Training in progress, step 23500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:607761a68d936fac3569465d3f0d20419715827f9c4331a285faee67e09f155c
+oid sha256:e5e8c57b0540b43477103991d94a0031665bde117b5025873ffe2f31a08ca029
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:80833488aa86fb4cba830c432f5aa3abb1b8d8ff38f3b1e8b83ce7834612b917
+oid sha256:dcb60822099d74359c979b9b670c4aabd0f2ec34f416894d009bb290edd18fb0
 size 1736

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4035be8d85b44dc1d28bbbd28b0c42e51f7f77114f423b7a9f9fd7587aca606
+oid sha256:a87d8a39dc457bb1ae5a270a42ba3d23dcea31e23f4fc916e1a9528a3ddc6ce1
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ce24974df420c8b58ce59acb329c359d968e0b98672cd1798e7bedf4f10e13c3
-size 14244
+oid sha256:72244ff2f8e1c50254cff18940eace7cf87f75ca483c4693d1f58d7d7fc6e225
+size 14180

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36d8e43084f8796853eb654313d4c1bc1085274ed04c979e93b6fe0324b7620f
+oid sha256:1455d8897bfe51e2e7ebee8ab560ed718f3a24db9dec5faff5c72296032decc1
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e50d97e6b844f88427c38d3f0fdd176e281de54a4fb5c4780fea01f20111dfc1
+oid sha256:3a48972742d0232c3fc2605c286832dbdeaceae1cad253f23d64ff38411b92e0
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 23000,
-  "best_metric": 0.7635026574134827,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-23000",
-  "epoch": 1.7690946850242288,
+  "best_global_step": 23500,
+  "best_metric": 0.7594859600067139,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-23500",
+  "epoch": 1.8075532651334512,
   "eval_steps": 250,
-  "global_step": 23000,
+  "global_step": 23500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3964,6 +3964,92 @@
       "eval_samples_per_second": 55.875,
       "eval_steps_per_second": 13.969,
       "step": 23000
+    },
+    {
+      "epoch": 1.7729405430351513,
+      "grad_norm": 0.7365099787712097,
+      "learning_rate": 8.291301976469366e-05,
+      "loss": 0.7915,
+      "step": 23050
+    },
+    {
+      "epoch": 1.7767864010460732,
+      "grad_norm": 0.7158268690109253,
+      "learning_rate": 8.265329974287718e-05,
+      "loss": 0.8089,
+      "step": 23100
+    },
+    {
+      "epoch": 1.7806322590569956,
+      "grad_norm": 0.7917172312736511,
+      "learning_rate": 8.23935797210607e-05,
+      "loss": 0.7881,
+      "step": 23150
+    },
+    {
+      "epoch": 1.7844781170679178,
+      "grad_norm": 0.9002280831336975,
+      "learning_rate": 8.213385969924422e-05,
+      "loss": 0.7861,
+      "step": 23200
+    },
+    {
+      "epoch": 1.78832397507884,
+      "grad_norm": 1.118498682975769,
+      "learning_rate": 8.187413967742774e-05,
+      "loss": 0.7787,
+      "step": 23250
+    },
+    {
+      "epoch": 1.78832397507884,
+      "eval_loss": 0.7597912549972534,
+      "eval_runtime": 18.0126,
+      "eval_samples_per_second": 55.517,
+      "eval_steps_per_second": 13.879,
+      "step": 23250
+    },
+    {
+      "epoch": 1.7921698330897624,
+      "grad_norm": 0.555014967918396,
+      "learning_rate": 8.161441965561125e-05,
+      "loss": 0.7698,
+      "step": 23300
+    },
+    {
+      "epoch": 1.7960156911006846,
+      "grad_norm": 0.7749983072280884,
+      "learning_rate": 8.135469963379477e-05,
+      "loss": 0.7976,
+      "step": 23350
+    },
+    {
+      "epoch": 1.7998615491116068,
+      "grad_norm": 0.8833787441253662,
+      "learning_rate": 8.10949796119783e-05,
+      "loss": 0.7901,
+      "step": 23400
+    },
+    {
+      "epoch": 1.8037074071225292,
+      "grad_norm": 1.099992036819458,
+      "learning_rate": 8.08352595901618e-05,
+      "loss": 0.7843,
+      "step": 23450
+    },
+    {
+      "epoch": 1.8075532651334512,
+      "grad_norm": 0.7118529677391052,
+      "learning_rate": 8.057553956834533e-05,
+      "loss": 0.7818,
+      "step": 23500
+    },
+    {
+      "epoch": 1.8075532651334512,
+      "eval_loss": 0.7594859600067139,
+      "eval_runtime": 17.8501,
+      "eval_samples_per_second": 56.022,
+      "eval_steps_per_second": 14.005,
+      "step": 23500
     }
   ],
   "logging_steps": 50,