Training in progress, step 27500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f6888cf33bfb468f85e5dcc21cc0f17b5b74bef4405ed14d519bb99a297d1c8b
+oid sha256:395d6f2f3db4017a2737909d6b4ea9e9031c52589c6f6d040f85bb9bc9a61574
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c7be3b232938e666a118057487d49dbdeac28f922d4f76b4b64c4ac34b65dec
+oid sha256:33bfe006e215fece89dfe8777344c7d6ce9d541fd4ce48109267ec6314d1a517
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b85d9af76ebb8d64bfb9efff12b696b6e9cc741ff3de19d27300fc289ad65d69
+oid sha256:b9df54078255f6bdb0657e3789fa1dc2e87d9dd9aedcc964174fb1c8597c5616
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:67321b1783cd5b265c92a36f00016768c3971f088d66dcd615ac78f9d75ae5ec
+oid sha256:547f82cfc12d87310740e670a337bffc2b27a4a004bc784a756302a28851318f
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2864a03b74f869a8e9c75a4e96db8943ce0e8fc7dd9949c7e834dc3430789e00
+oid sha256:aafb160e86ca3dae2151e7d5410c1f0f3d6a45d02cb075a2fd74acbbb6438223
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7dcbf4a011b987cda27e2ddb4d5d90c40d1d68052b9a56d8eccc78c40a20d31d
+oid sha256:00e4057ddb93d4201ababcbc4c1556bd476cd599690ef79c5880a1b8a05ecfad
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 26750,
-  "best_metric": 0.7409418821334839,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-26500",
-  "epoch": 2.076763325898008,
+  "best_global_step": 27500,
+  "best_metric": 0.7388279438018799,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-27500",
+  "epoch": 2.11522190600723,
   "eval_steps": 250,
-  "global_step": 27000,
+  "global_step": 27500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4652,6 +4652,92 @@
       "eval_samples_per_second": 55.444,
       "eval_steps_per_second": 13.861,
       "step": 27000
+    },
+    {
+      "epoch": 2.0806091839089302,
+      "grad_norm": 0.4836456775665283,
+      "learning_rate": 6.21510012206841e-05,
+      "loss": 0.7725,
+      "step": 27050
+    },
+    {
+      "epoch": 2.084455041919852,
+      "grad_norm": 0.9285927414894104,
+      "learning_rate": 6.189128119886762e-05,
+      "loss": 0.7167,
+      "step": 27100
+    },
+    {
+      "epoch": 2.0883008999307746,
+      "grad_norm": 0.8826911449432373,
+      "learning_rate": 6.163156117705115e-05,
+      "loss": 0.7204,
+      "step": 27150
+    },
+    {
+      "epoch": 2.0921467579416966,
+      "grad_norm": 1.0057801008224487,
+      "learning_rate": 6.137184115523465e-05,
+      "loss": 0.713,
+      "step": 27200
+    },
+    {
+      "epoch": 2.095992615952619,
+      "grad_norm": 1.2974227666854858,
+      "learning_rate": 6.111212113341818e-05,
+      "loss": 0.6974,
+      "step": 27250
+    },
+    {
+      "epoch": 2.095992615952619,
+      "eval_loss": 0.7402239441871643,
+      "eval_runtime": 17.8723,
+      "eval_samples_per_second": 55.953,
+      "eval_steps_per_second": 13.988,
+      "step": 27250
+    },
+    {
+      "epoch": 2.0998384739635414,
+      "grad_norm": 0.7047484517097473,
+      "learning_rate": 6.0852401111601696e-05,
+      "loss": 0.7377,
+      "step": 27300
+    },
+    {
+      "epoch": 2.1036843319744634,
+      "grad_norm": 1.0780186653137207,
+      "learning_rate": 6.0592681089785216e-05,
+      "loss": 0.7153,
+      "step": 27350
+    },
+    {
+      "epoch": 2.1075301899853858,
+      "grad_norm": 0.6844059228897095,
+      "learning_rate": 6.033296106796873e-05,
+      "loss": 0.7077,
+      "step": 27400
+    },
+    {
+      "epoch": 2.111376047996308,
+      "grad_norm": 0.7109357118606567,
+      "learning_rate": 6.007324104615225e-05,
+      "loss": 0.6574,
+      "step": 27450
+    },
+    {
+      "epoch": 2.11522190600723,
+      "grad_norm": 1.0174553394317627,
+      "learning_rate": 5.9813521024335775e-05,
+      "loss": 0.7269,
+      "step": 27500
+    },
+    {
+      "epoch": 2.11522190600723,
+      "eval_loss": 0.7388279438018799,
+      "eval_runtime": 17.9018,
+      "eval_samples_per_second": 55.86,
+      "eval_steps_per_second": 13.965,
+      "step": 27500
     }
   ],
   "logging_steps": 50,