Training in progress, step 36500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74ae96a07f0a1364fae979dc8192670789abe5ddf742fd023b1f7e6cff9f0baf
+oid sha256:00da868c16026f8b91fd852d53e5b1e8b84bd1c76aefa3ce6543ff5082c8a8c1
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14c071741a7326f16a83b0de63af44ed2b3c7961447b59e277e8e3f8fd62b865
+oid sha256:04e0ccc68b635616af9e0dbead055411d758f4619095d018f21601b71ae1dad9
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae2cee13b4f8b39a57ec4a0861bd7b8c66161ea17bdf36a9f9ee8d1ce1e2759c
+oid sha256:81484f3b03bf089df3397a87ac9d9ffd5e1f946de717b51932df743ee8d9f915
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6591f94e486820dc9de5c16ae67ac55b097fbd6c09cd7885ab8396b22d79615
+oid sha256:5d10178046632fe47bb7008cd9e1a9974e0ea49262160950299e919075603393
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f183c8ebab3f56f9652055ab8d2536f1affa446052e58c49d6e1ac9410f236aa
+oid sha256:81382d5bc209ea226a1381f99c5863a7a3af33f40b32204afd2996013b6f09c3
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:906c43e30b9edc8f130b3f77317e01f206c6e5f1267c22ca4899b9c5968ba8ca
+oid sha256:0cf91a495853427b5f841ab1be3c35766ad0edbc12ff06e2c15160a47811ed22
 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": 2.7690177678640104,
+  "epoch": 2.8074763479732328,
   "eval_steps": 250,
-  "global_step": 36000,
+  "global_step": 36500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -6200,6 +6200,92 @@
       "eval_samples_per_second": 57.898,
       "eval_steps_per_second": 14.474,
       "step": 36000
+    },
+    {
+      "epoch": 2.772863625874933,
+      "grad_norm": 0.5958703756332397,
+      "learning_rate": 1.5422174895462695e-05,
+      "loss": 0.9503,
+      "step": 36050
+    },
+    {
+      "epoch": 2.776709483885855,
+      "grad_norm": 1.1087392568588257,
+      "learning_rate": 1.516245487364621e-05,
+      "loss": 0.9725,
+      "step": 36100
+    },
+    {
+      "epoch": 2.780555341896777,
+      "grad_norm": 1.2012007236480713,
+      "learning_rate": 1.4902734851829727e-05,
+      "loss": 1.0098,
+      "step": 36150
+    },
+    {
+      "epoch": 2.784401199907699,
+      "grad_norm": 0.9333285093307495,
+      "learning_rate": 1.4643014830013247e-05,
+      "loss": 0.9764,
+      "step": 36200
+    },
+    {
+      "epoch": 2.7882470579186216,
+      "grad_norm": 0.9407594203948975,
+      "learning_rate": 1.4383294808196765e-05,
+      "loss": 0.9404,
+      "step": 36250
+    },
+    {
+      "epoch": 2.7882470579186216,
+      "eval_loss": 0.9894633889198303,
+      "eval_runtime": 17.3239,
+      "eval_samples_per_second": 57.724,
+      "eval_steps_per_second": 14.431,
+      "step": 36250
+    },
+    {
+      "epoch": 2.792092915929544,
+      "grad_norm": 0.9794307947158813,
+      "learning_rate": 1.4123574786380284e-05,
+      "loss": 0.9445,
+      "step": 36300
+    },
+    {
+      "epoch": 2.795938773940466,
+      "grad_norm": 1.1447358131408691,
+      "learning_rate": 1.3863854764563799e-05,
+      "loss": 1.0056,
+      "step": 36350
+    },
+    {
+      "epoch": 2.7997846319513884,
+      "grad_norm": 0.8741857409477234,
+      "learning_rate": 1.3604134742747318e-05,
+      "loss": 0.9511,
+      "step": 36400
+    },
+    {
+      "epoch": 2.803630489962311,
+      "grad_norm": 1.0769715309143066,
+      "learning_rate": 1.3344414720930836e-05,
+      "loss": 0.9789,
+      "step": 36450
+    },
+    {
+      "epoch": 2.8074763479732328,
+      "grad_norm": 1.1045129299163818,
+      "learning_rate": 1.3084694699114356e-05,
+      "loss": 0.9958,
+      "step": 36500
+    },
+    {
+      "epoch": 2.8074763479732328,
+      "eval_loss": 0.9914500117301941,
+      "eval_runtime": 17.3793,
+      "eval_samples_per_second": 57.54,
+      "eval_steps_per_second": 14.385,
+      "step": 36500
     }
   ],
   "logging_steps": 50,