Training in progress, step 23500, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:829898806c2cf797a60646c5ed75fbf0d1a577e9d79be79d379be1431ad4345b
+oid sha256:47666e73d822697d81bbfccb0535c0d56782b13d9471c108cfb5f708ab05a025
 size 36730224

last-checkpoint/ar_diffusion_info.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:314892f8ce2942ea9f92638fd38beb1039385352c76b29835054762eadd01ceb
+oid sha256:8b33800467cd4e0ea352bddd3d6bc949226fd682d012e035e3cd5842672b679a
 size 1544

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:332c8cdea66a1010f828c6ecd205e8e1ab1047e6bc38f9436e154fb2f1bf959d
+oid sha256:f1890e293af670b81b954d87d565c793549dd2a93009af624045ee0011125001
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f21304ce44e93f8f86da1b431eb1e188b0a7d5ce22c8a8c84f5d679245daaffa
+oid sha256:9cb49390cd54c8aca2233c64b9e33d7396d51837a032de9f6d8f9419982ba2ef
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e49d1b644e105b5cfb9b398b2668e7af4b7561d44b430dc9a36c09d64fe8801
+oid sha256:b4b01d3ddac56b25619c1ead241e352b13330923fdc417898657168ee04f3af4
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9322bae8c5f36989bed1e15319c91ee4f9f6797c8ed15e1827437e61c8ea85d
+oid sha256:e8a75b227d1139c1950381f2862e7e3504b8ee20dbb6bf732f2c5e660ca929c9
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 22250,
   "best_metric": 1.4596961736679077,
   "best_model_checkpoint": "./ar-diffusion-checkpoints-fixed/checkpoint-21000",
-  "epoch": 1.7690946850242288,
+  "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": 56.616,
       "eval_steps_per_second": 14.154,
       "step": 23000
+    },
+    {
+      "epoch": 1.7729405430351513,
+      "grad_norm": 1.6651790142059326,
+      "learning_rate": 8.292860296600265e-05,
+      "loss": 1.4537,
+      "step": 23050
+    },
+    {
+      "epoch": 1.7767864010460732,
+      "grad_norm": 1.5910587310791016,
+      "learning_rate": 8.266888294418618e-05,
+      "loss": 1.5305,
+      "step": 23100
+    },
+    {
+      "epoch": 1.7806322590569956,
+      "grad_norm": 1.2711199522018433,
+      "learning_rate": 8.240916292236969e-05,
+      "loss": 1.4877,
+      "step": 23150
+    },
+    {
+      "epoch": 1.7844781170679178,
+      "grad_norm": 1.7133463621139526,
+      "learning_rate": 8.21494429005532e-05,
+      "loss": 1.4547,
+      "step": 23200
+    },
+    {
+      "epoch": 1.78832397507884,
+      "grad_norm": 1.6850301027297974,
+      "learning_rate": 8.188972287873673e-05,
+      "loss": 1.4228,
+      "step": 23250
+    },
+    {
+      "epoch": 1.78832397507884,
+      "eval_loss": 1.4735645055770874,
+      "eval_runtime": 17.7983,
+      "eval_samples_per_second": 56.185,
+      "eval_steps_per_second": 14.046,
+      "step": 23250
+    },
+    {
+      "epoch": 1.7921698330897624,
+      "grad_norm": 1.1047898530960083,
+      "learning_rate": 8.163000285692024e-05,
+      "loss": 1.4817,
+      "step": 23300
+    },
+    {
+      "epoch": 1.7960156911006846,
+      "grad_norm": 1.8970743417739868,
+      "learning_rate": 8.137028283510376e-05,
+      "loss": 1.4884,
+      "step": 23350
+    },
+    {
+      "epoch": 1.7998615491116068,
+      "grad_norm": 1.4730740785598755,
+      "learning_rate": 8.111056281328728e-05,
+      "loss": 1.4436,
+      "step": 23400
+    },
+    {
+      "epoch": 1.8037074071225292,
+      "grad_norm": 1.8288697004318237,
+      "learning_rate": 8.08508427914708e-05,
+      "loss": 1.4614,
+      "step": 23450
+    },
+    {
+      "epoch": 1.8075532651334512,
+      "grad_norm": 1.2339516878128052,
+      "learning_rate": 8.059112276965432e-05,
+      "loss": 1.4276,
+      "step": 23500
+    },
+    {
+      "epoch": 1.8075532651334512,
+      "eval_loss": 1.4655163288116455,
+      "eval_runtime": 17.9781,
+      "eval_samples_per_second": 55.623,
+      "eval_steps_per_second": 13.906,
+      "step": 23500
     }
   ],
   "logging_steps": 50,