Training in progress, step 35500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e859ba9561b55c4713545dc47c401bfb524b089ca9535702dcd1494059b3820
+oid sha256:c4d51eeeda51bc19f5c6d7222c3569a8b30c8252006cab154f166725a5d6a0a8
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f14825ac12e33d198b451f7e856b892bba4b3f9adb0c77a6ed825bd184145a9
+oid sha256:fc1b1084a32b0cff2b333748a7e838a8590ff81e228b7c3e840ae1cbbc113c15
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f37ff6c2cf1b8fc28fe79a96b16d82c777a40f9ae1a62213e0d40ea69913852
+oid sha256:86e904a91e2cb70946038bcc4ef0f399dffd5e68a67a3db4e06d21064557dbb3
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:420ba1bac282b79318028dd15898492f4873416a2a773ea36e1d51488979f42b
+oid sha256:c53689ea6f99d2293000a099d231d87ec2b98e3ae16c084822ed187b69c19695
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:512c2f45e554afd2937a0615962c874313642439d4843c36db50a18a20e03902
+oid sha256:c0b69fcd3d6df327b4f7a7f693a8457e7a615ee5a38d8957008f6b6d0162dd3f
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cc40176b44fff927c20f1bac9b67d6ff31ea020f24aee448725b6d82a306f911
+oid sha256:f72dabb66926d5f103787feb6d06e607565f49ca5105e75822f116bfd973c31b
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 34000,
   "best_metric": 0.987713634967804,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-34000",
-  "epoch": 2.692100607645566,
+  "epoch": 2.730559187754788,
   "eval_steps": 250,
-  "global_step": 35000,
+  "global_step": 35500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -6028,6 +6028,92 @@
       "eval_samples_per_second": 57.95,
       "eval_steps_per_second": 14.488,
       "step": 35000
+    },
+    {
+      "epoch": 2.6959464656564878,
+      "grad_norm": 1.3044216632843018,
+      "learning_rate": 2.0611380931356e-05,
+      "loss": 0.9787,
+      "step": 35050
+    },
+    {
+      "epoch": 2.69979232366741,
+      "grad_norm": 1.4273097515106201,
+      "learning_rate": 2.0351660909539518e-05,
+      "loss": 0.9775,
+      "step": 35100
+    },
+    {
+      "epoch": 2.7036381816783326,
+      "grad_norm": 1.6097638607025146,
+      "learning_rate": 2.0091940887723034e-05,
+      "loss": 0.9926,
+      "step": 35150
+    },
+    {
+      "epoch": 2.7074840396892546,
+      "grad_norm": 1.4893895387649536,
+      "learning_rate": 1.9832220865906553e-05,
+      "loss": 1.0435,
+      "step": 35200
+    },
+    {
+      "epoch": 2.711329897700177,
+      "grad_norm": 1.1620233058929443,
+      "learning_rate": 1.9572500844090073e-05,
+      "loss": 0.9525,
+      "step": 35250
+    },
+    {
+      "epoch": 2.711329897700177,
+      "eval_loss": 0.9926208257675171,
+      "eval_runtime": 17.3977,
+      "eval_samples_per_second": 57.479,
+      "eval_steps_per_second": 14.37,
+      "step": 35250
+    },
+    {
+      "epoch": 2.7151757557110994,
+      "grad_norm": 1.2087517976760864,
+      "learning_rate": 1.931278082227359e-05,
+      "loss": 1.0005,
+      "step": 35300
+    },
+    {
+      "epoch": 2.7190216137220213,
+      "grad_norm": 0.8885460495948792,
+      "learning_rate": 1.9058255200893437e-05,
+      "loss": 0.9911,
+      "step": 35350
+    },
+    {
+      "epoch": 2.7228674717329437,
+      "grad_norm": 0.9543077945709229,
+      "learning_rate": 1.8798535179076957e-05,
+      "loss": 1.0437,
+      "step": 35400
+    },
+    {
+      "epoch": 2.7267133297438657,
+      "grad_norm": 1.2362306118011475,
+      "learning_rate": 1.8538815157260476e-05,
+      "loss": 0.9766,
+      "step": 35450
+    },
+    {
+      "epoch": 2.730559187754788,
+      "grad_norm": 1.1946227550506592,
+      "learning_rate": 1.8279095135443993e-05,
+      "loss": 1.0333,
+      "step": 35500
+    },
+    {
+      "epoch": 2.730559187754788,
+      "eval_loss": 0.9953948855400085,
+      "eval_runtime": 17.1888,
+      "eval_samples_per_second": 58.178,
+      "eval_steps_per_second": 14.544,
+      "step": 35500
     }
   ],
   "logging_steps": 50,