Training in progress, step 25500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31864ac4bb9726b9569409ab5161fad26e94cbf3e7dd95686402bacb97d569e6
+oid sha256:8492be170f7ead3e1de3a8a4310c2bd0ea5b6f9e4d86bcd7d0ba2698651e52bd
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7d52ac292e08ab1865944b2cee1c9d3b2be89e70ca0ae0fa8ad99b662c99bf03
+oid sha256:ab9a2b23fe7fd93945a85cfc1259da800f7071f45eb31fd4ed7e630b682c35ca
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:306ba75cd660966ab767dfb4cbb174b1b61b5756f39410aac084e39a57af127d
+oid sha256:32da047c4d4b572ed1b8f48061cec167cdf94b2d474f921721549d72ae412177
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f580d496f0048fe1cffdac35f5b2b7b4ede7ac0df721e48429fe4715fc745efa
+oid sha256:fa86ead0557145b8c2ea7bbd3e68631cb5af19c8565a5c11d759d3e4217c56ab
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d0369b7df9f4f6057f54d12203a572c70823bfca8ac268e71fe1a019e47a330
+oid sha256:98ac3712aca76800ec0076b19c413d355103c14ed7b94cc79d36faba9193644b
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c9b07666b0c0a9f4236eb42880b6e9eedd3084d2bd56250a745ae04b4f3412ef
+oid sha256:394944f715215f9b5f72270954d983bc0552f0bf05252a3f3b1abae560af7063
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 24750,
-  "best_metric": 0.7527515888214111,
+  "best_global_step": 25250,
+  "best_metric": 0.7520478963851929,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-24500",
-  "epoch": 1.9229290054611183,
+  "epoch": 1.9613875855703409,
   "eval_steps": 250,
-  "global_step": 25000,
+  "global_step": 25500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4308,6 +4308,92 @@
       "eval_samples_per_second": 55.873,
       "eval_steps_per_second": 13.968,
       "step": 25000
+    },
+    {
+      "epoch": 1.9267748634720405,
+      "grad_norm": 0.6314563751220703,
+      "learning_rate": 7.25242188920344e-05,
+      "loss": 0.8278,
+      "step": 25050
+    },
+    {
+      "epoch": 1.930620721482963,
+      "grad_norm": 0.38249602913856506,
+      "learning_rate": 7.22644988702179e-05,
+      "loss": 0.7948,
+      "step": 25100
+    },
+    {
+      "epoch": 1.9344665794938851,
+      "grad_norm": 0.8241211771965027,
+      "learning_rate": 7.200477884840142e-05,
+      "loss": 0.8152,
+      "step": 25150
+    },
+    {
+      "epoch": 1.9383124375048073,
+      "grad_norm": 0.4248273968696594,
+      "learning_rate": 7.175025322702127e-05,
+      "loss": 0.8083,
+      "step": 25200
+    },
+    {
+      "epoch": 1.9421582955157297,
+      "grad_norm": 1.0574986934661865,
+      "learning_rate": 7.14905332052048e-05,
+      "loss": 0.7644,
+      "step": 25250
+    },
+    {
+      "epoch": 1.9421582955157297,
+      "eval_loss": 0.7520478963851929,
+      "eval_runtime": 17.9125,
+      "eval_samples_per_second": 55.827,
+      "eval_steps_per_second": 13.957,
+      "step": 25250
+    },
+    {
+      "epoch": 1.9460041535266517,
+      "grad_norm": 0.957831084728241,
+      "learning_rate": 7.123081318338831e-05,
+      "loss": 0.8145,
+      "step": 25300
+    },
+    {
+      "epoch": 1.949850011537574,
+      "grad_norm": 0.7300383448600769,
+      "learning_rate": 7.097109316157184e-05,
+      "loss": 0.8021,
+      "step": 25350
+    },
+    {
+      "epoch": 1.9536958695484963,
+      "grad_norm": 0.6103696227073669,
+      "learning_rate": 7.071137313975535e-05,
+      "loss": 0.7561,
+      "step": 25400
+    },
+    {
+      "epoch": 1.9575417275594185,
+      "grad_norm": 0.9353188276290894,
+      "learning_rate": 7.045165311793886e-05,
+      "loss": 0.76,
+      "step": 25450
+    },
+    {
+      "epoch": 1.9613875855703409,
+      "grad_norm": 0.7097103595733643,
+      "learning_rate": 7.019193309612239e-05,
+      "loss": 0.8125,
+      "step": 25500
+    },
+    {
+      "epoch": 1.9613875855703409,
+      "eval_loss": 0.7521655559539795,
+      "eval_runtime": 17.8845,
+      "eval_samples_per_second": 55.914,
+      "eval_steps_per_second": 13.979,
+      "step": 25500
     }
   ],
   "logging_steps": 50,