Training in progress, step 1500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f02ec6b82a831f2902a910cf8fafe953154448b3b45acb102d57d8391c23b96
+oid sha256:45253cd6a22264fcd7ca6604e87717a83e19ed0d2d4435aba88387d4537d6529
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbe39982c4bebe4e0b3f0f76afe1c485909ad223aac0a865ef49d364896c3cff
+oid sha256:0b17ffeb44d19aff3ac9c819aa738ee5ab85c93083a940599d75b2a7ffda4f79
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c326c13c75092245145edf95faedf4bbc2e6d95bbc78269d58ccc6219fd83b99
+oid sha256:6ad00466a0acdb40237dae4c0b4283060a706960c994c5977eb55d8a845c8d60
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31bb3f2abd407f62d555fa53e113286139c6abb36d00dab486b238b592c7d71a
+oid sha256:97a66f532d27e837c455454684371b16caffae096e4b153335efaed54e929716
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2229a15bdf42c8dbae70a65f18e33cde5c16363745edce9ae10c3d4c94d5ed8e
+oid sha256:8167a7a628c5d6d5061531d993a99d8970329d2f198daf9e2bd5698793632aed
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d88837be00a1ec9be1702d7436c136fd03c05aff969a737326a957a88901e17
+oid sha256:8648ee281a6907a159c6d2a012bfd17e90caa435b3ae05a6accbf2bacc5a6ed9
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 1000,
-  "best_metric": 1.6695575714111328,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-1000",
-  "epoch": 0.07691716021844473,
+  "best_global_step": 1500,
+  "best_metric": 1.6279098987579346,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-1500",
+  "epoch": 0.1153757403276671,
   "eval_steps": 250,
-  "global_step": 1000,
+  "global_step": 1500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -180,6 +180,92 @@
       "eval_samples_per_second": 59.406,
       "eval_steps_per_second": 14.851,
       "step": 1000
+    },
+    {
+      "epoch": 0.08076301822936698,
+      "grad_norm": 2.8161075115203857,
+      "learning_rate": 0.00019716905176220034,
+      "loss": 1.6006,
+      "step": 1050
+    },
+    {
+      "epoch": 0.0846088762402892,
+      "grad_norm": 1.2744417190551758,
+      "learning_rate": 0.00019690933174038388,
+      "loss": 1.6306,
+      "step": 1100
+    },
+    {
+      "epoch": 0.08845473425121145,
+      "grad_norm": 1.0525975227355957,
+      "learning_rate": 0.0001966496117185674,
+      "loss": 1.6414,
+      "step": 1150
+    },
+    {
+      "epoch": 0.09230059226213368,
+      "grad_norm": 1.7405261993408203,
+      "learning_rate": 0.0001963898916967509,
+      "loss": 1.5997,
+      "step": 1200
+    },
+    {
+      "epoch": 0.09614645027305592,
+      "grad_norm": 2.664368152618408,
+      "learning_rate": 0.00019613017167493445,
+      "loss": 1.6865,
+      "step": 1250
+    },
+    {
+      "epoch": 0.09614645027305592,
+      "eval_loss": 1.653170108795166,
+      "eval_runtime": 16.8496,
+      "eval_samples_per_second": 59.348,
+      "eval_steps_per_second": 14.837,
+      "step": 1250
+    },
+    {
+      "epoch": 0.09999230828397816,
+      "grad_norm": 2.0688107013702393,
+      "learning_rate": 0.00019587045165311796,
+      "loss": 1.6575,
+      "step": 1300
+    },
+    {
+      "epoch": 0.10383816629490039,
+      "grad_norm": 0.8756892085075378,
+      "learning_rate": 0.00019561073163130145,
+      "loss": 1.6426,
+      "step": 1350
+    },
+    {
+      "epoch": 0.10768402430582263,
+      "grad_norm": 1.398770809173584,
+      "learning_rate": 0.000195351011609485,
+      "loss": 1.5807,
+      "step": 1400
+    },
+    {
+      "epoch": 0.11152988231674486,
+      "grad_norm": 2.2140140533447266,
+      "learning_rate": 0.0001950912915876685,
+      "loss": 1.5556,
+      "step": 1450
+    },
+    {
+      "epoch": 0.1153757403276671,
+      "grad_norm": 1.9329601526260376,
+      "learning_rate": 0.000194831571565852,
+      "loss": 1.6597,
+      "step": 1500
+    },
+    {
+      "epoch": 0.1153757403276671,
+      "eval_loss": 1.6279098987579346,
+      "eval_runtime": 16.9897,
+      "eval_samples_per_second": 58.859,
+      "eval_steps_per_second": 14.715,
+      "step": 1500
     }
   ],
   "logging_steps": 50,