56%|█████▋    | 8036/14250 [00:06<00:10, 582.00 examples/s]
tokenizing (num_proc=8):  57%|█████▋    | 8131/14250 [00:06<00:09, 612.78 examples/s]
tokenizing (num_proc=8):  58%|█████▊    | 8222/14250 [00:06<00:09, 657.87 examples/s]
tokenizing (num_proc=8):  59%|█████▊    | 8347/14250 [00:06<00:07, 773.66 examples/s]
tokenizing (num_proc=8):  59%|█████▉    | 8465/14250 [00:06<00:06, 849.51 examples/s]
tokenizing (num_proc=8):  60%|██████    | 8618/14250 [00:06<00:05, 1007.43 examples/s]
tokenizing (num_proc=8):  62%|██████▏   | 8791/14250 [00:07<00:04, 1172.57 examples/s]
tokenizing (num_proc=8):  63%|██████▎   | 8932/14250 [00:07<00:04, 1227.26 examples/s]
tokenizing (num_proc=8):  64%|██████▎   | 9075/14250 [00:07<00:04, 1276.87 examples/s]
tokenizing (num_proc=8):  65%|██████▍   | 9234/14250 [00:07<00:03, 1363.29 examples/s]
tokenizing (num_proc=8):  66%|██████▌   | 9387/14250 [00:07<00:03, 1396.37 examples/s]
tokenizing (num_proc=8):  67%|██████▋   | 9558/14250 [00:07<00:03, 1453.05 examples/s]
tokenizing (num_proc=8):  68%|██████▊   | 9711/14250 [00:07<00:03, 1460.08 examples/s]
tokenizing (num_proc=8):  69%|██████▉   | 9881/14250 [00:07<00:02, 1517.12 examples/s]
tokenizing (num_proc=8):  71%|███████   | 10093/14250 [00:07<00:02, 1666.38 examples/s]
tokenizing (num_proc=8):  72%|███████▏  | 10287/14250 [00:08<00:02, 1716.59 examples/s]
tokenizing (num_proc=8):  73%|███████▎  | 10464/14250 [00:08<00:02, 1642.87 examples/s]
tokenizing (num_proc=8):  75%|███████▍  | 10645/14250 [00:08<00:02, 1640.68 examples/s]
tokenizing (num_proc=8):  76%|███████▌  | 10820/14250 [00:08<00:02, 1631.71 examples/s]
tokenizing (num_proc=8):  77%|███████▋  | 10995/14250 [00:08<00:02, 1612.62 examples/s]
tokenizing (num_proc=8):  78%|███████▊  | 11176/14250 [00:08<00:01, 1615.53 examples/s]
tokenizing (num_proc=8):  80%|███████▉  | 11352/14250 [00:08<00:01, 1628.01 examples/s]
tokenizing (num_proc=8):  81%|████████  | 11518/14250 [00:08<00:01, 1633.52 examples/s]
tokenizing (num_proc=8):  82%|████████▏ | 11706/14250 [00:08<00:01, 1679.80 examples/s]
tokenizing (num_proc=8):  83%|████████▎ | 11894/14250 [00:08<00:01, 1724.75 examples/s]
tokenizing (num_proc=8):  85%|████████▍ | 12078/14250 [00:09<00:01, 1751.27 examples/s]
tokenizing (num_proc=8):  86%|████████▌ | 12262/14250 [00:09<00:01, 1719.10 examples/s]
tokenizing (num_proc=8):  87%|████████▋ | 12463/14250 [00:09<00:01, 1750.59 examples/s]
tokenizing (num_proc=8):  89%|████████▊ | 12640/14250 [00:09<00:00, 1683.74 examples/s]
tokenizing (num_proc=8):  90%|████████▉ | 12821/14250 [00:09<00:00, 1692.39 examples/s]
tokenizing (num_proc=8):  91%|█████████ | 13002/14250 [00:09<00:00, 1722.06 examples/s]
tokenizing (num_proc=8):  93%|█████████▎| 13183/14250 [00:09<00:00, 1671.07 examples/s]
tokenizing (num_proc=8):  94%|█████████▍| 13360/14250 [00:09<00:00, 1557.49 examples/s]
tokenizing (num_proc=8):  95%|█████████▍| 13519/14250 [00:10<00:00, 1452.19 examples/s]
tokenizing (num_proc=8):  96%|█████████▌| 13671/14250 [00:10<00:00, 1366.25 examples/s]
tokenizing (num_proc=8):  97%|█████████▋| 13818/14250 [00:10<00:00, 1238.60 examples/s]
tokenizing (num_proc=8):  98%|█████████▊| 13948/14250 [00:10<00:00, 1094.77 examples/s]
tokenizing (num_proc=8):  99%|█████████▉| 14072/14250 [00:10<00:00, 989.95 examples/s] 
tokenizing (num_proc=8):  99%|█████████▉| 14177/14250 [00:10<00:00, 884.58 examples/s]
tokenizing (num_proc=8): 100%|██████████| 14250/14250 [00:11<00:00, 1245.71 examples/s]

Filter:   0%|          | 0/14250 [00:00<?, ? examples/s]
Filter:   7%|▋         | 1000/14250 [00:00<00:03, 3895.81 examples/s]
Filter:  14%|█▍        | 2000/14250 [00:00<00:02, 4620.30 examples/s]
Filter:  21%|██        | 3000/14250 [00:00<00:02, 4453.77 examples/s]
Filter:  28%|██▊       | 4000/14250 [00:00<00:02, 4681.14 examples/s]
Filter:  35%|███▌      | 5000/14250 [00:01<00:01, 4759.22 examples/s]
Filter:  42%|████▏     | 6000/14250 [00:01<00:01, 4693.79 examples/s]
Filter:  49%|████▉     | 7000/14250 [00:01<00:01, 4815.32 examples/s]
Filter:  56%|█████▌    | 8000/14250 [00:01<00:01, 4963.41 examples/s]
Filter:  63%|██████▎   | 9000/14250 [00:01<00:01, 4978.38 examples/s]
Filter:  70%|███████   | 10000/14250 [00:02<00:00, 5111.16 examples/s]
Filter:  77%|███████▋  | 11000/14250 [00:02<00:00, 5018.46 examples/s]
Filter:  84%|████████▍ | 12000/14250 [00:02<00:00, 4852.32 examples/s]
Filter:  91%|█████████ | 13000/14250 [00:02<00:00, 4956.03 examples/s]
Filter:  98%|█████████▊| 14000/14250 [00:02<00:00, 4742.63 examples/s]
Filter: 100%|██████████| 14250/14250 [00:02<00:00, 4829.84 examples/s]

Filter:   0%|          | 0/14250 [00:00<?, ? examples/s]
Filter:   7%|▋         | 1000/14250 [00:00<00:02, 4719.50 examples/s]
Filter:  14%|█▍        | 2000/14250 [00:00<00:02, 4881.84 examples/s]
Filter:  21%|██        | 3000/14250 [00:00<00:02, 4545.55 examples/s]
Filter:  28%|██▊       | 4000/14250 [00:00<00:02, 4662.41 examples/s]
Filter:  35%|███▌      | 5000/14250 [00:01<00:01, 4656.56 examples/s]
Filter:  42%|████▏     | 6000/14250 [00:01<00:01, 4753.82 examples/s]
Filter:  49%|████▉     | 7000/14250 [00:01<00:01, 4746.76 examples/s]
Filter:  56%|█████▌    | 8000/14250 [00:01<00:01, 4691.69 examples/s]
Filter:  63%|██████▎   | 9000/14250 [00:01<00:01, 4833.64 examples/s]
Filter:  70%|███████   | 10000/14250 [00:02<00:00, 4612.92 examples/s]
Filter:  77%|███████▋  | 11000/14250 [00:02<00:00, 4673.15 examples/s]
Filter:  84%|████████▍ | 12000/14250 [00:02<00:00, 4669.64 examples/s]
Filter:  91%|█████████ | 13000/14250 [00:02<00:00, 4606.14 examples/s]
Filter:  98%|█████████▊| 14000/14250 [00:03<00:00, 4419.02 examples/s]
Filter: 100%|██████████| 14250/14250 [00:03<00:00, 4627.10 examples/s]
supervised tokens: tool 4221981, replay 0 (0.0% replay)
rows: 14250 raw -> 14138 kept; by source: {'apigen': 4888, 'recovery': 3750, 'xlam': 5000, 'efficiency': 500}
[manifest] git=534143b028a9765e904b4d6d1b22205027804f19  data_sha256={'train_olmo-pretrained_replay50.jsonl': '64d5d5af76e7', 'xlam5000_max1call.jsonl': 'a8aae6cf5c36'}
wandb: WARNING git root /workspace/anchor/olmo-pretrained/wandb does not exist
wandb: Tracking run with wandb version 0.30.0
wandb: W&B syncing is set to `offline` in this directory. Run `wandb online` or set WANDB_MODE=online to enable cloud syncing.
wandb: Run data is saved locally in /workspace/anchor/olmo-pretrained/wandb/wandb/offline-run-20261016_031238-pqlcax1d
wandb: View this run in the terminal with `wandb leet`
wandb: Detected [huggingface_hub.inference] in use.
wandb: Use W&B Weave for improved LLM call tracing. Install Weave with `pip install weave` then add `import weave` to the top of your script.
wandb: For more information, check out the docs at: https://weave-docs.wandb.ai
==((====))==  Unsloth - 2x faster free finetuning | Num GPUs used = 1
   \\   /|    Num examples = 13,784 | Num Epochs = 1 | Total steps = 862
O^O/ \_/ \    Batch size per device = 2 | Gradient accumulation steps = 8
\        /    Data Parallel GPUs = 1 | Total batch size (2 x 8 x 1) = 16
 "-____-"     Trainable parameters = 981,385,216 of 8,279,396,352 (11.85% trained)
Unsloth: Setting lr = 2.00e-05 instead of 2.00e-04 for embed_tokens.
Unsloth: Setting lr = 2.00e-05 instead of 2.00e-04 for lm_head.

  0%|          | 0/862 [00:00<?, ?it/s]
  0%|          | 1/862 [00:44<10:42:30, 44.77s/it]
  0%|          | 2/862 [01:12<8:15:15, 34.55s/it] 
  0%|          | 3/862 [01:34<6:57:15, 29.15s/it]
  0%|          | 4/862 [01:59<6:30:55, 27.34s/it]
  1%|          | 5/862 [02:23<6:12:21, 26.07s/it]
                                                 

  1%|          | 5/862 [02:23<6:12:21, 26.07s/it]
  1%|          | 6/862 [02:43<5:43:34, 24.08s/it]
  1%|          | 7/862 [03:10<5:57:20, 25.08s/it]
  1%|          | 8/862 [03:35<5:57:27, 25.11s/it]
  1%|          | 9/862 [03:58<5:44:53, 24.26s/it]
  1%|          | 10/862 [04:21<5:40:37, 23.99s/it]
                                                  

  1%|          | 10/862 [04:21<5:40:37, 23.99s/it]
  1%|▏         | 11/862 [04:42<5:28:19, 23.15s/it]
  1%|▏         | 12/862 [05:06<5:30:25, 23.32s/it]
  2%|▏         | 13/862 [05:26<5:16:39, 22.38s/it]
  2%|▏         | 14/862 [05:49<5:18:53, 22.56s/it]
  2%|▏         | 15/862 [06:13<5:24:49, 23.01s/it]
                                                  

  2%|▏         | 15/862 [06:13<5:24:49, 23.01s/it]
  2%|▏         | 16/862 [06:38<5:31:17, 23.50s/it]
  2%|▏         | 17/862 [07:00<5:24:03, 23.01s/it]
  2%|▏         | 18/862 [07:18<5:04:18, 21.63s/it]
  2%|▏         | 19/862 [07:41<5:09:50, 22.05s/it]
  2%|▏         | 20/862 [08:06<5:19:10, 22.74s/it]
                                                  

  2%|▏         | 20/862 [08:06<5:19:10, 22.74s/it]
  2%|▏         | 21/862 [08:31<5:28:57, 23.47s/it]
  3%|▎         | 22/862 [08:53<5:25:10, 23.23s/it]
  3%|▎         | 23/862 [09:09<4:52:17, 20.90s/it]
  3%|▎         | 24/862 [09:31<4:55:56, 21.19s/it]
  3%|▎         | 25/862 [09:52<4:56:06, 21.23s/it]
                                                  

  3%|▎         | 25/862 [09:52<4:56:06, 21.23s/it]
  3%|▎         | 26/862 [10:13<4:54:07, 21.11s/it]
  3%|▎         | 27/862 [10:37<5:04:40, 21.89s/it]
  3%|▎         | 28/862 [10:51<4:34:34, 19.75s/it]
  3%|▎         | 29/862 [11:15<4:51:40, 21.01s/it]
  3%|▎         | 30/862 [11:39<5:01:39, 21.75s/it]
                                                  

  3%|▎         | 30/862 [11:39<5:01:39, 21.75s/it]
  4%|▎         | 31/862 [12:02<5:07:38, 22.21s/it]
  4%|▎         | 32/862 [12:21<4:51:57, 21.11s/it]
  4%|▍         | 33/862 [12:46<5:10:15, 22.45s/it]
  4%|▍         | 34/862 [13:06<4:57:46, 21.58s/it]
  4%|▍         | 35/862 [13:25<4:49:17, 20.99s/it]
                                                  

  4%|▍         | 35/862 [13:25<4:49:17, 20.99s/it]
  4%|▍         | 36/862 [13:48<4:57:37, 21.62s/it]
  4%|▍         | 37/862 [14:08<4:48:28, 20.98s/it]
  4%|▍         | 38/862 [14:33<5:06:16, 22.30s/it]
  5%|▍         | 39/862 [14:56<5:05:56, 22.30s/it]
  5%|▍         | 40/862 [15:16<4:56:12, 21.62s/it]
                                                  

  5%|▍         | 40/862 [15:16<4:56:12, 21.62s/it]
  5%|▍         | 41/862 [15:37<4:54:10, 21.50s/it]
  5%|▍         | 42/862 [16:00<5:00:48, 22.01s/it]
  5%|▍         | 43/862 [16:26<5:15:50, 23.14s/it]
  5%|▌         | 44/862 [16:44<4:53:24, 21.52s/it]
  5%|▌         | 45/862 [17:08<5:06:27, 22.51s/it]
                                                  

  5%|▌         | 45/862 [17:08<5:06:27, 22.51s/it]
  5%|▌         | 46/862 [17:28<4:53:48, 21.60s/it]
  5%|▌         | 47/862 [17:55<5:16:23, 23.29s/it]
  6%|▌         | 48/862 [18:17<5:09:44, 22.83s/it]
  6%|▌         | 49/862 [18:40<5:10:24, 22.91s/it]
  6%|▌         | 50/862 [19:03<5:12:02, 23.06s/it]
                                                  

  6%|▌         | 50/862 [19:03<5:12:02, 23.06s/it]Unsloth: Will smartly offload gradients to save VRAM!
Unsloth: Double buffering enabled (parallel H2D + compute) for backward pass.
{'loss': 0.9941, 'grad_norm': 1.8992259502410889, 'learning_rate': 1.8181818181818182e-05, 'epoch': 0.01}
{'loss': 0.8727, 'grad_norm': 1.5108100175857544, 'learning_rate': 4.0909090909090915e-05, 'epoch': 0.01}
{'loss': 0.7164, 'grad_norm': 1.313894271850586, 'learning_rate': 6.363636363636364e-05, 'epoch': 0.02}
{'loss': 0.5324, 'grad_norm': 1.5354955196380615, 'learning_rate': 8.636363636363637e-05, 'epoch': 0.02}
{'loss': 0.4492, 'grad_norm': 1.4390828609466553, 'learning_rate': 0.00010909090909090909, 'epoch': 0.03}
{'loss': 0.4172, 'grad_norm': 1.8266173601150513, 'learning_rate': 0.0001318181818181818, 'epoch': 0.03}
{'loss': 0.3966, 'grad_norm': 1.5203430652618408, 'learning_rate': 0.00015454545454545454, 'epoch': 0.04}
{'loss': 0.3584, 'grad_norm': 1.6018481254577637, 'learning_rate': 0.00017727272727272728, 'epoch': 0.05}
{'loss': 0.3571, 'grad_norm': 1.3058902025222778, 'learning_rate': 0.0002, 'epoch': 0.05}
{'loss': 0.2975, 'grad_norm': 1.388102650642395, 'learning_rate': 0.00019998156301911284, 'epoch': 0.06}


Filter:   0%|          | 0/354 [00:00<?, ? examples/s]

Filter: 100%|██████████| 354/354 [00:00<00:00, 1481.01 examples/s]
Filter: 100%|██████████| 354/354 [00:00<00:00, 1467.34 examples/s]


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:01<02:04,  1.40it/s]

  2%|▏         | 3/177 [00:02<02:28,  1.17it/s]

  2%|▏         | 4/177 [00:03<02:37,  1.10it/s]

  3%|▎         | 5/177 [00:04<02:33,  1.12it/s]

  3%|▎         | 6/177 [00:04<01:58,  1.44it/s]

  4%|▍         | 7/177 [00:05<02:18,  1.23it/s]

  5%|▍         | 8/177 [00:06<02:18,  1.22it/s]

  5%|▌         | 9/177 [00:07<02:20,  1.20it/s]

  6%|▌         | 10/177 [00:08<02:21,  1.18it/s]

  6%|▌         | 11/177 [00:09<02:30,  1.10it/s]

  7%|▋         | 12/177 [00:09<01:55,  1.42it/s]

  7%|▋         | 13/177 [00:09<01:39,  1.65it/s]

  8%|▊         | 14/177 [00:10<02:00,  1.36it/s]

  8%|▊         | 15/177 [00:11<02:03,  1.31it/s]

  9%|▉         | 16/177 [00:12<02:13,  1.20it/s]

 10%|▉         | 17/177 [00:13<01:46,  1.50it/s]

 10%|█         | 18/177 [00:13<01:53,  1.41it/s]

 11%|█         | 19/177 [00:14<02:07,  1.24it/s]

 11%|█▏        | 20/177 [00:15<02:05,  1.26it/s]

 12%|█▏        | 21/177 [00:16<02:10,  1.19it/s]

 12%|█▏        | 22/177 [00:18<02:35,  1.00s/it]

 13%|█▎        | 23/177 [00:19<02:54,  1.13s/it]

 14%|█▎        | 24/177 [00:20<02:56,  1.15s/it]

 14%|█▍        | 25/177 [00:20<02:16,  1.12it/s]

 15%|█▍        | 26/177 [00:21<02:09,  1.17it/s]

 15%|█▌        | 27/177 [00:22<02:06,  1.19it/s]

 16%|█▌        | 28/177 [00:23<02:01,  1.22it/s]

 16%|█▋        | 29/177 [00:24<02:22,  1.04it/s]

 17%|█▋        | 30/177 [00:25<02:36,  1.06s/it]

 18%|█▊        | 31/177 [00:27<02:39,  1.09s/it]

 18%|█▊        | 32/177 [00:27<02:26,  1.01s/it]

 19%|█▊        | 33/177 [00:28<02:19,  1.03it/s]

 19%|█▉        | 34/177 [00:29<02:01,  1.18it/s]

 20%|█▉        | 35/177 [00:30<02:07,  1.11it/s]

 20%|██        | 36/177 [00:30<01:38,  1.43it/s]

 21%|██        | 37/177 [00:31<01:44,  1.34it/s]

 21%|██▏       | 38/177 [00:32<01:43,  1.34it/s]

 22%|██▏       | 39/177 [00:33<01:57,  1.17it/s]

 23%|██▎       | 40/177 [00:34<02:07,  1.07it/s]

 23%|██▎       | 41/177 [00:35<02:03,  1.10it/s]

 24%|██▎       | 42/177 [00:36<02:07,  1.06it/s]

 24%|██▍       | 43/177 [00:37<02:30,  1.12s/it]

 25%|██▍       | 44/177 [00:39<02:34,  1.16s/it]

 25%|██▌       | 45/177 [00:40<02:35,  1.18s/it]

 26%|██▌       | 46/177 [00:41<02:35,  1.19s/it]

 27%|██▋       | 47/177 [00:42<02:38,  1.22s/it]

 27%|██▋       | 48/177 [00:43<02:37,  1.22s/it]

 28%|██▊       | 49/177 [00:45<02:37,  1.23s/it]

 28%|██▊       | 50/177 [00:46<02:35,  1.22s/it]

 29%|██▉       | 51/177 [00:47<02:39,  1.26s/it]

 29%|██▉       | 52/177 [00:48<02:12,  1.06s/it]

 30%|██▉       | 53/177 [00:49<02:12,  1.07s/it]

 31%|███       | 54/177 [00:50<02:00,  1.02it/s]

 31%|███       | 55/177 [00:51<01:55,  1.05it/s]

 32%|███▏      | 56/177 [00:51<01:50,  1.10it/s]

 32%|███▏      | 57/177 [00:52<01:42,  1.17it/s]

 33%|███▎      | 58/177 [00:53<01:36,  1.23it/s]

 33%|███▎      | 59/177 [00:54<01:36,  1.22it/s]

 34%|███▍      | 60/177 [00:54<01:27,  1.34it/s]

 34%|███▍      | 61/177 [00:55<01:28,  1.31it/s]

 35%|███▌      | 62/177 [00:56<01:34,  1.22it/s]

 36%|███▌      | 63/177 [00:57<01:43,  1.10it/s]

 36%|███▌      | 64/177 [00:58<01:46,  1.06it/s]

 37%|███▋      | 65/177 [00:59<01:42,  1.09it/s]

 37%|███▋      | 66/177 [01:00<01:38,  1.13it/s]

 38%|███▊      | 67/177 [01:00<01:11,  1.54it/s]

 38%|███▊      | 68/177 [01:01<01:14,  1.47it/s]

 39%|███▉      | 69/177 [01:01<00:59,  1.80it/s]

 40%|████      | 71/177 [01:02<00:54,  1.94it/s]

 41%|████      | 72/177 [01:02<00:47,  2.19it/s]

 41%|████      | 73/177 [01:03<00:52,  1.97it/s]

 42%|████▏     | 74/177 [01:04<01:00,  1.72it/s]

 42%|████▏     | 75/177 [01:05<01:12,  1.40it/s]

 43%|████▎     | 76/177 [01:05<01:08,  1.47it/s]

 44%|████▎     | 77/177 [01:06<01:20,  1.24it/s]

 44%|████▍     | 78/177 [01:07<01:26,  1.15it/s]

 45%|████▍     | 79/177 [01:08<01:25,  1.15it/s]

 45%|████▌     | 80/177 [01:09<01:21,  1.19it/s]

 46%|████▌     | 81/177 [01:10<01:20,  1.19it/s]

 46%|████▋     | 82/177 [01:11<01:27,  1.08it/s]

 47%|████▋     | 83/177 [01:13<01:47,  1.14s/it]

 47%|████▋     | 84/177 [01:14<01:37,  1.05s/it]

 48%|████▊     | 85/177 [01:14<01:24,  1.09it/s]

 49%|████▊     | 86/177 [01:15<01:14,  1.23it/s]

 49%|████▉     | 87/177 [01:16<01:14,  1.21it/s]

 50%|████▉     | 88/177 [01:16<01:07,  1.32it/s]

 50%|█████     | 89/177 [01:17<01:02,  1.40it/s]

 51%|█████     | 90/177 [01:17<00:57,  1.51it/s]

 52%|█████▏    | 92/177 [01:18<00:50,  1.70it/s]

 53%|█████▎    | 93/177 [01:19<00:41,  2.00it/s]

 53%|█████▎    | 94/177 [01:20<00:55,  1.50it/s]

 54%|█████▎    | 95/177 [01:21<01:03,  1.30it/s]

 54%|█████▍    | 96/177 [01:22<01:10,  1.15it/s]

 55%|█████▍    | 97/177 [01:22<01:00,  1.33it/s]

 55%|█████▌    | 98/177 [01:23<01:04,  1.22it/s]

 56%|█████▌    | 99/177 [01:24<01:10,  1.10it/s]

 56%|█████▋    | 100/177 [01:25<01:01,  1.25it/s]

 57%|█████▋    | 101/177 [01:26<01:05,  1.15it/s]

 58%|█████▊    | 102/177 [01:27<01:11,  1.05it/s]

 58%|█████▊    | 103/177 [01:28<01:07,  1.09it/s]

 59%|█████▉    | 104/177 [01:29<01:07,  1.08it/s]

 59%|█████▉    | 105/177 [01:30<01:06,  1.09it/s]

 60%|█████▉    | 106/177 [01:31<01:03,  1.11it/s]

 60%|██████    | 107/177 [01:32<01:05,  1.08it/s]

 61%|██████    | 108/177 [01:32<01:01,  1.12it/s]

 62%|██████▏   | 109/177 [01:33<00:54,  1.24it/s]

 63%|██████▎   | 111/177 [01:34<00:42,  1.55it/s]

 63%|██████▎   | 112/177 [01:35<00:44,  1.47it/s]

 64%|██████▍   | 113/177 [01:35<00:39,  1.62it/s]

 64%|██████▍   | 114/177 [01:36<00:46,  1.35it/s]

 65%|██████▍   | 115/177 [01:37<00:46,  1.33it/s]

 66%|██████▌   | 116/177 [01:38<00:49,  1.24it/s]

 66%|██████▌   | 117/177 [01:39<00:45,  1.33it/s]

 67%|██████▋   | 118/177 [01:40<00:48,  1.21it/s]

 67%|██████▋   | 119/177 [01:41<00:55,  1.04it/s]

 68%|██████▊   | 120/177 [01:42<00:55,  1.02it/s]

 68%|██████▊   | 121/177 [01:42<00:40,  1.39it/s]

 69%|██████▉   | 122/177 [01:43<00:37,  1.47it/s]

 69%|██████▉   | 123/177 [01:43<00:37,  1.44it/s]

 70%|███████   | 124/177 [01:44<00:42,  1.24it/s]

 71%|███████   | 125/177 [01:46<00:50,  1.03it/s]

 71%|███████   | 126/177 [01:47<00:51,  1.00s/it]

 72%|███████▏  | 127/177 [01:47<00:41,  1.21it/s]

 72%|███████▏  | 128/177 [01:48<00:43,  1.12it/s]

 73%|███████▎  | 129/177 [01:49<00:42,  1.13it/s]

 73%|███████▎  | 130/177 [01:50<00:42,  1.10it/s]

 74%|███████▍  | 131/177 [01:51<00:42,  1.08it/s]

 75%|███████▍  | 132/177 [01:52<00:41,  1.07it/s]

 75%|███████▌  | 133/177 [01:52<00:32,  1.34it/s]

 76%|███████▌  | 134/177 [01:53<00:34,  1.23it/s]

 76%|███████▋  | 135/177 [01:54<00:27,  1.51it/s]

 77%|███████▋  | 137/177 [01:54<00:15,  2.56it/s]

 78%|███████▊  | 138/177 [01:54<00:17,  2.26it/s]

 79%|███████▊  | 139/177 [01:55<00:20,  1.84it/s]

 79%|███████▉  | 140/177 [01:56<00:22,  1.67it/s]

 80%|███████▉  | 141/177 [01:57<00:29,  1.24it/s]

 80%|████████  | 142/177 [01:58<00:31,  1.12it/s]

 81%|████████  | 143/177 [01:59<00:26,  1.26it/s]

 81%|████████▏ | 144/177 [02:00<00:27,  1.19it/s]

 82%|████████▏ | 145/177 [02:01<00:31,  1.02it/s]

 82%|████████▏ | 146/177 [02:02<00:30,  1.02it/s]

 83%|████████▎ | 147/177 [02:03<00:30,  1.01s/it]

 84%|████████▎ | 148/177 [02:04<00:25,  1.14it/s]

 84%|████████▍ | 149/177 [02:04<00:20,  1.37it/s]

 85%|████████▍ | 150/177 [02:05<00:18,  1.45it/s]

 85%|████████▌ | 151/177 [02:06<00:21,  1.21it/s]

 86%|████████▌ | 152/177 [02:07<00:22,  1.11it/s]

 86%|████████▋ | 153/177 [02:08<00:20,  1.16it/s]

 87%|████████▋ | 154/177 [02:08<00:15,  1.51it/s]

 88%|████████▊ | 155/177 [02:09<00:15,  1.39it/s]

 88%|████████▊ | 156/177 [02:10<00:15,  1.35it/s]

 89%|████████▊ | 157/177 [02:11<00:15,  1.31it/s]

 89%|████████▉ | 158/177 [02:11<00:14,  1.29it/s]

 90%|████████▉ | 159/177 [02:12<00:14,  1.28it/s]

 90%|█████████ | 160/177 [02:13<00:12,  1.38it/s]

 91%|█████████ | 161/177 [02:14<00:12,  1.33it/s]

 92%|█████████▏| 162/177 [02:14<00:10,  1.44it/s]

 92%|█████████▏| 163/177 [02:15<00:09,  1.43it/s]

 93%|█████████▎| 164/177 [02:16<00:09,  1.32it/s]

 93%|█████████▎| 165/177 [02:17<00:11,  1.09it/s]

 94%|█████████▍| 166/177 [02:18<00:10,  1.06it/s]

 94%|█████████▍| 167/177 [02:19<00:09,  1.02it/s]

 95%|█████████▍| 168/177 [02:20<00:09,  1.05s/it]

 95%|█████████▌| 169/177 [02:21<00:08,  1.09s/it]

 96%|█████████▌| 170/177 [02:23<00:07,  1.09s/it]

 97%|█████████▋| 171/177 [02:23<00:05,  1.16it/s]

 97%|█████████▋| 172/177 [02:23<00:03,  1.39it/s]

 98%|█████████▊| 173/177 [02:24<00:03,  1.30it/s]

 98%|█████████▊| 174/177 [02:25<00:02,  1.40it/s]

 99%|█████████▉| 176/177 [02:26<00:00,  1.54it/s]

100%|██████████| 177/177 [02:27<00:00,  1.34it/s]
                                                  

                                                 

  6%|▌         | 50/862 [21:36<5:12:02, 23.06s/it]

100%|██████████| 177/177 [02:27<00:00,  1.34it/s]

                                                 
  6%|▌         | 51/862 [21:54<15:11:06, 67.41s/it]
  6%|▌         | 52/862 [22:16<12:04:25, 53.66s/it]
  6%|▌         | 53/862 [22:41<10:06:34, 44.99s/it]
  6%|▋         | 54/862 [23:04<8:38:55, 38.53s/it] 
  6%|▋         | 55/862 [23:27<7:37:06, 33.99s/it]
                                                  

  6%|▋         | 55/862 [23:27<7:37:06, 33.99s/it]
  6%|▋         | 56/862 [23:48<6:43:25, 30.03s/it]
  7%|▋         | 57/862 [24:10<6:10:36, 27.62s/it]
  7%|▋         | 58/862 [24:30<5:38:49, 25.29s/it]
  7%|▋         | 59/862 [24:49<5:12:30, 23.35s/it]
  7%|▋         | 60/862 [25:10<5:02:52, 22.66s/it]
                                                  

  7%|▋         | 60/862 [25:10<5:02:52, 22.66s/it]
  7%|▋         | 61/862 [25:29<4:48:18, 21.60s/it]
  7%|▋         | 62/862 [25:49<4:41:25, 21.11s/it]
  7%|▋         | 63/862 [26:13<4:50:19, 21.80s/it]
  7%|▋         | 64/862 [26:34<4:48:05, 21.66s/it]
  8%|▊         | 65/862 [26:54<4:39:53, 21.07s/it]
                                                  

  8%|▊         | 65/862 [26:54<4:39:53, 21.07s/it]
  8%|▊         | 66/862 [27:09<4:17:15, 19.39s/it]
  8%|▊         | 67/862 [27:31<4:26:41, 20.13s/it]
  8%|▊         | 68/862 [27:51<4:24:46, 20.01s/it]
  8%|▊         | 69/862 [28:11<4:26:46, 20.18s/it]
  8%|▊         | 70/862 [28:26<4:06:32, 18.68s/it]
                                                  

  8%|▊         | 70/862 [28:26<4:06:32, 18.68s/it]
  8%|▊         | 71/862 [28:45<4:07:16, 18.76s/it]
  8%|▊         | 72/862 [29:08<4:21:11, 19.84s/it]
  8%|▊         | 73/862 [29:28<4:23:54, 20.07s/it]
  9%|▊         | 74/862 [29:45<4:10:43, 19.09s/it]
  9%|▊         | 75/862 [30:04<4:11:16, 19.16s/it]
                                                  

  9%|▊         | 75/862 [30:04<4:11:16, 19.16s/it]
  9%|▉         | 76/862 [30:32<4:43:28, 21.64s/it]
  9%|▉         | 77/862 [30:56<4:51:52, 22.31s/it]
  9%|▉         | 78/862 [31:15<4:39:30, 21.39s/it]
  9%|▉         | 79/862 [31:39<4:51:33, 22.34s/it]
  9%|▉         | 80/862 [31:58<4:36:42, 21.23s/it]
                                                  

  9%|▉         | 80/862 [31:58<4:36:42, 21.23s/it]
  9%|▉         | 81/862 [32:15<4:18:42, 19.88s/it]
 10%|▉         | 82/862 [32:39<4:35:19, 21.18s/it]
 10%|▉         | 83/862 [32:56<4:19:50, 20.01s/it]
 10%|▉         | 84/862 [33:13<4:06:52, 19.04s/it]
 10%|▉         | 85/862 [33:31<4:03:53, 18.83s/it]
                                                  

 10%|▉         | 85/862 [33:31<4:03:53, 18.83s/it]
 10%|▉         | 86/862 [33:49<3:57:15, 18.35s/it]
 10%|█         | 87/862 [34:03<3:40:22, 17.06s/it]
 10%|█         | 88/862 [34:23<3:51:10, 17.92s/it]
 10%|█         | 89/862 [34:48<4:20:04, 20.19s/it]
 10%|█         | 90/862 [35:11<4:29:48, 20.97s/it]
                                                  

 10%|█         | 90/862 [35:11<4:29:48, 20.97s/it]
 11%|█         | 91/862 [35:33<4:32:16, 21.19s/it]
 11%|█         | 92/862 [35:52<4:23:15, 20.51s/it]
 11%|█         | 93/862 [36:14<4:31:28, 21.18s/it]
 11%|█         | 94/862 [36:33<4:19:49, 20.30s/it]
 11%|█         | 95/862 [36:56<4:31:41, 21.25s/it]
                                                  

 11%|█         | 95/862 [36:56<4:31:41, 21.25s/it]
 11%|█         | 96/862 [37:19<4:37:46, 21.76s/it]
 11%|█▏        | 97/862 [37:36<4:18:24, 20.27s/it]
 11%|█▏        | 98/862 [37:56<4:17:27, 20.22s/it]
 11%|█▏        | 99/862 [38:12<4:03:00, 19.11s/it]
 12%|█▏        | 100/862 [38:33<4:09:12, 19.62s/it]
                                                   

 12%|█▏        | 100/862 [38:33<4:09:12, 19.62s/it]{'eval_loss': 0.2423536628484726, 'eval_runtime': 152.5048, 'eval_samples_per_second': 2.321, 'eval_steps_per_second': 1.161, 'epoch': 0.06}
{'loss': 0.3098, 'grad_norm': 1.3211257457733154, 'learning_rate': 0.0001999262588748966, 'epoch': 0.06}
{'loss': 0.27, 'grad_norm': 1.0737800598144531, 'learning_rate': 0.00019983410796018032, 'epoch': 0.07}
{'loss': 0.2597, 'grad_norm': 1.062996745109558, 'learning_rate': 0.00019970514425465705, 'epoch': 0.08}
{'loss': 0.2213, 'grad_norm': 0.8997095227241516, 'learning_rate': 0.00019953941531235424, 'epoch': 0.08}
{'loss': 0.2616, 'grad_norm': 0.6930145025253296, 'learning_rate': 0.00019933698224409876, 'epoch': 0.09}
{'loss': 0.2218, 'grad_norm': 0.5634636282920837, 'learning_rate': 0.0001990979196949828, 'epoch': 0.09}
{'loss': 0.238, 'grad_norm': 0.545459508895874, 'learning_rate': 0.00019882231581683937, 'epoch': 0.1}
{'loss': 0.2145, 'grad_norm': 0.4912775456905365, 'learning_rate': 0.0001985102722357371, 'epoch': 0.1}
{'loss': 0.2336, 'grad_norm': 0.4108012914657593, 'learning_rate': 0.0001981619040145068, 'epoch': 0.11}
{'loss': 0.1963, 'grad_norm': 0.521858274936676, 'learning_rate': 0.00019777733961031326, 'epoch': 0.12}


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:00<00:49,  3.52it/s]

  2%|▏         | 3/177 [00:01<01:19,  2.18it/s]

  2%|▏         | 4/177 [00:01<01:08,  2.53it/s]

  3%|▎         | 5/177 [00:01<00:50,  3.42it/s]

  3%|▎         | 6/177 [00:01<00:50,  3.37it/s]

  4%|▍         | 7/177 [00:02<00:52,  3.24it/s]

  5%|▍         | 8/177 [00:02<00:52,  3.19it/s]

  5%|▌         | 9/177 [00:02<00:55,  3.02it/s]

  6%|▌         | 10/177 [00:03<00:56,  2.95it/s]

  6%|▌         | 11/177 [00:03<01:06,  2.50it/s]

  7%|▋         | 12/177 [00:04<00:57,  2.85it/s]

  7%|▋         | 13/177 [00:04<00:59,  2.76it/s]

  8%|▊         | 14/177 [00:04<00:56,  2.89it/s]

  8%|▊         | 15/177 [00:04<00:44,  3.65it/s]

  9%|▉         | 16/177 [00:05<00:43,  3.70it/s]

 10%|▉         | 17/177 [00:05<00:44,  3.62it/s]

 11%|█         | 19/177 [00:05<00:37,  4.22it/s]

 11%|█▏        | 20/177 [00:06<00:38,  4.04it/s]

 12%|█▏        | 21/177 [00:06<00:46,  3.36it/s]

 12%|█▏        | 22/177 [00:07<01:01,  2.53it/s]

 13%|█▎        | 23/177 [00:07<01:15,  2.05it/s]

 14%|█▎        | 24/177 [00:08<01:14,  2.06it/s]

 14%|█▍        | 25/177 [00:08<01:05,  2.33it/s]

 15%|█▍        | 26/177 [00:09<00:57,  2.62it/s]

 15%|█▌        | 27/177 [00:09<00:53,  2.81it/s]

 16%|█▌        | 28/177 [00:09<00:48,  3.06it/s]

 16%|█▋        | 29/177 [00:10<00:58,  2.53it/s]

 17%|█▋        | 30/177 [00:10<01:05,  2.24it/s]

 18%|█▊        | 31/177 [00:11<01:14,  1.96it/s]

 18%|█▊        | 32/177 [00:11<01:05,  2.20it/s]

 19%|█▊        | 33/177 [00:11<00:51,  2.80it/s]

 20%|█▉        | 35/177 [00:12<00:38,  3.65it/s]

 20%|██        | 36/177 [00:12<00:37,  3.79it/s]

 21%|██        | 37/177 [00:12<00:31,  4.38it/s]

 21%|██▏       | 38/177 [00:12<00:32,  4.31it/s]

 22%|██▏       | 39/177 [00:13<00:46,  2.99it/s]

 23%|██▎       | 40/177 [00:13<00:47,  2.88it/s]

 23%|██▎       | 41/177 [00:14<00:47,  2.87it/s]

 24%|██▎       | 42/177 [00:14<00:53,  2.50it/s]

 24%|██▍       | 43/177 [00:15<01:01,  2.19it/s]

 25%|██▍       | 44/177 [00:15<00:59,  2.23it/s]

 25%|██▌       | 45/177 [00:16<01:00,  2.17it/s]

 26%|██▌       | 46/177 [00:16<01:06,  1.98it/s]

 27%|██▋       | 47/177 [00:17<01:03,  2.05it/s]

 27%|██▋       | 48/177 [00:17<01:03,  2.04it/s]

 28%|██▊       | 49/177 [00:18<01:04,  1.97it/s]

 28%|██▊       | 50/177 [00:18<01:02,  2.02it/s]

 29%|██▉       | 51/177 [00:19<01:08,  1.85it/s]

 29%|██▉       | 52/177 [00:19<01:09,  1.80it/s]

 30%|██▉       | 53/177 [00:20<01:10,  1.77it/s]

 31%|███       | 54/177 [00:20<00:58,  2.11it/s]

 31%|███       | 55/177 [00:21<00:54,  2.26it/s]

 32%|███▏      | 56/177 [00:21<00:49,  2.47it/s]

 32%|███▏      | 57/177 [00:21<00:42,  2.84it/s]

 33%|███▎      | 58/177 [00:21<00:36,  3.28it/s]

 33%|███▎      | 59/177 [00:21<00:28,  4.08it/s]

 34%|███▍      | 61/177 [00:22<00:18,  6.20it/s]

 35%|███▌      | 62/177 [00:22<00:20,  5.54it/s]

 36%|███▌      | 63/177 [00:22<00:32,  3.53it/s]

 36%|███▌      | 64/177 [00:23<00:40,  2.82it/s]

 37%|███▋      | 65/177 [00:23<00:39,  2.85it/s]

 38%|███▊      | 67/177 [00:23<00:25,  4.30it/s]

 38%|███▊      | 68/177 [00:24<00:25,  4.26it/s]

 39%|███▉      | 69/177 [00:24<00:26,  4.11it/s]

 40%|████      | 71/177 [00:24<00:24,  4.32it/s]

 41%|████      | 72/177 [00:25<00:25,  4.11it/s]

 41%|████      | 73/177 [00:25<00:22,  4.61it/s]

 42%|████▏     | 74/177 [00:25<00:24,  4.27it/s]

 42%|████▏     | 75/177 [00:25<00:26,  3.90it/s]

 44%|████▎     | 77/177 [00:26<00:25,  4.00it/s]

 44%|████▍     | 78/177 [00:26<00:31,  3.14it/s]

 45%|████▍     | 79/177 [00:27<00:32,  3.04it/s]

 45%|████▌     | 80/177 [00:27<00:30,  3.19it/s]

 46%|████▌     | 81/177 [00:27<00:30,  3.18it/s]

 46%|████▋     | 82/177 [00:28<00:30,  3.12it/s]

 47%|████▋     | 83/177 [00:28<00:38,  2.44it/s]

 47%|████▋     | 84/177 [00:29<00:35,  2.59it/s]

 49%|████▊     | 86/177 [00:29<00:22,  4.07it/s]

 50%|████▉     | 88/177 [00:29<00:16,  5.36it/s]

 50%|█████     | 89/177 [00:29<00:14,  5.93it/s]

 51%|█████▏    | 91/177 [00:29<00:10,  7.88it/s]

 53%|█████▎    | 93/177 [00:30<00:13,  6.25it/s]

 53%|█████▎    | 94/177 [00:30<00:17,  4.71it/s]

 54%|█████▎    | 95/177 [00:31<00:23,  3.47it/s]

 54%|█████▍    | 96/177 [00:31<00:25,  3.18it/s]

 55%|█████▍    | 97/177 [00:32<00:28,  2.84it/s]

 55%|█████▌    | 98/177 [00:32<00:30,  2.58it/s]

 56%|█████▌    | 99/177 [00:33<00:35,  2.22it/s]

 56%|█████▋    | 100/177 [00:33<00:37,  2.04it/s]

 57%|█████▋    | 101/177 [00:34<00:37,  2.02it/s]

 58%|█████▊    | 102/177 [00:34<00:35,  2.13it/s]

 58%|█████▊    | 103/177 [00:34<00:31,  2.34it/s]

 59%|█████▉    | 104/177 [00:35<00:26,  2.73it/s]

 59%|█████▉    | 105/177 [00:35<00:27,  2.66it/s]

 60%|█████▉    | 106/177 [00:35<00:26,  2.71it/s]

 60%|██████    | 107/177 [00:36<00:23,  2.95it/s]

 62%|██████▏   | 109/177 [00:36<00:14,  4.60it/s]

 63%|██████▎   | 111/177 [00:36<00:14,  4.71it/s]

 63%|██████▎   | 112/177 [00:37<00:14,  4.37it/s]

 64%|██████▍   | 113/177 [00:37<00:17,  3.61it/s]

 64%|██████▍   | 114/177 [00:37<00:18,  3.39it/s]

 65%|██████▍   | 115/177 [00:38<00:18,  3.44it/s]

 66%|██████▌   | 116/177 [00:38<00:16,  3.66it/s]

 66%|██████▌   | 117/177 [00:38<00:13,  4.37it/s]

 67%|██████▋   | 118/177 [00:38<00:14,  4.07it/s]

 67%|██████▋   | 119/177 [00:39<00:19,  2.97it/s]

 68%|██████▊   | 120/177 [00:39<00:18,  3.10it/s]

 68%|██████▊   | 121/177 [00:39<00:14,  3.85it/s]

 69%|██████▉   | 123/177 [00:39<00:11,  4.78it/s]

 70%|███████   | 124/177 [00:40<00:12,  4.13it/s]

 71%|███████   | 125/177 [00:40<00:17,  3.02it/s]

 71%|███████   | 126/177 [00:41<00:16,  3.00it/s]

 72%|███████▏  | 127/177 [00:41<00:17,  2.79it/s]

 72%|███████▏  | 128/177 [00:42<00:17,  2.88it/s]

 73%|███████▎  | 129/177 [00:42<00:16,  2.85it/s]

 73%|███████▎  | 130/177 [00:42<00:18,  2.58it/s]

 74%|███████▍  | 131/177 [00:43<00:18,  2.42it/s]

 75%|███████▍  | 132/177 [00:43<00:15,  2.86it/s]

 75%|███████▌  | 133/177 [00:43<00:14,  2.93it/s]

 76%|███████▌  | 134/177 [00:44<00:16,  2.65it/s]

 76%|███████▋  | 135/177 [00:44<00:15,  2.74it/s]

 77%|███████▋  | 137/177 [00:44<00:09,  4.30it/s]

 79%|███████▊  | 139/177 [00:44<00:06,  5.70it/s]

 79%|███████▉  | 140/177 [00:45<00:07,  5.22it/s]

 80%|███████▉  | 141/177 [00:45<00:09,  3.71it/s]

 80%|████████  | 142/177 [00:46<00:11,  3.17it/s]

 81%|████████  | 143/177 [00:46<00:12,  2.65it/s]

 81%|████████▏ | 144/177 [00:47<00:13,  2.51it/s]

 82%|████████▏ | 145/177 [00:47<00:14,  2.20it/s]

 82%|████████▏ | 146/177 [00:48<00:14,  2.15it/s]

 83%|████████▎ | 147/177 [00:48<00:15,  1.91it/s]

 84%|████████▎ | 148/177 [00:49<00:15,  1.88it/s]

 84%|████████▍ | 149/177 [00:49<00:13,  2.04it/s]

 85%|████████▌ | 151/177 [00:50<00:09,  2.63it/s]

 86%|████████▌ | 152/177 [00:50<00:09,  2.68it/s]

 86%|████████▋ | 153/177 [00:50<00:08,  2.90it/s]

 87%|████████▋ | 154/177 [00:51<00:06,  3.29it/s]

 88%|████████▊ | 155/177 [00:51<00:05,  3.94it/s]

 89%|████████▊ | 157/177 [00:51<00:03,  5.65it/s]

 89%|████████▉ | 158/177 [00:51<00:03,  4.86it/s]

 90%|████████▉ | 159/177 [00:52<00:04,  4.38it/s]

 91%|█████████ | 161/177 [00:52<00:02,  6.02it/s]

 92%|█████████▏| 163/177 [00:52<00:02,  6.46it/s]

 93%|█████████▎| 164/177 [00:52<00:02,  4.92it/s]

 93%|█████████▎| 165/177 [00:53<00:03,  3.46it/s]

 94%|█████████▍| 166/177 [00:53<00:03,  2.92it/s]

 94%|█████████▍| 167/177 [00:54<00:03,  2.92it/s]

 95%|█████████▍| 168/177 [00:54<00:03,  2.71it/s]

 95%|█████████▌| 169/177 [00:55<00:03,  2.62it/s]

 96%|█████████▌| 170/177 [00:55<00:02,  2.78it/s]

 97%|█████████▋| 171/177 [00:55<00:02,  2.85it/s]

 97%|█████████▋| 172/177 [00:56<00:01,  2.78it/s]

 98%|█████████▊| 173/177 [00:56<00:01,  2.73it/s]

 99%|█████████▉| 175/177 [00:56<00:00,  4.20it/s]

 99%|█████████▉| 176/177 [00:57<00:00,  3.90it/s]

100%|██████████| 177/177 [00:57<00:00,  3.63it/s]
                                                   

                                                 

 12%|█▏        | 100/862 [39:31<4:09:12, 19.62s/it]

100%|██████████| 177/177 [00:57<00:00,  3.63it/s]

                                                 wandb: WARNING URL not available in offline run
`rope_scaling`'s beta_fast field must be a float, got 32
`rope_scaling`'s beta_slow field must be a float, got 1

 12%|█▏        | 101/862 [39:59<8:18:49, 39.33s/it]
 12%|█▏        | 102/862 [40:17<7:00:20, 33.19s/it]
 12%|█▏        | 103/862 [40:37<6:09:28, 29.21s/it]
 12%|█▏        | 104/862 [41:38<8:07:17, 38.57s/it]
 12%|█▏        | 105/862 [42:07<7:31:11, 35.76s/it]
                                                   

 12%|█▏        | 105/862 [42:07<7:31:11, 35.76s/it]
 12%|█▏        | 106/862 [42:25<6:23:19, 30.42s/it]
 12%|█▏        | 107/862 [42:48<5:55:01, 28.21s/it]
 13%|█▎        | 108/862 [43:01<4:58:43, 23.77s/it]
 13%|█▎        | 109/862 [43:15<4:19:05, 20.64s/it]
 13%|█▎        | 110/862 [43:28<3:51:39, 18.48s/it]
                                                   

 13%|█▎        | 110/862 [43:28<3:51:39, 18.48s/it]
 13%|█▎        | 111/862 [43:46<3:48:38, 18.27s/it]
 13%|█▎        | 112/862 [44:07<3:59:49, 19.19s/it]
 13%|█▎        | 113/862 [44:36<4:36:45, 22.17s/it]
 13%|█▎        | 114/862 [44:54<4:19:00, 20.78s/it]
 13%|█▎        | 115/862 [45:10<4:00:31, 19.32s/it]
                                                   

 13%|█▎        | 115/862 [45:10<4:00:31, 19.32s/it]
 13%|█▎        | 116/862 [45:23<3:37:13, 17.47s/it]
 14%|█▎        | 117/862 [45:45<3:52:11, 18.70s/it]
 14%|█▎        | 118/862 [46:03<3:50:20, 18.58s/it]
 14%|█▍        | 119/862 [46:19<3:39:45, 17.75s/it]
 14%|█▍        | 120/862 [46:39<3:49:27, 18.55s/it]
                                                   

 14%|█▍        | 120/862 [46:39<3:49:27, 18.55s/it]
 14%|█▍        | 121/862 [46:58<3:50:50, 18.69s/it]
 14%|█▍        | 122/862 [47:15<3:43:05, 18.09s/it]
 14%|█▍        | 123/862 [47:28<3:23:47, 16.55s/it]
 14%|█▍        | 124/862 [47:50<3:43:51, 18.20s/it]
 15%|█▍        | 125/862 [48:12<4:00:07, 19.55s/it]
                                                   

 15%|█▍        | 125/862 [48:12<4:00:07, 19.55s/it]
 15%|█▍        | 126/862 [48:30<3:52:26, 18.95s/it]
 15%|█▍        | 127/862 [48:50<3:54:34, 19.15s/it]
 15%|█▍        | 128/862 [49:06<3:42:15, 18.17s/it]
 15%|█▍        | 129/862 [49:26<3:48:45, 18.72s/it]
 15%|█▌        | 130/862 [49:45<3:52:27, 19.05s/it]
                                                   

 15%|█▌        | 130/862 [49:45<3:52:27, 19.05s/it]
 15%|█▌        | 131/862 [50:06<3:58:18, 19.56s/it]
 15%|█▌        | 132/862 [50:24<3:50:09, 18.92s/it]
 15%|█▌        | 133/862 [50:37<3:30:08, 17.30s/it]
 16%|█▌        | 134/862 [50:59<3:48:02, 18.79s/it]
 16%|█▌        | 135/862 [51:15<3:34:54, 17.74s/it]
                                                   

 16%|█▌        | 135/862 [51:15<3:34:54, 17.74s/it]
 16%|█▌        | 136/862 [51:34<3:40:21, 18.21s/it]
 16%|█▌        | 137/862 [51:51<3:34:13, 17.73s/it]
 16%|█▌        | 138/862 [52:08<3:33:55, 17.73s/it]
 16%|█▌        | 139/862 [52:29<3:42:57, 18.50s/it]
 16%|█▌        | 140/862 [52:47<3:44:12, 18.63s/it]
                                                   

 16%|█▌        | 140/862 [52:47<3:44:12, 18.63s/it]
 16%|█▋        | 141/862 [53:09<3:54:57, 19.55s/it]
 16%|█▋        | 142/862 [53:36<4:20:01, 21.67s/it]
 17%|█▋        | 143/862 [53:51<3:56:47, 19.76s/it]
 17%|█▋        | 144/862 [54:07<3:43:33, 18.68s/it]
 17%|█▋        | 145/862 [54:30<3:56:02, 19.75s/it]
                                                   

 17%|█▋        | 145/862 [54:30<3:56:02, 19.75s/it]
 17%|█▋        | 146/862 [54:44<3:35:18, 18.04s/it]
 17%|█▋        | 147/862 [55:03<3:38:32, 18.34s/it]
 17%|█▋        | 148/862 [55:22<3:40:31, 18.53s/it]
 17%|█▋        | 149/862 [55:40<3:39:20, 18.46s/it]
 17%|█▋        | 150/862 [56:03<3:55:15, 19.83s/it]
                                                   

 17%|█▋        | 150/862 [56:03<3:55:15, 19.83s/it]{'eval_loss': 0.1702147275209427, 'eval_runtime': 57.6184, 'eval_samples_per_second': 6.144, 'eval_steps_per_second': 3.072, 'epoch': 0.12}
{'loss': 0.2433, 'grad_norm': 0.5323580503463745, 'learning_rate': 0.00019735672082728782, 'epoch': 0.12}
{'loss': 0.2063, 'grad_norm': 0.5639608502388, 'learning_rate': 0.0001969002027642398, 'epoch': 0.13}
{'loss': 0.2208, 'grad_norm': 0.583233654499054, 'learning_rate': 0.00019640795375746518, 'epoch': 0.13}
{'loss': 0.204, 'grad_norm': 0.5018272399902344, 'learning_rate': 0.00019588015531867464, 'epoch': 0.14}
{'loss': 0.2155, 'grad_norm': 0.48238405585289, 'learning_rate': 0.00019531700206806274, 'epoch': 0.15}
{'loss': 0.208, 'grad_norm': 0.4539037048816681, 'learning_rate': 0.00019471870166254377, 'epoch': 0.15}
{'loss': 0.2292, 'grad_norm': 0.5225255489349365, 'learning_rate': 0.00019408547471918061, 'epoch': 0.16}
{'loss': 0.2074, 'grad_norm': 0.4645092785358429, 'learning_rate': 0.00019341755473383432, 'epoch': 0.16}
{'loss': 0.2301, 'grad_norm': 0.47805657982826233, 'learning_rate': 0.00019271518799506492, 'epoch': 0.17}
{'loss': 0.1916, 'grad_norm': 0.43033567070961, 'learning_rate': 0.00019197863349331522, 'epoch': 0.17}


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:00<00:50,  3.50it/s]

  2%|▏         | 3/177 [00:01<01:20,  2.17it/s]

  2%|▏         | 4/177 [00:01<01:08,  2.52it/s]

  3%|▎         | 5/177 [00:01<00:50,  3.39it/s]

  3%|▎         | 6/177 [00:01<00:51,  3.35it/s]

  4%|▍         | 7/177 [00:02<00:52,  3.22it/s]

  5%|▍         | 8/177 [00:02<00:53,  3.18it/s]

  5%|▌         | 9/177 [00:03<00:55,  3.01it/s]

  6%|▌         | 10/177 [00:03<00:56,  2.93it/s]

  6%|▌         | 11/177 [00:03<01:06,  2.50it/s]

  7%|▋         | 12/177 [00:04<00:57,  2.86it/s]

  7%|▋         | 13/177 [00:04<00:59,  2.75it/s]

  8%|▊         | 14/177 [00:04<00:56,  2.88it/s]

  8%|▊         | 15/177 [00:04<00:45,  3.55it/s]

  9%|▉         | 16/177 [00:05<00:44,  3.63it/s]

 10%|▉         | 17/177 [00:05<00:44,  3.57it/s]

 11%|█         | 19/177 [00:05<00:37,  4.19it/s]

 11%|█▏        | 20/177 [00:06<00:39,  4.02it/s]

 12%|█▏        | 21/177 [00:06<00:46,  3.36it/s]

 12%|█▏        | 22/177 [00:07<01:01,  2.51it/s]

 13%|█▎        | 23/177 [00:08<01:15,  2.04it/s]

 14%|█▎        | 24/177 [00:08<01:14,  2.05it/s]

 14%|█▍        | 25/177 [00:08<01:05,  2.31it/s]

 15%|█▍        | 26/177 [00:09<00:57,  2.60it/s]

 15%|█▌        | 27/177 [00:09<00:53,  2.79it/s]

 16%|█▌        | 28/177 [00:09<00:48,  3.06it/s]

 16%|█▋        | 29/177 [00:10<00:58,  2.53it/s]

 17%|█▋        | 30/177 [00:10<01:05,  2.24it/s]

 18%|█▊        | 31/177 [00:11<01:14,  1.96it/s]

 18%|█▊        | 32/177 [00:11<01:05,  2.20it/s]

 19%|█▊        | 33/177 [00:11<00:51,  2.81it/s]

 20%|█▉        | 35/177 [00:12<00:38,  3.65it/s]

 20%|██        | 36/177 [00:12<00:37,  3.80it/s]

 21%|██        | 37/177 [00:12<00:31,  4.40it/s]

 21%|██▏       | 38/177 [00:12<00:32,  4.34it/s]

 22%|██▏       | 39/177 [00:13<00:46,  2.99it/s]

 23%|██▎       | 40/177 [00:13<00:47,  2.89it/s]

 23%|██▎       | 41/177 [00:14<00:47,  2.87it/s]

 24%|██▎       | 42/177 [00:14<00:53,  2.50it/s]

 24%|██▍       | 43/177 [00:15<01:01,  2.19it/s]

 25%|██▍       | 44/177 [00:15<00:59,  2.23it/s]

 25%|██▌       | 45/177 [00:16<01:01,  2.16it/s]

 26%|██▌       | 46/177 [00:16<01:01,  2.12it/s]

 27%|██▋       | 47/177 [00:17<01:06,  1.97it/s]

 27%|██▋       | 48/177 [00:17<01:05,  1.97it/s]

 28%|██▊       | 49/177 [00:18<01:06,  1.92it/s]

 28%|██▊       | 50/177 [00:18<01:04,  1.97it/s]

 29%|██▉       | 51/177 [00:19<01:09,  1.82it/s]

 29%|██▉       | 52/177 [00:20<01:10,  1.78it/s]

 30%|██▉       | 53/177 [00:20<01:10,  1.75it/s]

 31%|███       | 54/177 [00:20<00:58,  2.09it/s]

 31%|███       | 55/177 [00:21<00:54,  2.25it/s]

 32%|███▏      | 56/177 [00:21<00:49,  2.47it/s]

 32%|███▏      | 57/177 [00:21<00:42,  2.84it/s]

 33%|███▎      | 58/177 [00:21<00:36,  3.28it/s]

 33%|███▎      | 59/177 [00:22<00:28,  4.09it/s]

 34%|███▍      | 61/177 [00:22<00:18,  6.22it/s]

 35%|███▌      | 62/177 [00:22<00:20,  5.56it/s]

 36%|███▌      | 63/177 [00:23<00:32,  3.54it/s]

 36%|███▌      | 64/177 [00:23<00:39,  2.83it/s]

 37%|███▋      | 65/177 [00:23<00:39,  2.85it/s]

 38%|███▊      | 67/177 [00:24<00:25,  4.32it/s]

 38%|███▊      | 68/177 [00:24<00:25,  4.29it/s]

 39%|███▉      | 69/177 [00:24<00:26,  4.13it/s]

 40%|████      | 71/177 [00:25<00:24,  4.32it/s]

 41%|████      | 72/177 [00:25<00:25,  4.10it/s]

 41%|████      | 73/177 [00:25<00:22,  4.61it/s]

 42%|████▏     | 74/177 [00:25<00:24,  4.26it/s]

 42%|████▏     | 75/177 [00:26<00:26,  3.90it/s]

 44%|████▎     | 77/177 [00:26<00:24,  4.01it/s]

 44%|████▍     | 78/177 [00:27<00:31,  3.14it/s]

 45%|████▍     | 79/177 [00:27<00:32,  3.03it/s]

 45%|████▌     | 80/177 [00:27<00:30,  3.19it/s]

 46%|████▌     | 81/177 [00:28<00:30,  3.18it/s]

 46%|████▋     | 82/177 [00:28<00:30,  3.12it/s]

 47%|████▋     | 83/177 [00:28<00:38,  2.43it/s]

 47%|████▋     | 84/177 [00:29<00:35,  2.59it/s]

 49%|████▊     | 86/177 [00:29<00:22,  4.06it/s]

 50%|████▉     | 88/177 [00:29<00:16,  5.36it/s]

 50%|█████     | 89/177 [00:29<00:14,  5.93it/s]

 51%|█████▏    | 91/177 [00:29<00:10,  7.86it/s]

 53%|█████▎    | 93/177 [00:30<00:13,  6.22it/s]

 53%|█████▎    | 94/177 [00:30<00:17,  4.70it/s]

 54%|█████▎    | 95/177 [00:31<00:23,  3.47it/s]

 54%|█████▍    | 96/177 [00:31<00:25,  3.16it/s]

 55%|█████▍    | 97/177 [00:32<00:32,  2.44it/s]

 55%|█████▌    | 98/177 [00:32<00:30,  2.58it/s]

 56%|█████▌    | 99/177 [00:33<00:34,  2.23it/s]

 56%|█████▋    | 100/177 [00:33<00:36,  2.11it/s]

 57%|█████▋    | 101/177 [00:34<00:37,  2.02it/s]

 58%|█████▊    | 102/177 [00:34<00:35,  2.11it/s]

 58%|█████▊    | 103/177 [00:35<00:31,  2.32it/s]

 59%|█████▉    | 104/177 [00:35<00:26,  2.72it/s]

 59%|█████▉    | 105/177 [00:35<00:27,  2.65it/s]

 60%|█████▉    | 106/177 [00:36<00:26,  2.70it/s]

 60%|██████    | 107/177 [00:36<00:23,  2.93it/s]

 62%|██████▏   | 109/177 [00:36<00:14,  4.57it/s]

 63%|██████▎   | 111/177 [00:36<00:14,  4.68it/s]

 63%|██████▎   | 112/177 [00:37<00:14,  4.35it/s]

 64%|██████▍   | 113/177 [00:37<00:17,  3.58it/s]

 64%|██████▍   | 114/177 [00:38<00:18,  3.38it/s]

 65%|██████▍   | 115/177 [00:38<00:18,  3.42it/s]

 66%|██████▌   | 116/177 [00:38<00:16,  3.66it/s]

 66%|██████▌   | 117/177 [00:38<00:13,  4.39it/s]

 67%|██████▋   | 118/177 [00:38<00:14,  4.09it/s]

 67%|██████▋   | 119/177 [00:39<00:19,  2.97it/s]

 68%|██████▊   | 120/177 [00:39<00:18,  3.09it/s]

 68%|██████▊   | 121/177 [00:39<00:14,  3.83it/s]

 69%|██████▉   | 123/177 [00:40<00:11,  4.75it/s]

 70%|███████   | 124/177 [00:40<00:12,  4.10it/s]

 71%|███████   | 125/177 [00:41<00:17,  2.99it/s]

 71%|███████   | 126/177 [00:41<00:17,  2.98it/s]

 72%|███████▏  | 127/177 [00:41<00:18,  2.77it/s]

 72%|███████▏  | 128/177 [00:42<00:17,  2.87it/s]

 73%|███████▎  | 129/177 [00:42<00:16,  2.85it/s]

 73%|███████▎  | 130/177 [00:43<00:18,  2.59it/s]

 74%|███████▍  | 131/177 [00:43<00:19,  2.42it/s]

 75%|███████▍  | 132/177 [00:43<00:15,  2.85it/s]

 75%|███████▌  | 133/177 [00:44<00:15,  2.93it/s]

 76%|███████▌  | 134/177 [00:44<00:16,  2.62it/s]

 76%|███████▋  | 135/177 [00:44<00:15,  2.73it/s]

 77%|███████▋  | 137/177 [00:44<00:09,  4.29it/s]

 79%|███████▊  | 139/177 [00:45<00:06,  5.68it/s]

 79%|███████▉  | 140/177 [00:45<00:07,  5.23it/s]

 80%|███████▉  | 141/177 [00:45<00:09,  3.71it/s]

 80%|████████  | 142/177 [00:46<00:11,  3.17it/s]

 81%|████████  | 143/177 [00:46<00:12,  2.65it/s]

 81%|████████▏ | 144/177 [00:47<00:13,  2.51it/s]

 82%|████████▏ | 145/177 [00:47<00:14,  2.20it/s]

 82%|████████▏ | 146/177 [00:48<00:14,  2.15it/s]

 83%|████████▎ | 147/177 [00:49<00:14,  2.00it/s]

 84%|████████▎ | 148/177 [00:49<00:14,  1.94it/s]

 84%|████████▍ | 149/177 [00:50<00:13,  2.08it/s]

 85%|████████▌ | 151/177 [00:50<00:09,  2.66it/s]

 86%|████████▌ | 152/177 [00:50<00:09,  2.70it/s]

 86%|████████▋ | 153/177 [00:51<00:08,  2.90it/s]

 87%|████████▋ | 154/177 [00:51<00:07,  3.27it/s]

 88%|████████▊ | 155/177 [00:51<00:05,  3.84it/s]

 89%|████████▊ | 157/177 [00:51<00:03,  5.36it/s]

 89%|████████▉ | 158/177 [00:51<00:04,  4.74it/s]

 90%|████████▉ | 159/177 [00:52<00:04,  4.16it/s]

 91%|█████████ | 161/177 [00:52<00:02,  5.62it/s]

 92%|█████████▏| 163/177 [00:52<00:02,  5.90it/s]

 93%|█████████▎| 164/177 [00:53<00:02,  4.66it/s]

 93%|█████████▎| 165/177 [00:53<00:03,  3.30it/s]

 94%|█████████▍| 166/177 [00:54<00:03,  2.85it/s]

 94%|█████████▍| 167/177 [00:54<00:03,  2.85it/s]

 95%|█████████▍| 168/177 [00:55<00:03,  2.66it/s]

 95%|█████████▌| 169/177 [00:55<00:03,  2.58it/s]

 96%|█████████▌| 170/177 [00:55<00:02,  2.75it/s]

 97%|█████████▋| 171/177 [00:56<00:02,  2.82it/s]

 97%|█████████▋| 172/177 [00:56<00:01,  2.76it/s]

 98%|█████████▊| 173/177 [00:56<00:01,  2.72it/s]

 99%|█████████▉| 175/177 [00:57<00:00,  4.18it/s]

 99%|█████████▉| 176/177 [00:57<00:00,  3.88it/s]

100%|██████████| 177/177 [00:57<00:00,  3.64it/s]
                                                   

                                                 

 17%|█▋        | 150/862 [57:01<3:55:15, 19.83s/it]

100%|██████████| 177/177 [00:57<00:00,  3.64it/s]

                                                 
 18%|█▊        | 151/862 [57:22<7:27:21, 37.75s/it]
 18%|█▊        | 152/862 [57:42<6:22:24, 32.32s/it]
 18%|█▊        | 153/862 [58:06<5:50:44, 29.68s/it]
 18%|█▊        | 154/862 [58:26<5:19:01, 27.04s/it]
 18%|█▊        | 155/862 [58:44<4:43:55, 24.09s/it]
                                                   

 18%|█▊        | 155/862 [58:44<4:43:55, 24.09s/it]
 18%|█▊        | 156/862 [59:06<4:35:27, 23.41s/it]
 18%|█▊        | 157/862 [59:24<4:16:18, 21.81s/it]
 18%|█▊        | 158/862 [59:44<4:10:49, 21.38s/it]
 18%|█▊        | 159/862 [1:00:00<3:53:14, 19.91s/it]
 19%|█▊        | 160/862 [1:00:14<3:29:48, 17.93s/it]
                                                     

 19%|█▊        | 160/862 [1:00:14<3:29:48, 17.93s/it]
 19%|█▊        | 161/862 [1:00:37<3:47:19, 19.46s/it]
 19%|█▉        | 162/862 [1:00:49<3:19:55, 17.14s/it]
 19%|█▉        | 163/862 [1:01:09<3:30:08, 18.04s/it]
 19%|█▉        | 164/862 [1:01:29<3:36:40, 18.63s/it]
 19%|█▉        | 165/862 [1:01:46<3:32:47, 18.32s/it]
                                                     

 19%|█▉        | 165/862 [1:01:46<3:32:47, 18.32s/it]
 19%|█▉        | 166/862 [1:02:01<3:18:20, 17.10s/it]
 19%|█▉        | 167/862 [1:02:12<2:59:18, 15.48s/it]
 19%|█▉        | 168/862 [1:02:31<3:11:46, 16.58s/it]
 20%|█▉        | 169/862 [1:02:48<3:12:26, 16.66s/it]
 20%|█▉        | 170/862 [1:03:07<3:20:34, 17.39s/it]
                                                     

 20%|█▉        | 170/862 [1:03:07<3:20:34, 17.39s/it]
 20%|█▉        | 171/862 [1:03:26<3:25:45, 17.87s/it]
 20%|█▉        | 172/862 [1:03:43<3:22:05, 17.57s/it]
 20%|██        | 173/862 [1:03:55<3:01:00, 15.76s/it]
 20%|██        | 174/862 [1:04:08<2:50:50, 14.90s/it]
 20%|██        | 175/862 [1:04:27<3:04:32, 16.12s/it]
                                                     

 20%|██        | 175/862 [1:04:27<3:04:32, 16.12s/it]
 20%|██        | 176/862 [1:04:38<2:47:13, 14.63s/it]
 21%|██        | 177/862 [1:04:53<2:50:23, 14.93s/it]
 21%|██        | 178/862 [1:05:13<3:05:55, 16.31s/it]
 21%|██        | 179/862 [1:05:30<3:07:24, 16.46s/it]
 21%|██        | 180/862 [1:05:47<3:08:54, 16.62s/it]
                                                     

 21%|██        | 180/862 [1:05:47<3:08:54, 16.62s/it]
 21%|██        | 181/862 [1:06:03<3:09:22, 16.68s/it]
 21%|██        | 182/862 [1:06:20<3:07:12, 16.52s/it]
 21%|██        | 183/862 [1:06:36<3:05:21, 16.38s/it]
 21%|██▏       | 184/862 [1:07:00<3:32:38, 18.82s/it]
 21%|██▏       | 185/862 [1:07:18<3:28:28, 18.48s/it]
                                                     

 21%|██▏       | 185/862 [1:07:18<3:28:28, 18.48s/it]
 22%|██▏       | 186/862 [1:07:37<3:28:54, 18.54s/it]
 22%|██▏       | 187/862 [1:07:51<3:15:56, 17.42s/it]
 22%|██▏       | 188/862 [1:08:10<3:18:28, 17.67s/it]
 22%|██▏       | 189/862 [1:08:32<3:35:34, 19.22s/it]
 22%|██▏       | 190/862 [1:08:56<3:51:16, 20.65s/it]
                                                     

 22%|██▏       | 190/862 [1:08:56<3:51:16, 20.65s/it]
 22%|██▏       | 191/862 [1:09:14<3:41:03, 19.77s/it]
 22%|██▏       | 192/862 [1:09:32<3:33:28, 19.12s/it]
 22%|██▏       | 193/862 [1:09:45<3:12:46, 17.29s/it]
 23%|██▎       | 194/862 [1:09:56<2:51:10, 15.38s/it]
 23%|██▎       | 195/862 [1:10:12<2:52:26, 15.51s/it]
                                                     

 23%|██▎       | 195/862 [1:10:12<2:52:26, 15.51s/it]
 23%|██▎       | 196/862 [1:10:30<3:03:26, 16.53s/it]
 23%|██▎       | 197/862 [1:10:46<3:00:45, 16.31s/it]
 23%|██▎       | 198/862 [1:11:07<3:13:45, 17.51s/it]
 23%|██▎       | 199/862 [1:11:24<3:13:01, 17.47s/it]
 23%|██▎       | 200/862 [1:11:36<2:56:00, 15.95s/it]
                                                     

 23%|██▎       | 200/862 [1:11:36<2:56:00, 15.95s/it]{'eval_loss': 0.1569332629442215, 'eval_runtime': 57.9274, 'eval_samples_per_second': 6.111, 'eval_steps_per_second': 3.056, 'epoch': 0.17}
{'loss': 0.2054, 'grad_norm': 0.5715239644050598, 'learning_rate': 0.00019120816282541063, 'epoch': 0.18}
{'loss': 0.2156, 'grad_norm': 0.5529009699821472, 'learning_rate': 0.00019040406009441073, 'epoch': 0.19}
{'loss': 0.227, 'grad_norm': 0.5217124819755554, 'learning_rate': 0.00018956662180484913, 'epoch': 0.19}
{'loss': 0.2191, 'grad_norm': 0.4713853895664215, 'learning_rate': 0.00018869615675340068, 'epoch': 0.2}
{'loss': 0.2052, 'grad_norm': 0.44989559054374695, 'learning_rate': 0.00018779298591501564, 'epoch': 0.2}
{'loss': 0.2009, 'grad_norm': 0.5454024076461792, 'learning_rate': 0.00018685744232456374, 'epoch': 0.21}
{'loss': 0.175, 'grad_norm': 0.5243939161300659, 'learning_rate': 0.0001858898709540309, 'epoch': 0.21}
{'loss': 0.1861, 'grad_norm': 0.4877641499042511, 'learning_rate': 0.0001848906285853145, 'epoch': 0.22}
{'loss': 0.1775, 'grad_norm': 0.45391249656677246, 'learning_rate': 0.0001838600836786635, 'epoch': 0.23}
{'loss': 0.1791, 'grad_norm': 0.49695494771003723, 'learning_rate': 0.0001827986162368129, 'epoch': 0.23}


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:00<00:49,  3.52it/s]

  2%|▏         | 3/177 [00:01<01:20,  2.17it/s]

  2%|▏         | 4/177 [00:01<01:08,  2.52it/s]

  3%|▎         | 5/177 [00:01<00:52,  3.26it/s]

  3%|▎         | 6/177 [00:02<00:52,  3.26it/s]

  4%|▍         | 7/177 [00:02<00:53,  3.17it/s]

  5%|▍         | 8/177 [00:02<00:53,  3.14it/s]

  5%|▌         | 9/177 [00:03<00:56,  2.99it/s]

  6%|▌         | 10/177 [00:03<00:57,  2.92it/s]

  6%|▌         | 11/177 [00:03<01:06,  2.50it/s]

  7%|▋         | 12/177 [00:04<00:58,  2.84it/s]

  7%|▋         | 13/177 [00:04<00:59,  2.76it/s]

  8%|▊         | 14/177 [00:04<00:56,  2.89it/s]

  8%|▊         | 15/177 [00:04<00:44,  3.65it/s]

  9%|▉         | 16/177 [00:05<00:43,  3.70it/s]

 10%|▉         | 17/177 [00:05<00:44,  3.62it/s]

 11%|█         | 19/177 [00:05<00:37,  4.19it/s]

 11%|█▏        | 20/177 [00:06<00:39,  3.96it/s]

 12%|█▏        | 21/177 [00:06<00:47,  3.30it/s]

 12%|█▏        | 22/177 [00:07<01:01,  2.51it/s]

 13%|█▎        | 23/177 [00:08<01:16,  2.02it/s]

 14%|█▎        | 24/177 [00:08<01:14,  2.04it/s]

 14%|█▍        | 25/177 [00:08<01:07,  2.26it/s]

 15%|█▍        | 26/177 [00:09<00:58,  2.59it/s]

 15%|█▌        | 27/177 [00:09<00:53,  2.78it/s]

 16%|█▌        | 28/177 [00:09<00:48,  3.05it/s]

 16%|█▋        | 29/177 [00:10<00:58,  2.53it/s]

 17%|█▋        | 30/177 [00:10<01:05,  2.24it/s]

 18%|█▊        | 31/177 [00:11<01:14,  1.96it/s]

 18%|█▊        | 32/177 [00:11<01:05,  2.21it/s]

 19%|█▊        | 33/177 [00:11<00:51,  2.81it/s]

 20%|█▉        | 35/177 [00:12<00:38,  3.66it/s]

 20%|██        | 36/177 [00:12<00:37,  3.79it/s]

 21%|██        | 37/177 [00:12<00:31,  4.38it/s]

 21%|██▏       | 38/177 [00:12<00:32,  4.31it/s]

 22%|██▏       | 39/177 [00:13<00:46,  2.98it/s]

 23%|██▎       | 40/177 [00:13<00:47,  2.88it/s]

 23%|██▎       | 41/177 [00:14<00:47,  2.86it/s]

 24%|██▎       | 42/177 [00:14<00:54,  2.50it/s]

 24%|██▍       | 43/177 [00:15<01:01,  2.19it/s]

 25%|██▍       | 44/177 [00:15<00:59,  2.22it/s]

 25%|██▌       | 45/177 [00:16<01:00,  2.17it/s]

 26%|██▌       | 46/177 [00:16<01:01,  2.13it/s]

 27%|██▋       | 47/177 [00:17<01:05,  2.00it/s]

 27%|██▋       | 48/177 [00:17<01:04,  1.99it/s]

 28%|██▊       | 49/177 [00:18<01:05,  1.94it/s]

 28%|██▊       | 50/177 [00:18<01:03,  1.99it/s]

 29%|██▉       | 51/177 [00:19<01:08,  1.83it/s]

 29%|██▉       | 52/177 [00:20<01:09,  1.79it/s]

 30%|██▉       | 53/177 [00:20<01:10,  1.76it/s]

 31%|███       | 54/177 [00:20<00:58,  2.10it/s]

 31%|███       | 55/177 [00:21<00:54,  2.25it/s]

 32%|███▏      | 56/177 [00:21<00:49,  2.46it/s]

 32%|███▏      | 57/177 [00:21<00:44,  2.73it/s]

 33%|███▎      | 58/177 [00:21<00:35,  3.32it/s]

 33%|███▎      | 59/177 [00:22<00:32,  3.65it/s]

 34%|███▍      | 61/177 [00:22<00:20,  5.68it/s]

 35%|███▌      | 62/177 [00:22<00:22,  5.22it/s]

 36%|███▌      | 63/177 [00:23<00:33,  3.43it/s]

 36%|███▌      | 64/177 [00:23<00:40,  2.77it/s]

 37%|███▋      | 65/177 [00:24<00:39,  2.80it/s]

 38%|███▊      | 67/177 [00:24<00:25,  4.24it/s]

 38%|███▊      | 68/177 [00:24<00:25,  4.23it/s]

 39%|███▉      | 69/177 [00:24<00:26,  4.11it/s]

 40%|████      | 71/177 [00:25<00:24,  4.28it/s]

 41%|████      | 72/177 [00:25<00:25,  4.11it/s]

 41%|████      | 73/177 [00:25<00:22,  4.61it/s]

 42%|████▏     | 74/177 [00:25<00:24,  4.26it/s]

 42%|████▏     | 75/177 [00:26<00:26,  3.89it/s]

 44%|████▎     | 77/177 [00:26<00:24,  4.01it/s]

 44%|████▍     | 78/177 [00:27<00:31,  3.15it/s]

 45%|████▍     | 79/177 [00:27<00:32,  3.04it/s]

 45%|████▌     | 80/177 [00:27<00:30,  3.19it/s]

 46%|████▌     | 81/177 [00:28<00:30,  3.18it/s]

 46%|████▋     | 82/177 [00:28<00:30,  3.12it/s]

 47%|████▋     | 83/177 [00:29<00:38,  2.43it/s]

 47%|████▋     | 84/177 [00:29<00:36,  2.58it/s]

 49%|████▊     | 86/177 [00:29<00:22,  4.05it/s]

 50%|████▉     | 88/177 [00:29<00:16,  5.31it/s]

 50%|█████     | 89/177 [00:29<00:14,  5.89it/s]

 51%|█████▏    | 91/177 [00:30<00:11,  7.79it/s]

 53%|█████▎    | 93/177 [00:30<00:13,  6.23it/s]

 53%|█████▎    | 94/177 [00:30<00:17,  4.69it/s]

 54%|█████▎    | 95/177 [00:31<00:23,  3.47it/s]

 54%|█████▍    | 96/177 [00:31<00:25,  3.16it/s]

 55%|█████▍    | 97/177 [00:32<00:28,  2.82it/s]

 55%|█████▌    | 98/177 [00:32<00:30,  2.55it/s]

 56%|█████▌    | 99/177 [00:33<00:35,  2.20it/s]

 56%|█████▋    | 100/177 [00:33<00:36,  2.09it/s]

 57%|█████▋    | 101/177 [00:34<00:37,  2.02it/s]

 58%|█████▊    | 102/177 [00:34<00:35,  2.11it/s]

 58%|█████▊    | 103/177 [00:35<00:31,  2.33it/s]

 59%|█████▉    | 104/177 [00:35<00:26,  2.72it/s]

 59%|█████▉    | 105/177 [00:35<00:27,  2.65it/s]

 60%|█████▉    | 106/177 [00:36<00:26,  2.70it/s]

 60%|██████    | 107/177 [00:36<00:23,  2.94it/s]

 62%|██████▏   | 109/177 [00:36<00:14,  4.59it/s]

 63%|██████▎   | 111/177 [00:37<00:14,  4.68it/s]

 63%|██████▎   | 112/177 [00:37<00:14,  4.37it/s]

 64%|██████▍   | 113/177 [00:37<00:17,  3.59it/s]

 64%|██████▍   | 114/177 [00:38<00:18,  3.38it/s]

 65%|██████▍   | 115/177 [00:38<00:18,  3.43it/s]

 66%|██████▌   | 116/177 [00:38<00:16,  3.66it/s]

 66%|██████▌   | 117/177 [00:38<00:13,  4.40it/s]

 67%|██████▋   | 118/177 [00:38<00:14,  4.09it/s]

 67%|██████▋   | 119/177 [00:39<00:19,  2.98it/s]

 68%|██████▊   | 120/177 [00:39<00:18,  3.10it/s]

 68%|██████▊   | 121/177 [00:39<00:14,  3.85it/s]

 69%|██████▉   | 123/177 [00:40<00:11,  4.77it/s]

 70%|███████   | 124/177 [00:40<00:12,  4.12it/s]

 71%|███████   | 125/177 [00:41<00:17,  3.01it/s]

 71%|███████   | 126/177 [00:41<00:17,  2.99it/s]

 72%|███████▏  | 127/177 [00:41<00:18,  2.77it/s]

 72%|███████▏  | 128/177 [00:42<00:17,  2.81it/s]

 73%|███████▎  | 129/177 [00:42<00:16,  2.82it/s]

 73%|███████▎  | 130/177 [00:43<00:18,  2.57it/s]

 74%|███████▍  | 131/177 [00:43<00:19,  2.38it/s]

 75%|███████▍  | 132/177 [00:43<00:16,  2.74it/s]

 75%|███████▌  | 133/177 [00:44<00:15,  2.81it/s]

 76%|███████▌  | 134/177 [00:44<00:16,  2.57it/s]

 76%|███████▋  | 135/177 [00:45<00:15,  2.65it/s]

 77%|███████▋  | 137/177 [00:45<00:09,  4.09it/s]

 79%|███████▊  | 139/177 [00:45<00:07,  5.43it/s]

 79%|███████▉  | 140/177 [00:45<00:07,  5.07it/s]

 80%|███████▉  | 141/177 [00:46<00:09,  3.66it/s]

 80%|████████  | 142/177 [00:46<00:11,  3.14it/s]

 81%|████████  | 143/177 [00:47<00:12,  2.63it/s]

 81%|████████▏ | 144/177 [00:47<00:13,  2.50it/s]

 82%|████████▏ | 145/177 [00:48<00:14,  2.19it/s]

 82%|████████▏ | 146/177 [00:48<00:14,  2.15it/s]

 83%|████████▎ | 147/177 [00:49<00:15,  2.00it/s]

 84%|████████▎ | 148/177 [00:49<00:14,  1.94it/s]

 84%|████████▍ | 149/177 [00:50<00:13,  2.08it/s]

 85%|████████▌ | 151/177 [00:50<00:09,  2.67it/s]

 86%|████████▌ | 152/177 [00:51<00:09,  2.72it/s]

 86%|████████▋ | 153/177 [00:51<00:08,  2.92it/s]

 87%|████████▋ | 154/177 [00:51<00:06,  3.31it/s]

 88%|████████▊ | 155/177 [00:51<00:05,  3.96it/s]

 89%|████████▊ | 157/177 [00:51<00:03,  5.69it/s]

 89%|████████▉ | 158/177 [00:52<00:03,  4.87it/s]

 90%|████████▉ | 159/177 [00:52<00:04,  4.38it/s]

 91%|█████████ | 161/177 [00:52<00:02,  6.03it/s]

 92%|█████████▏| 163/177 [00:52<00:02,  6.45it/s]

 93%|█████████▎| 164/177 [00:53<00:02,  4.91it/s]

 93%|█████████▎| 165/177 [00:53<00:03,  3.45it/s]

 94%|█████████▍| 166/177 [00:54<00:03,  2.92it/s]

 94%|█████████▍| 167/177 [00:54<00:03,  2.92it/s]

 95%|█████████▍| 168/177 [00:55<00:03,  2.71it/s]

 95%|█████████▌| 169/177 [00:55<00:03,  2.62it/s]

 96%|█████████▌| 170/177 [00:55<00:02,  2.78it/s]

 97%|█████████▋| 171/177 [00:56<00:02,  2.85it/s]

 97%|█████████▋| 172/177 [00:56<00:01,  2.78it/s]

 98%|█████████▊| 173/177 [00:56<00:01,  2.74it/s]

 99%|█████████▉| 175/177 [00:57<00:00,  4.21it/s]

 99%|█████████▉| 176/177 [00:57<00:00,  3.90it/s]

100%|██████████| 177/177 [00:57<00:00,  3.66it/s]
                                                     

                                                 

 23%|██▎       | 200/862 [1:12:34<2:56:00, 15.95s/it]

100%|██████████| 177/177 [00:57<00:00,  3.66it/s]

                                                 wandb: WARNING URL not available in offline run
`rope_scaling`'s beta_fast field must be a float, got 32
`rope_scaling`'s beta_slow field must be a float, got 1

 23%|██▎       | 201/862 [1:12:52<6:12:46, 33.84s/it]
 23%|██▎       | 202/862 [1:13:19<5:51:25, 31.95s/it]
 24%|██▎       | 203/862 [1:14:05<6:36:52, 36.13s/it]
 24%|██▎       | 204/862 [1:14:49<7:02:16, 38.51s/it]
 24%|██▍       | 205/862 [1:15:10<6:03:43, 33.22s/it]
                                                     

 24%|██▍       | 205/862 [1:15:10<6:03:43, 33.22s/it]
 24%|██▍       | 206/862 [1:15:31<5:22:17, 29.48s/it]
 24%|██▍       | 207/862 [1:15:53<4:58:34, 27.35s/it]
 24%|██▍       | 208/862 [1:16:09<4:19:59, 23.85s/it]
 24%|██▍       | 209/862 [1:16:40<4:44:17, 26.12s/it]
 24%|██▍       | 210/862 [1:16:58<4:15:58, 23.56s/it]
                                                     

 24%|██▍       | 210/862 [1:16:58<4:15:58, 23.56s/it]
 24%|██▍       | 211/862 [1:17:16<3:57:09, 21.86s/it]
 25%|██▍       | 212/862 [1:17:28<3:26:21, 19.05s/it]
 25%|██▍       | 213/862 [1:17:43<3:12:50, 17.83s/it]
 25%|██▍       | 214/862 [1:17:57<2:57:37, 16.45s/it]
 25%|██▍       | 215/862 [1:18:08<2:40:25, 14.88s/it]
                                                     

 25%|██▍       | 215/862 [1:18:08<2:40:25, 14.88s/it]
 25%|██▌       | 216/862 [1:18:25<2:46:12, 15.44s/it]
 25%|██▌       | 217/862 [1:18:40<2:46:09, 15.46s/it]
 25%|██▌       | 218/862 [1:18:57<2:51:56, 16.02s/it]
 25%|██▌       | 219/862 [1:19:09<2:37:44, 14.72s/it]
 26%|██▌       | 220/862 [1:19:24<2:38:33, 14.82s/it]
                                                     

 26%|██▌       | 220/862 [1:19:24<2:38:33, 14.82s/it]
 26%|██▌       | 221/862 [1:19:47<3:02:28, 17.08s/it]
 26%|██▌       | 222/862 [1:19:59<2:48:16, 15.78s/it]
 26%|██▌       | 223/862 [1:20:19<2:59:42, 16.87s/it]
 26%|██▌       | 224/862 [1:20:36<3:00:20, 16.96s/it]
 26%|██▌       | 225/862 [1:20:48<2:45:04, 15.55s/it]
                                                     

 26%|██▌       | 225/862 [1:20:48<2:45:04, 15.55s/it]
 26%|██▌       | 226/862 [1:20:57<2:22:05, 13.41s/it]
 26%|██▋       | 227/862 [1:21:13<2:31:22, 14.30s/it]
 26%|██▋       | 228/862 [1:21:27<2:29:56, 14.19s/it]
 27%|██▋       | 229/862 [1:21:43<2:36:02, 14.79s/it]
 27%|██▋       | 230/862 [1:21:57<2:33:46, 14.60s/it]
                                                     

 27%|██▋       | 230/862 [1:21:57<2:33:46, 14.60s/it]
 27%|██▋       | 231/862 [1:22:11<2:31:44, 14.43s/it]
 27%|██▋       | 232/862 [1:22:34<2:57:54, 16.94s/it]
 27%|██▋       | 233/862 [1:22:49<2:51:51, 16.39s/it]
 27%|██▋       | 234/862 [1:23:04<2:45:18, 15.79s/it]
 27%|██▋       | 235/862 [1:23:31<3:20:55, 19.23s/it]
                                                     

 27%|██▋       | 235/862 [1:23:31<3:20:55, 19.23s/it]
 27%|██▋       | 236/862 [1:23:44<3:00:26, 17.30s/it]
 27%|██▋       | 237/862 [1:24:04<3:09:25, 18.18s/it]
 28%|██▊       | 238/862 [1:24:19<2:59:09, 17.23s/it]
 28%|██▊       | 239/862 [1:24:40<3:09:53, 18.29s/it]
 28%|██▊       | 240/862 [1:24:58<3:10:34, 18.38s/it]
                                                     

 28%|██▊       | 240/862 [1:24:58<3:10:34, 18.38s/it]
 28%|██▊       | 241/862 [1:25:11<2:52:17, 16.65s/it]
 28%|██▊       | 242/862 [1:25:29<2:55:59, 17.03s/it]
 28%|██▊       | 243/862 [1:25:47<3:00:30, 17.50s/it]
 28%|██▊       | 244/862 [1:26:02<2:52:50, 16.78s/it]
 28%|██▊       | 245/862 [1:26:14<2:36:15, 15.20s/it]
                                                     

 28%|██▊       | 245/862 [1:26:14<2:36:15, 15.20s/it]
 29%|██▊       | 246/862 [1:26:27<2:29:29, 14.56s/it]
 29%|██▊       | 247/862 [1:26:41<2:27:27, 14.39s/it]
 29%|██▉       | 248/862 [1:26:54<2:24:03, 14.08s/it]
 29%|██▉       | 249/862 [1:27:14<2:39:40, 15.63s/it]
 29%|██▉       | 250/862 [1:27:31<2:44:38, 16.14s/it]
                                                     

 29%|██▉       | 250/862 [1:27:31<2:44:38, 16.14s/it]{'eval_loss': 0.14798370003700256, 'eval_runtime': 57.9291, 'eval_samples_per_second': 6.111, 'eval_steps_per_second': 3.055, 'epoch': 0.23}
{'loss': 0.2095, 'grad_norm': 0.3668426275253296, 'learning_rate': 0.0001817066176648615, 'epoch': 0.24}
{'loss': 0.2096, 'grad_norm': 0.48783811926841736, 'learning_rate': 0.0001805844906259452, 'epoch': 0.24}
{'loss': 0.1579, 'grad_norm': 0.3652338981628418, 'learning_rate': 0.00017943264889275944, 'epoch': 0.25}
{'loss': 0.169, 'grad_norm': 0.4943305552005768, 'learning_rate': 0.00017825151719498466, 'epoch': 0.26}
{'loss': 0.1633, 'grad_norm': 0.4297771453857422, 'learning_rate': 0.0001770415310626715, 'epoch': 0.26}
{'loss': 0.1713, 'grad_norm': 0.5426889657974243, 'learning_rate': 0.00017580313666564395, 'epoch': 0.27}
{'loss': 0.1921, 'grad_norm': 0.5065458416938782, 'learning_rate': 0.0001745367906489786, 'epoch': 0.27}
{'loss': 0.2107, 'grad_norm': 0.452774316072464, 'learning_rate': 0.00017324295996462146, 'epoch': 0.28}
{'loss': 0.1468, 'grad_norm': 0.37844130396842957, 'learning_rate': 0.00017192212169920459, 'epoch': 0.28}
{'loss': 0.1421, 'grad_norm': 0.3819119930267334, 'learning_rate': 0.00017057476289812504, 'epoch': 0.29}


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:00<00:50,  3.49it/s]

  2%|▏         | 3/177 [00:01<01:22,  2.11it/s]

  2%|▏         | 4/177 [00:01<01:10,  2.47it/s]

  3%|▎         | 5/177 [00:01<00:52,  3.29it/s]

  3%|▎         | 6/177 [00:02<00:52,  3.27it/s]

  4%|▍         | 7/177 [00:02<00:54,  3.11it/s]

  5%|▍         | 8/177 [00:02<00:54,  3.08it/s]

  5%|▌         | 9/177 [00:03<00:57,  2.92it/s]

  6%|▌         | 10/177 [00:03<00:58,  2.87it/s]

  6%|▌         | 11/177 [00:03<01:07,  2.47it/s]

  7%|▋         | 12/177 [00:04<00:58,  2.83it/s]

  7%|▋         | 13/177 [00:04<00:59,  2.75it/s]

  8%|▊         | 14/177 [00:04<00:56,  2.88it/s]

  8%|▊         | 15/177 [00:05<00:44,  3.65it/s]

  9%|▉         | 16/177 [00:05<00:43,  3.71it/s]

 10%|▉         | 17/177 [00:05<00:44,  3.62it/s]

 11%|█         | 19/177 [00:05<00:37,  4.23it/s]

 11%|█▏        | 20/177 [00:06<00:38,  4.06it/s]

 12%|█▏        | 21/177 [00:06<00:46,  3.37it/s]

 12%|█▏        | 22/177 [00:07<01:01,  2.54it/s]

 13%|█▎        | 23/177 [00:08<01:14,  2.06it/s]

 14%|█▎        | 24/177 [00:08<01:14,  2.06it/s]

 14%|█▍        | 25/177 [00:08<01:05,  2.33it/s]

 15%|█▍        | 26/177 [00:09<00:57,  2.61it/s]

 15%|█▌        | 27/177 [00:09<00:53,  2.80it/s]

 16%|█▌        | 28/177 [00:09<00:48,  3.06it/s]

 16%|█▋        | 29/177 [00:10<00:58,  2.54it/s]

 17%|█▋        | 30/177 [00:10<01:05,  2.25it/s]

 18%|█▊        | 31/177 [00:11<01:14,  1.97it/s]

 18%|█▊        | 32/177 [00:11<01:05,  2.21it/s]

 19%|█▊        | 33/177 [00:11<00:51,  2.82it/s]

 20%|█▉        | 35/177 [00:12<00:38,  3.67it/s]

 20%|██        | 36/177 [00:12<00:37,  3.80it/s]

 21%|██        | 37/177 [00:12<00:31,  4.40it/s]

 21%|██▏       | 38/177 [00:12<00:32,  4.33it/s]

 22%|██▏       | 39/177 [00:13<00:46,  2.98it/s]

 23%|██▎       | 40/177 [00:13<00:47,  2.88it/s]

 23%|██▎       | 41/177 [00:14<00:47,  2.87it/s]

 24%|██▎       | 42/177 [00:14<00:53,  2.50it/s]

 24%|██▍       | 43/177 [00:15<01:01,  2.20it/s]

 25%|██▍       | 44/177 [00:15<00:59,  2.23it/s]

 25%|██▌       | 45/177 [00:16<01:00,  2.18it/s]

 26%|██▌       | 46/177 [00:16<01:01,  2.14it/s]

 27%|██▋       | 47/177 [00:17<01:04,  2.00it/s]

 27%|██▋       | 48/177 [00:17<01:04,  2.00it/s]

 28%|██▊       | 49/177 [00:18<01:05,  1.95it/s]

 28%|██▊       | 50/177 [00:18<01:03,  1.99it/s]

 29%|██▉       | 51/177 [00:19<01:08,  1.84it/s]

 29%|██▉       | 52/177 [00:19<01:09,  1.79it/s]

 30%|██▉       | 53/177 [00:20<01:10,  1.76it/s]

 31%|███       | 54/177 [00:20<00:58,  2.11it/s]

 31%|███       | 55/177 [00:21<00:53,  2.26it/s]

 32%|███▏      | 56/177 [00:21<00:48,  2.48it/s]

 32%|███▏      | 57/177 [00:21<00:42,  2.85it/s]

 33%|███▎      | 58/177 [00:21<00:36,  3.30it/s]

 33%|███▎      | 59/177 [00:22<00:28,  4.10it/s]

 34%|███▍      | 61/177 [00:22<00:18,  6.22it/s]

 35%|███▌      | 62/177 [00:22<00:20,  5.55it/s]

 36%|███▌      | 63/177 [00:22<00:32,  3.54it/s]

 36%|███▌      | 64/177 [00:23<00:40,  2.82it/s]

 37%|███▋      | 65/177 [00:23<00:39,  2.85it/s]

 38%|███▊      | 67/177 [00:24<00:25,  4.30it/s]

 38%|███▊      | 68/177 [00:24<00:25,  4.27it/s]

 39%|███▉      | 69/177 [00:24<00:26,  4.13it/s]

 40%|████      | 71/177 [00:24<00:24,  4.33it/s]

 41%|████      | 72/177 [00:25<00:25,  4.11it/s]

 41%|████      | 73/177 [00:25<00:22,  4.62it/s]

 42%|████▏     | 74/177 [00:25<00:24,  4.27it/s]

 42%|████▏     | 75/177 [00:26<00:26,  3.91it/s]

 44%|████▎     | 77/177 [00:26<00:30,  3.31it/s]

 44%|████▍     | 78/177 [00:27<00:33,  2.98it/s]

 45%|████▍     | 79/177 [00:27<00:33,  2.91it/s]

 45%|████▌     | 80/177 [00:27<00:31,  3.08it/s]

 46%|████▌     | 81/177 [00:28<00:30,  3.10it/s]

 46%|████▋     | 82/177 [00:28<00:31,  3.00it/s]

 47%|████▋     | 83/177 [00:29<00:38,  2.43it/s]

 47%|████▋     | 84/177 [00:29<00:35,  2.59it/s]

 49%|████▊     | 86/177 [00:29<00:22,  4.07it/s]

 50%|████▉     | 88/177 [00:29<00:16,  5.38it/s]

 50%|█████     | 89/177 [00:29<00:14,  5.95it/s]

 51%|█████▏    | 91/177 [00:29<00:10,  7.85it/s]

 53%|█████▎    | 93/177 [00:30<00:13,  6.25it/s]

 53%|█████▎    | 94/177 [00:30<00:17,  4.69it/s]

 54%|█████▎    | 95/177 [00:31<00:23,  3.46it/s]

 54%|█████▍    | 96/177 [00:31<00:25,  3.15it/s]

 55%|█████▍    | 97/177 [00:32<00:28,  2.83it/s]

 55%|█████▌    | 98/177 [00:32<00:30,  2.56it/s]

 56%|█████▌    | 99/177 [00:33<00:35,  2.20it/s]

 56%|█████▋    | 100/177 [00:33<00:36,  2.09it/s]

 57%|█████▋    | 101/177 [00:34<00:37,  2.02it/s]

 58%|█████▊    | 102/177 [00:34<00:35,  2.11it/s]

 58%|█████▊    | 103/177 [00:35<00:31,  2.33it/s]

 59%|█████▉    | 104/177 [00:35<00:26,  2.72it/s]

 59%|█████▉    | 105/177 [00:35<00:27,  2.64it/s]

 60%|█████▉    | 106/177 [00:36<00:26,  2.69it/s]

 60%|██████    | 107/177 [00:36<00:24,  2.85it/s]

 62%|██████▏   | 109/177 [00:36<00:15,  4.28it/s]

 63%|██████▎   | 111/177 [00:37<00:14,  4.50it/s]

 63%|██████▎   | 112/177 [00:37<00:15,  4.10it/s]

 64%|██████▍   | 113/177 [00:37<00:18,  3.43it/s]

 64%|██████▍   | 114/177 [00:38<00:19,  3.27it/s]

 65%|██████▍   | 115/177 [00:38<00:18,  3.32it/s]

 66%|██████▌   | 116/177 [00:38<00:17,  3.54it/s]

 66%|██████▌   | 117/177 [00:38<00:14,  4.22it/s]

 67%|██████▋   | 118/177 [00:39<00:14,  3.97it/s]

 67%|██████▋   | 119/177 [00:39<00:19,  2.93it/s]

 68%|██████▊   | 120/177 [00:39<00:18,  3.07it/s]

 68%|██████▊   | 121/177 [00:40<00:14,  3.81it/s]

 69%|██████▉   | 123/177 [00:40<00:11,  4.75it/s]

 70%|███████   | 124/177 [00:40<00:12,  4.10it/s]

 71%|███████   | 125/177 [00:41<00:17,  3.00it/s]

 71%|███████   | 126/177 [00:41<00:17,  2.99it/s]

 72%|███████▏  | 127/177 [00:42<00:17,  2.78it/s]

 72%|███████▏  | 128/177 [00:42<00:16,  2.88it/s]

 73%|███████▎  | 129/177 [00:42<00:16,  2.87it/s]

 73%|███████▎  | 130/177 [00:43<00:18,  2.60it/s]

 74%|███████▍  | 131/177 [00:43<00:18,  2.42it/s]

 75%|███████▍  | 132/177 [00:43<00:15,  2.86it/s]

 75%|███████▌  | 133/177 [00:44<00:14,  2.94it/s]

 76%|███████▌  | 134/177 [00:44<00:16,  2.63it/s]

 76%|███████▋  | 135/177 [00:45<00:15,  2.75it/s]

 77%|███████▋  | 137/177 [00:45<00:09,  4.33it/s]

 79%|███████▊  | 139/177 [00:45<00:06,  5.72it/s]

 79%|███████▉  | 140/177 [00:45<00:07,  5.26it/s]

 80%|███████▉  | 141/177 [00:46<00:09,  3.72it/s]

 80%|████████  | 142/177 [00:46<00:11,  3.17it/s]

 81%|████████  | 143/177 [00:47<00:12,  2.66it/s]

 81%|████████▏ | 144/177 [00:47<00:13,  2.51it/s]

 82%|████████▏ | 145/177 [00:48<00:14,  2.20it/s]

 82%|████████▏ | 146/177 [00:48<00:14,  2.15it/s]

 83%|████████▎ | 147/177 [00:49<00:15,  2.00it/s]

 84%|████████▎ | 148/177 [00:49<00:14,  1.93it/s]

 84%|████████▍ | 149/177 [00:50<00:13,  2.08it/s]

 85%|████████▌ | 151/177 [00:50<00:09,  2.66it/s]

 86%|████████▌ | 152/177 [00:51<00:09,  2.71it/s]

 86%|████████▋ | 153/177 [00:51<00:08,  2.93it/s]

 87%|████████▋ | 154/177 [00:51<00:06,  3.32it/s]

 88%|████████▊ | 155/177 [00:51<00:05,  3.98it/s]

 89%|████████▊ | 157/177 [00:51<00:03,  5.71it/s]

 89%|████████▉ | 158/177 [00:52<00:03,  4.89it/s]

 90%|████████▉ | 159/177 [00:52<00:04,  4.39it/s]

 91%|█████████ | 161/177 [00:52<00:02,  6.05it/s]

 92%|█████████▏| 163/177 [00:52<00:02,  6.50it/s]

 93%|█████████▎| 164/177 [00:53<00:02,  4.94it/s]

 93%|█████████▎| 165/177 [00:53<00:03,  3.47it/s]

 94%|█████████▍| 166/177 [00:54<00:03,  2.93it/s]

 94%|█████████▍| 167/177 [00:54<00:03,  2.92it/s]

 95%|█████████▍| 168/177 [00:55<00:03,  2.71it/s]

 95%|█████████▌| 169/177 [00:55<00:03,  2.62it/s]

 96%|█████████▌| 170/177 [00:55<00:02,  2.79it/s]

 97%|█████████▋| 171/177 [00:56<00:02,  2.85it/s]

 97%|█████████▋| 172/177 [00:56<00:01,  2.79it/s]

 98%|█████████▊| 173/177 [00:56<00:01,  2.74it/s]

 99%|█████████▉| 175/177 [00:57<00:00,  4.20it/s]

 99%|█████████▉| 176/177 [00:57<00:00,  3.89it/s]

100%|██████████| 177/177 [00:57<00:00,  3.65it/s]
                                                     

                                                 

 29%|██▉       | 250/862 [1:28:29<2:44:38, 16.14s/it]

100%|██████████| 177/177 [00:57<00:00,  3.65it/s]

                                                 
 29%|██▉       | 251/862 [1:28:50<5:56:21, 34.99s/it]
 29%|██▉       | 252/862 [1:29:10<5:11:30, 30.64s/it]
 29%|██▉       | 253/862 [1:29:25<4:23:29, 25.96s/it]
 29%|██▉       | 254/862 [1:29:49<4:15:26, 25.21s/it]
 30%|██▉       | 255/862 [1:29:58<3:26:01, 20.36s/it]
                                                     

 30%|██▉       | 255/862 [1:29:58<3:26:01, 20.36s/it]
 30%|██▉       | 256/862 [1:30:10<2:59:58, 17.82s/it]
 30%|██▉       | 257/862 [1:30:29<3:04:39, 18.31s/it]
 30%|██▉       | 258/862 [1:30:48<3:05:17, 18.41s/it]
 30%|███       | 259/862 [1:31:05<3:02:10, 18.13s/it]
 30%|███       | 260/862 [1:31:29<3:19:22, 19.87s/it]
                                                     

 30%|███       | 260/862 [1:31:29<3:19:22, 19.87s/it]
 30%|███       | 261/862 [1:31:50<3:20:35, 20.03s/it]
 30%|███       | 262/862 [1:32:05<3:06:01, 18.60s/it]
 31%|███       | 263/862 [1:32:16<2:42:33, 16.28s/it]
 31%|███       | 264/862 [1:32:33<2:43:34, 16.41s/it]
 31%|███       | 265/862 [1:32:49<2:44:40, 16.55s/it]
                                                     

 31%|███       | 265/862 [1:32:49<2:44:40, 16.55s/it]
 31%|███       | 266/862 [1:33:00<2:27:20, 14.83s/it]
 31%|███       | 267/862 [1:33:09<2:09:37, 13.07s/it]
 31%|███       | 268/862 [1:33:32<2:39:20, 16.09s/it]
 31%|███       | 269/862 [1:33:48<2:39:02, 16.09s/it]
 31%|███▏      | 270/862 [1:34:04<2:36:51, 15.90s/it]
                                                     

 31%|███▏      | 270/862 [1:34:04<2:36:51, 15.90s/it]
 31%|███▏      | 271/862 [1:34:24<2:48:06, 17.07s/it]
 32%|███▏      | 272/862 [1:34:40<2:45:05, 16.79s/it]
 32%|███▏      | 273/862 [1:34:52<2:31:32, 15.44s/it]
 32%|███▏      | 274/862 [1:35:08<2:31:14, 15.43s/it]
 32%|███▏      | 275/862 [1:35:23<2:32:29, 15.59s/it]
                                                     

 32%|███▏      | 275/862 [1:35:23<2:32:29, 15.59s/it]
 32%|███▏      | 276/862 [1:35:43<2:44:53, 16.88s/it]
 32%|███▏      | 277/862 [1:36:03<2:51:15, 17.57s/it]
 32%|███▏      | 278/862 [1:36:16<2:39:15, 16.36s/it]
 32%|███▏      | 279/862 [1:36:35<2:46:26, 17.13s/it]
 32%|███▏      | 280/862 [1:36:58<3:02:07, 18.78s/it]
                                                     

 32%|███▏      | 280/862 [1:36:58<3:02:07, 18.78s/it]
 33%|███▎      | 281/862 [1:37:10<2:43:11, 16.85s/it]
 33%|███▎      | 282/862 [1:37:24<2:35:45, 16.11s/it]
 33%|███▎      | 283/862 [1:37:43<2:41:38, 16.75s/it]
 33%|███▎      | 284/862 [1:38:01<2:46:17, 17.26s/it]
 33%|███▎      | 285/862 [1:38:20<2:50:57, 17.78s/it]
                                                     

 33%|███▎      | 285/862 [1:38:20<2:50:57, 17.78s/it]
 33%|███▎      | 286/862 [1:38:30<2:26:47, 15.29s/it]
 33%|███▎      | 287/862 [1:38:41<2:15:07, 14.10s/it]
 33%|███▎      | 288/862 [1:39:02<2:35:31, 16.26s/it]
 34%|███▎      | 289/862 [1:39:20<2:40:13, 16.78s/it]
 34%|███▎      | 290/862 [1:39:34<2:32:01, 15.95s/it]
                                                     

 34%|███▎      | 290/862 [1:39:34<2:32:01, 15.95s/it]
 34%|███▍      | 291/862 [1:39:51<2:35:27, 16.33s/it]
 34%|███▍      | 292/862 [1:40:03<2:21:13, 14.87s/it]
 34%|███▍      | 293/862 [1:40:22<2:34:19, 16.27s/it]
 34%|███▍      | 294/862 [1:40:32<2:16:24, 14.41s/it]
 34%|███▍      | 295/862 [1:40:51<2:28:33, 15.72s/it]
                                                     

 34%|███▍      | 295/862 [1:40:51<2:28:33, 15.72s/it]
 34%|███▍      | 296/862 [1:40:59<2:06:09, 13.37s/it]
 34%|███▍      | 297/862 [1:41:17<2:18:22, 14.69s/it]
 35%|███▍      | 298/862 [1:41:36<2:30:11, 15.98s/it]
 35%|███▍      | 299/862 [1:41:53<2:34:21, 16.45s/it]
 35%|███▍      | 300/862 [1:42:09<2:31:03, 16.13s/it]
                                                     

 35%|███▍      | 300/862 [1:42:09<2:31:03, 16.13s/it]{'eval_loss': 0.1410292237997055, 'eval_runtime': 57.9112, 'eval_samples_per_second': 6.113, 'eval_steps_per_second': 3.056, 'epoch': 0.29}
{'loss': 0.1774, 'grad_norm': 0.41748952865600586, 'learning_rate': 0.00016920138038595216, 'epoch': 0.3}
{'loss': 0.207, 'grad_norm': 0.4169933795928955, 'learning_rate': 0.0001678024805832284, 'epoch': 0.3}
{'loss': 0.1492, 'grad_norm': 0.3983752727508545, 'learning_rate': 0.0001663785793197323, 'epoch': 0.31}
{'loss': 0.1154, 'grad_norm': 0.3772595524787903, 'learning_rate': 0.00016493020164427152, 'epoch': 0.31}
{'loss': 0.177, 'grad_norm': 0.3978261947631836, 'learning_rate': 0.00016345788163107646, 'epoch': 0.32}
{'loss': 0.2136, 'grad_norm': 0.41389644145965576, 'learning_rate': 0.0001619621621828659, 'epoch': 0.33}
{'loss': 0.1915, 'grad_norm': 0.42459338903427124, 'learning_rate': 0.0001604435948306575, 'epoch': 0.33}
{'loss': 0.1747, 'grad_norm': 0.41835540533065796, 'learning_rate': 0.0001589027395303962, 'epoch': 0.34}
{'loss': 0.1698, 'grad_norm': 0.5012316703796387, 'learning_rate': 0.0001573401644564764, 'epoch': 0.34}
{'loss': 0.1693, 'grad_norm': 0.43907007575035095, 'learning_rate': 0.00015575644579223362, 'epoch': 0.35}


  0%|          | 0/177 [00:00<?, ?it/s]

  1%|          | 2/177 [00:00<00:49,  3.52it/s]

  2%|▏         | 3/177 [00:01<01:20,  2.17it/s]

  2%|▏         | 4/177 [00:01<01:08,  2.53it/s]

  3%|▎         | 5/177 [00:01<00:50,  3.41it/s]

  3%|▎         | 6/177 [00:01<00:50,  3.36it/s]

  4%|▍         | 7/177 [00:02<00:52,  3.24it/s]

  5%|▍         | 8/177 [00:02<00:52,  3.19it/s]

  5%|▌         | 9/177 [00:02<00:55,  3.03it/s]

  6%|▌         | 10/177 [00:03<00:59,  2.79it/s]

  6%|▌         | 11/177 [00:03<01:05,  2.53it/s]

  7%|▋         | 12/177 [00:04<00:57,  2.88it/s]

  7%|▋         | 13/177 [00:04<00:58,  2.78it/s]

  8%|▊         | 14/177 [00:04<00:56,  2.90it/s]

  8%|▊         | 15/177 [00:04<00:44,  3.67it/s]

  9%|▉         | 16/177 [00:05<00:43,  3.72it/s]

 10%|▉         | 17/177 [00:05<00:44,  3.61it/s]

 11%|█         | 19/177 [00:05<00:37,  4.22it/s]

 11%|█▏        | 20/177 [00:06<00:38,  4.04it/s]

 12%|█▏        | 21/177 [00:06<00:46,  3.38it/s]

 12%|█▏        | 22/177 [00:07<01:01,  2.53it/s]

 13%|█▎        | 23/177 [00:07<01:14,  2.06it/s]

 14%|█▎        | 24/177 [00:08<01:14,  2.07it/s]

 14%|█▍        | 25/177 [00:08<01:05,  2.33it/s]

 15%|█▍        | 26/177 [00:09<00:57,  2.62it/s]

 15%|█▌        | 27/177 [00:09<00:53,  2.80it/s]

 16%|█▌        | 28/177 [00:09<00:48,  3.06it/s]

 16%|█▋        | 29/177 [00:10<00:58,  2.53it/s]

 17%|█▋        | 30/177 [00:10<01:05,  2.24it/s]

 18%|█▊        | 31/177 [00:11<01:14,  1.96it/s]

 18%|█▊        | 32/177 [00:11<01:05,  2.21it/s]

 19%|█▊        | 33/177 [00:11<00:51,  2.81it/s]

 20%|█▉        | 35/177 [00:12<00:38,  3.66it/s]

 20%|██        | 36/177 [00:12<00:37,  3.79it/s]

 21%|██        | 37/177 [00:12<00:32,  4.31it/s]

 21%|██▏       | 38/177 [00:12<00:32,  4.21it/s]

 22%|██▏       | 39/177 [00:13<00:47,  2.91it/s]

 23%|██▎       | 40/177 [00:13<00:49,  2.77it/s]

 23%|██▎       | 41/177 [00:14<00:49,  2.72it/s]

 24%|██▎       | 42/177 [00:14<00:55,  2.43it/s]

 24%|██▍       | 43/177 [00:15<01:02,  2.13it/s]

 25%|██▍       | 44/177 [00:15<01:00,  2.19it/s]

 25%|██▌       | 45/177 [00:16<01:01,  2.14it/s]

 26%|██▌       | 46/177 [00:16<01:02,  2.11it/s]

 27%|██▋       | 47/177 [00:17<01:05,  1.98it/s]

 27%|██▋       | 48/177 [00:17<01:05,  1.98it/s]

 28%|██▊       | 49/177 [00:18<01:06,  1.94it/s]

 28%|██▊       | 50/177 [00:18<01:04,  1.98it/s]

 29%|██▉       | 51/177 [00:19<01:08,  1.83it/s]

 29%|██▉       | 52/177 [00:20<01:09,  1.79it/s]

 30%|██▉       | 53/177 [00:20<01:10,  1.76it/s]

 31%|███       | 54/177 [00:20<00:58,  2.10it/s]

 31%|███       | 55/177 [00:21<00:54,  2.25it/s]

 32%|███▏      | 56/177 [00:21<00:48,  2.47it/s]

 32%|███▏      | 57/177 [00:21<00:42,  2.85it/s]

 33%|███▎      | 58/177 [00:21<00:36,  3.29it/s]

 33%|███▎      | 59/177 [00:22<00:28,  4.10it/s]

 34%|███▍      | 61/177 [00:22<00:18,  6.22it/s]

 35%|███▌      | 62/177 [00:22<00:20,  5.55it/s]

 36%|███▌      | 63/177 [00:23<00:32,  3.54it/s]

 36%|███▌      | 64/177 [00:23<00:39,  2.83it/s]

 37%|███▋      | 65/177 [00:23<00:39,  2.86it/s]

 38%|███▊      | 67/177 [00:24<00:25,  4.32it/s]

 38%|███▊      | 68/177 [00:24<00:25,  4.30it/s]

 39%|███▉      | 69/177 [00:24<00:26,  4.15it/s]

 40%|████      | 71/177 [00:25<00:24,  4.35it/s]

 41%|████      | 72/177 [00:25<00:25,  4.13it/s]

 41%|████      | 73/177 [00:25<00:22,  4.64it/s]

 42%|████▏     | 74/177 [00:25<00:24,  4.28it/s]

 42%|████▏     | 75/177 [00:26<00:26,  3.91it/s]

 44%|████▎     | 77/177 [00:26<00:24,  4.02it/s]

 44%|████▍     | 78/177 [00:27<00:31,  3.15it/s]

 45%|████▍     | 79/177 [00:27<00:32,  3.05it/s]

 45%|████▌     | 80/177 [00:27<00:30,  3.18it/s]

 46%|████▌     | 81/177 [00:28<00:30,  3.19it/s]

 46%|████▋     | 82/177 [00:28<00:30,  3.12it/s]

 47%|████▋     | 83/177 [00:28<00:38,  2.44it/s]

 47%|████▋     | 84/177 [00:29<00:35,  2.59it/s]

 48%|████▊     | 85/177 [00:29<00:27,  3.31it/s]

 49%|████▉     | 87/177 [00:29<00:18,  4.91it/s]

 50%|█████     | 89/177 [00:29<00:14,  6.16it/s]

 51%|█████▏    | 91/177 [00:29<00:10,  7.96it/s]

 53%|█████▎    | 93/177 [00:30<00:13,  6.36it/s]

 53%|█████▎    | 94/177 [00:30<00:17,  4.79it/s]

 54%|█████▎    | 95/177 [00:31<00:23,  3.54it/s]

 54%|█████▍    | 96/177 [00:31<00:25,  3.21it/s]

 55%|█████▍    | 97/177 [00:32<00:27,  2.86it/s]

 55%|█████▌    | 98/177 [00:32<00:30,  2.58it/s]

 56%|█████▌    | 99/177 [00:33<00:35,  2.22it/s]

 56%|█████▋    | 100/177 [00:33<00:36,  2.10it/s]

 57%|█████▋    | 101/177 [00:34<00:37,  2.03it/s]

 58%|█████▊    | 102/177 [00:34<00:35,  2.12it/s]

 58%|█████▊    | 103/177 [00:35<00:31,  2.35it/s]

 59%|█████▉    | 104/177 [00:35<00:26,  2.74it/s]

 59%|█████▉    | 105/177 [00:35<00:27,  2.66it/s]

 60%|█████▉    | 106/177 [00:36<00:26,  2.70it/s]

 60%|██████    | 107/177 [00:36<00:23,  2.94it/s]

 62%|██████▏   | 109/177 [00:36<00:14,  4.59it/s]

 63%|██████▎   | 111/177 [00:36<00:14,  4.70it/s]

 63%|██████▎   | 112/177 [00:37<00:14,  4.36it/s]

 64%|██████▍   | 113/177 [00:37<00:17,  3.60it/s]

 64%|██████▍   | 114/177 [00:37<00:18,  3.39it/s]

 65%|██████▍   | 115/177 [00:38<00:18,  3.43it/s]

 66%|██████▌   | 116/177 [00:38<00:16,  3.66it/s]

 66%|██████▌   | 117/177 [00:38<00:13,  4.40it/s]

 67%|██████▋   | 118/177 [00:38<00:14,  4.10it/s]

 67%|██████▋   | 119/177 [00:39<00:19,  2.98it/s]

 68%|██████▊   | 120/177 [00:39<00:18,  3.10it/s]

 68%|██████▊   | 121/177 [00:39<00:14,  3.81it/s]

 69%|██████▉   | 123/177 [00:40<00:11,  4.74it/s]

 70%|███████   | 124/177 [00:40<00:12,  4.09it/s]

 71%|███████   | 125/177 [00:41<00:17,  3.00it/s]

 71%|███████   | 126/177 [00:41<00:17,  2.99it/s]

 72%|███████▏  | 127/177 [00:41<00:18,  2.78it/s]

 72%|███████▏  | 128/177 [00:42<00:17,  2.87it/s]

 73%|███████▎  | 129/177 [00:42<00:16,  2.86it/s]

 73%|███████▎  | 130/177 [00:42<00:18,  2.58it/s]

 74%|███████▍  | 131/177 [00:43<00:19,  2.42it/s]

 75%|███████▍  | 132/177 [00:43<00:15,  2.85it/s]

 75%|███████▌  | 133/177 [00:43<00:14,  2.94it/s]

 76%|███████▌  | 134/177 [00:44<00:16,  2.63it/s]

 76%|███████▋  | 135/177 [00:44<00:15,  2.73it/s]

 77%|███████▋  | 137/177 [00:44<00:09,  4.31it/s]

 79%|███████▊  | 139/177 [00:45<00:06,  5.69it/s]

 79%|███████▉  | 140/177 [00:45<00:07,  5.24it/s]

 80%|███████▉  | 141/177 [00:45<00:09,  3.73it/s]

 80%|████████  | 142/177 [00:46<00:11,  3.17it/s]

 81%|████████  | 143/177 [00:46<00:12,  2.66it/s]

 81%|████████▏ | 144/177 [00:47<00:13,  2.51it/s]

 82%|████████▏ | 145/177 [00:47<00:14,  2.20it/s]

 82%|████████▏ | 146/177 [00:48<00:14,  2.15it/s]

 83%|████████▎ | 147/177 [00:49<00:15,  1.98it/s]

 84%|████████▎ | 148/177 [00:49<00:15,  1.93it/s]

 84%|████████▍ | 149/177 [00:49<00:13,  2.02it/s]

 85%|████████▌ | 151/177 [00:50<00:09,  2.61it/s]

 86%|████████▌ | 152/177 [00:50<00:09,  2.65it/s]

 86%|████████▋ | 153/177 [00:51<00:08,  2.88it/s]

 87%|████████▋ | 154/177 [00:51<00:07,  3.25it/s]

 88%|████████▊ | 155/177 [00:51<00:05,  3.87it/s]

 89%|████████▊ | 157/177 [00:51<00:03,  5.42it/s]

 89%|████████▉ | 158/177 [00:51<00:03,  4.81it/s]

 90%|████████▉ | 159/177 [00:52<00:04,  4.33it/s]

 91%|█████████ | 161/177 [00:52<00:02,  5.97it/s]

 92%|█████████▏| 163/177 [00:52<00:02,  6.41it/s]

 93%|█████████▎| 164/177 [00:53<00:02,  4.89it/s]

 93%|█████████▎| 165/177 [00:53<00:03,  3.44it/s]

 94%|█████████▍| 166/177 [00:54<00:03,  2.92it/s]

 94%|█████████▍| 167/177 [00:54<00:03,  2.92it/s]

 95%|█████████▍| 168/177 [00:54<00:03,  2.71it/s]

 95%|█████████▌| 169/177 [00:55<00:03,  2.62it/s]

 96%|█████████▌| 170/177 [00:55<00:02,  2.78it/s]

 97%|█████████▋| 171/177 [00:55<00:02,  2.85it/s]

 97%|█████████▋| 172/177 [00:56<00:01,  2.78it/s]

 98%|█████████▊| 173/177 [00:56<00:01,  2.74it/s]

 99%|█████████▉| 175/177 [00:56<00:00,  4.21it/s]

 99%|█████████▉| 176/177 [00:57<00:00,  3.90it/s]

100%|██████████| 177/177 [00:57<00:00,  3.65it/s]
                                                     

                                                 

 35%|███▍      | 300/862 [1:43:07<2:31:03, 16.13s/it]

100%|██████████| 177/177 [00:57<00:00,  3.65it/s]

                                                 wandb: WARNING URL not available in offline run
`rope_scaling`'s beta_fast field must be a float, got 32
`rope_scaling`'s beta_slow field must be a float, got 1

 35%|███▍      | 301/862 [1:43:27<5:24:50, 34.74s/it]
 35%|███▌      | 302/862 [1:43:40<4:24:37, 28.35s/it]
 35%|███▌      | 303/862 [1:43:58<3:53:09, 25.03s/it]
 35%|███▌      | 304/862 [1:44:19<3:43:30, 24.03s/it]
 35%|███▌      | 305/862 [1:44:35<3:20:08, 21.56s/it]
                                                     

 35%|███▌      | 305/862 [1:44:35<3:20:08, 21.56s/it]
 35%|███▌      | 306/862 [1:45:12<4:01:19, 26.04s/it]
 36%|███▌      | 307/862 [1:45:45<4:21:35, 28.28s/it]
 36%|███▌      | 308/862 [1:46:11<4:14:09, 27.53s/it]
 36%|███▌      | 309/862 [1:46:30<3:49:33, 24.91s/it]
 36%|███▌      | 310/862 [1:46:43<3:16:37, 21.37s/it]
                                                     

 36%|███▌      | 310/862 [1:46:43<3:16:37, 21.37s/it]
 36%|███▌      | 311/862 [1:47:05<3:17:32, 21.51s/it]
 36%|███▌      | 312/862 [1:47:22<3:05:00, 20.18s/it]
 36%|███▋      | 313/862 [1:47:46<3:17:01, 21.53s/it]
 36%|███▋      | 314/862 [1:48:01<2:56:22, 19.31s/it]
 37%|███▋      | 315/862 [1:48:23<3:04:38, 20.25s/it]
                                                     

 37%|███▋      | 315/862 [1:48:23<3:04:38, 20.25s/it]
 37%|███▋      | 316/862 [1:48:58<3:44:05, 24.62s/it]
 37%|███▋      | 317/862 [1:49:16<3:27:16, 22.82s/it]
 37%|███▋      | 318/862 [1:49:39<3:27:24, 22.88s/it]
 37%|███▋      | 319/862 [1:49:55<3:05:59, 20.55s/it]
 37%|███▋      | 320/862 [1:50:17<3:11:31, 21.20s/it]
                                                     

 37%|███▋      | 320/862 [1:50:17<3:11:31, 21.20s/it]
 37%|███▋      | 321/862 [1:50:42<3:19:39, 22.14s/it]
 37%|███▋      | 322/862 [1:51:01<3:11:48, 21.31s/it]
 37%|███▋      | 323/862 [1:51:27<3:24:06, 22.72s/it]Terminated