Training in progress, step 28500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:327ebc57ff4a2fce5b1ff51eacb255f502d310697ef3cac70837eee01705b907
+oid sha256:ba0434cac786c71024c01559b6b7850fe5dbda706c5abe96ea5f84b39fc25a36
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82ce8b4d70a4e5ffdc955b54b4e61ae42e280ad3043bd6c657a5a4f94d25406f
+oid sha256:f60e86ef96a6048e32cbd10d68a32028312c3eb4c1fe5334839b8d0872e9442a
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cefd1fadeee425e44de130ce43706a90f27357be2970c4c0e99ec3480bec2b93
+oid sha256:70f64c96b37df69761195ec83337cbb613282041735f5a1741c63ec0a536dca4
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:675e8eec5b71db621439753580d23e328c8c5e2da6333850e0095f28c2ae9e4d
+oid sha256:d5e1e01486e5f149b266e0eeec8533923aa61f54f33b3d0447fd4d42461ef77d
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:df1b3ae13e2a70e8f07dd9d7c74ac5ab9a7fad8c5d805908d310a8c82d08cad1
+oid sha256:cf44b476c67ba6f9ada6529ca5d65598e077bf4445258a31e4213eee858ee11e
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:14c5f190b4eb0a7cb5ff6d1c912de09299ca659cbcc34c660c71e8207c2fc146
+oid sha256:da67868605cfbee80d944d1a20284bdc251f41c0584c59c98da7e787a25cfe98
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 27750,
-  "best_metric": -30.463788986206055,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-27500",
-  "epoch": 2.1536804861164525,
+  "best_global_step": 28500,
+  "best_metric": -30.463890075683594,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-progressive-attention/checkpoint-28500",
+  "epoch": 2.192139066225675,
   "eval_steps": 250,
-  "global_step": 28000,
+  "global_step": 28500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -4824,6 +4824,92 @@
       "eval_samples_per_second": 59.412,
       "eval_steps_per_second": 14.853,
       "step": 28000
+    },
+    {
+      "epoch": 2.157526344127375,
+      "grad_norm": 1.7737715244293213,
+      "learning_rate": 5.697737838609979e-05,
+      "loss": -30.4579,
+      "step": 28050
+    },
+    {
+      "epoch": 2.1613722021382973,
+      "grad_norm": 2.3100192546844482,
+      "learning_rate": 5.671765836428331e-05,
+      "loss": -30.4582,
+      "step": 28100
+    },
+    {
+      "epoch": 2.1652180601492192,
+      "grad_norm": 2.7968549728393555,
+      "learning_rate": 5.645793834246682e-05,
+      "loss": -30.4588,
+      "step": 28150
+    },
+    {
+      "epoch": 2.1690639181601417,
+      "grad_norm": 1.6727192401885986,
+      "learning_rate": 5.619821832065034e-05,
+      "loss": -30.4585,
+      "step": 28200
+    },
+    {
+      "epoch": 2.1729097761710636,
+      "grad_norm": 1.8535244464874268,
+      "learning_rate": 5.593849829883386e-05,
+      "loss": -30.4589,
+      "step": 28250
+    },
+    {
+      "epoch": 2.1729097761710636,
+      "eval_loss": -30.463205337524414,
+      "eval_runtime": 16.861,
+      "eval_samples_per_second": 59.308,
+      "eval_steps_per_second": 14.827,
+      "step": 28250
+    },
+    {
+      "epoch": 2.176755634181986,
+      "grad_norm": 1.9916008710861206,
+      "learning_rate": 5.5678778277017375e-05,
+      "loss": -30.4585,
+      "step": 28300
+    },
+    {
+      "epoch": 2.1806014921929084,
+      "grad_norm": 1.7134790420532227,
+      "learning_rate": 5.5419058255200894e-05,
+      "loss": -30.4589,
+      "step": 28350
+    },
+    {
+      "epoch": 2.1844473502038304,
+      "grad_norm": 1.7053008079528809,
+      "learning_rate": 5.5159338233384414e-05,
+      "loss": -30.4587,
+      "step": 28400
+    },
+    {
+      "epoch": 2.188293208214753,
+      "grad_norm": 2.964104413986206,
+      "learning_rate": 5.489961821156793e-05,
+      "loss": -30.4586,
+      "step": 28450
+    },
+    {
+      "epoch": 2.192139066225675,
+      "grad_norm": 1.901634931564331,
+      "learning_rate": 5.4639898189751446e-05,
+      "loss": -30.4595,
+      "step": 28500
+    },
+    {
+      "epoch": 2.192139066225675,
+      "eval_loss": -30.463890075683594,
+      "eval_runtime": 16.9187,
+      "eval_samples_per_second": 59.106,
+      "eval_steps_per_second": 14.777,
+      "step": 28500
     }
   ],
   "logging_steps": 50,