Training in progress, step 28500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62439bae41ea1998b90f6037dd9fa884cf659248f0c85d5c8a744390c2886fa9
+oid sha256:f6600bba0d4ff6b96aebe142a8f1d29d3689bd6d93b2ebd674227ef088bff7b8
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0db08bb0e64e6fdc76f068a4dce33c014fc080b283b9a64c9a60ab6193a7f49
+oid sha256:bf9bcfbac04fd11bcd15980c9df2ea854394b3bb5d97d731829d75371fbaa955
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:370e87188f637f83ed8349bf7a8b60a845c5cbc7b0913ca617016b425f823961
+oid sha256:72afdb8f3d5d0bd15446a4f8cd0fbad5718bb8f7796245e8f6d72c43db0cdada
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0f0206c699fd966569a4deff507474ca9114467664863e2a7c8efd89e088e4c4
+oid sha256:25bf93d553915ff8c50a7931af03f2d7b533b6db51da15a573325147ba05e268
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a80611879827bf00dd9a4f888e494a8a2af0eea4eada726997ec5489de20e821
+oid sha256:5bfb7a12fdede60278d2b32e52f50469e6e77b619720bcb0e4bd039e0a83e832
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:59377233a54e779461bcaa1b904cc8d8243d8ae8f5afd52f30216a544a19a619
+oid sha256:e4e46f9024365cd773095a35799f9dd86ead6c135c47c788f2a21bedfef6883c
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 27750,
-  "best_metric": 0.736056387424469,
+  "best_global_step": 28250,
+  "best_metric": 0.7313055396080017,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-27500",
-  "epoch": 2.1536804861164525,
+  "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": 55.73,
       "eval_steps_per_second": 13.933,
       "step": 28000
+    },
+    {
+      "epoch": 2.157526344127375,
+      "grad_norm": 0.9337176084518433,
+      "learning_rate": 5.6956600784354464e-05,
+      "loss": 0.7844,
+      "step": 28050
+    },
+    {
+      "epoch": 2.1613722021382973,
+      "grad_norm": 0.7867174744606018,
+      "learning_rate": 5.669688076253798e-05,
+      "loss": 0.7634,
+      "step": 28100
+    },
+    {
+      "epoch": 2.1652180601492192,
+      "grad_norm": 0.6526890397071838,
+      "learning_rate": 5.643716074072151e-05,
+      "loss": 0.743,
+      "step": 28150
+    },
+    {
+      "epoch": 2.1690639181601417,
+      "grad_norm": 1.1720079183578491,
+      "learning_rate": 5.617744071890503e-05,
+      "loss": 0.7165,
+      "step": 28200
+    },
+    {
+      "epoch": 2.1729097761710636,
+      "grad_norm": 1.029062032699585,
+      "learning_rate": 5.5917720697088535e-05,
+      "loss": 0.6854,
+      "step": 28250
+    },
+    {
+      "epoch": 2.1729097761710636,
+      "eval_loss": 0.7313055396080017,
+      "eval_runtime": 18.0992,
+      "eval_samples_per_second": 55.251,
+      "eval_steps_per_second": 13.813,
+      "step": 28250
+    },
+    {
+      "epoch": 2.176755634181986,
+      "grad_norm": 0.821443498134613,
+      "learning_rate": 5.565800067527206e-05,
+      "loss": 0.7209,
+      "step": 28300
+    },
+    {
+      "epoch": 2.1806014921929084,
+      "grad_norm": 0.6121924519538879,
+      "learning_rate": 5.539828065345558e-05,
+      "loss": 0.7371,
+      "step": 28350
+    },
+    {
+      "epoch": 2.1844473502038304,
+      "grad_norm": 0.8862821459770203,
+      "learning_rate": 5.5138560631639094e-05,
+      "loss": 0.684,
+      "step": 28400
+    },
+    {
+      "epoch": 2.188293208214753,
+      "grad_norm": 0.8110325336456299,
+      "learning_rate": 5.4878840609822614e-05,
+      "loss": 0.6843,
+      "step": 28450
+    },
+    {
+      "epoch": 2.192139066225675,
+      "grad_norm": 0.732751727104187,
+      "learning_rate": 5.461912058800613e-05,
+      "loss": 0.7229,
+      "step": 28500
+    },
+    {
+      "epoch": 2.192139066225675,
+      "eval_loss": 0.7322584986686707,
+      "eval_runtime": 18.025,
+      "eval_samples_per_second": 55.478,
+      "eval_steps_per_second": 13.87,
+      "step": 28500
     }
   ],
   "logging_steps": 50,