Training in progress, step 22000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a15735cee6a081c034ee0a7775a5bd0e6cd549717f7509e7cac3a604e2b5ece6
+oid sha256:d7eaab29466400eb2a657f3436b2818198ebc3b9fd87cb963c622b5d81bbf4d9
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b1989d46f84770def08bc86fa4f365fd487bca1f5ce4a5f6b3c39adc01e8043b
+oid sha256:55f3c6089dbca62c8b2bebd6faf4edc5b1bb8a8a960c2f857f9b4984ed1a89d0
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24cdd6f45f9ebce887e06d548bf949bd18ccf52b17999d18752823b6dd7dc03a
-size 14308
+oid sha256:625da12a62ce1ec712e7ea0b4fc06eed36ff20b9634198d1bc6029989e807f8d
+size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1bb9f4faaa4ec0530a4b6c0b900d92b0246f7d7e281fc5a146d4979b5987fc8e
+oid sha256:46c97cfd73b672df8a5afb8d418c975440124b784cf73259cde1c1d1e99e483d
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c15c154285d68d5fe6fd0358f59f07ef7cc4e4e79592695b393ca4cfdda9f3f
+oid sha256:137a43b60deca6dd3da445bd48fbb2a7c159344a7a642ca673dc22a6f6abb915
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 21000,
-  "best_metric": 1.467063307762146,
-  "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-21000",
-  "epoch": 1.6537189446965619,
+  "best_global_step": 22000,
+  "best_metric": 1.4664525985717773,
+  "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-22000",
+  "epoch": 1.6921775248057842,
   "eval_steps": 250,
-  "global_step": 21500,
+  "global_step": 22000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3706,6 +3706,92 @@
       "eval_samples_per_second": 55.376,
       "eval_steps_per_second": 13.844,
       "step": 21500
+    },
+    {
+      "epoch": 1.657564802707484,
+      "grad_norm": 1.2780442237854004,
+      "learning_rate": 2.9661409507003417e-05,
+      "loss": 1.5179,
+      "step": 21550
+    },
+    {
+      "epoch": 1.6614106607184063,
+      "grad_norm": 1.206725835800171,
+      "learning_rate": 2.94040660269843e-05,
+      "loss": 1.4438,
+      "step": 21600
+    },
+    {
+      "epoch": 1.6652565187293287,
+      "grad_norm": 2.1834638118743896,
+      "learning_rate": 2.914672254696518e-05,
+      "loss": 1.4783,
+      "step": 21650
+    },
+    {
+      "epoch": 1.6691023767402506,
+      "grad_norm": 1.5568137168884277,
+      "learning_rate": 2.8889379066946066e-05,
+      "loss": 1.38,
+      "step": 21700
+    },
+    {
+      "epoch": 1.672948234751173,
+      "grad_norm": 1.6938014030456543,
+      "learning_rate": 2.863203558692695e-05,
+      "loss": 1.3754,
+      "step": 21750
+    },
+    {
+      "epoch": 1.672948234751173,
+      "eval_loss": 1.466833472251892,
+      "eval_runtime": 18.1069,
+      "eval_samples_per_second": 55.228,
+      "eval_steps_per_second": 13.807,
+      "step": 21750
+    },
+    {
+      "epoch": 1.6767940927620952,
+      "grad_norm": 1.3192166090011597,
+      "learning_rate": 2.837469210690783e-05,
+      "loss": 1.4388,
+      "step": 21800
+    },
+    {
+      "epoch": 1.6806399507730174,
+      "grad_norm": 2.0135934352874756,
+      "learning_rate": 2.8117348626888716e-05,
+      "loss": 1.429,
+      "step": 21850
+    },
+    {
+      "epoch": 1.6844858087839398,
+      "grad_norm": 1.4457674026489258,
+      "learning_rate": 2.78600051468696e-05,
+      "loss": 1.5154,
+      "step": 21900
+    },
+    {
+      "epoch": 1.6883316667948618,
+      "grad_norm": 1.225411295890808,
+      "learning_rate": 2.760266166685048e-05,
+      "loss": 1.4658,
+      "step": 21950
+    },
+    {
+      "epoch": 1.6921775248057842,
+      "grad_norm": 1.8256678581237793,
+      "learning_rate": 2.7345318186831365e-05,
+      "loss": 1.5004,
+      "step": 22000
+    },
+    {
+      "epoch": 1.6921775248057842,
+      "eval_loss": 1.4664525985717773,
+      "eval_runtime": 18.0331,
+      "eval_samples_per_second": 55.454,
+      "eval_steps_per_second": 13.863,
+      "step": 22000
     }
   ],
   "logging_steps": 50,