Training in progress, step 18000, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5fd64b4fff9be5f0c3e2f6b4d9919b9e6375df4b9f578e7ba32211a7c5285de9
+oid sha256:f56b521d1a393817f17047a20ecf2bb1e83ee9f43692fbe30db28020b9d79b36
 size 36730224

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2d27c5a577064cfc46e35c17df10152b71102230382c1124b730922de85a7cde
+oid sha256:af75d0581fbdc90030a2d5f4ec75948b6c9a777f25105f2635a11a66e19a8991
 size 73588346

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c6f63d896b1e09539b72ac15eabafd681bf824376794bc90d0981fb00940917
+oid sha256:aec74914111c893531d77390ed26a8e62f47c5fd368563e3e2ea395d9971bdd0
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a3d47a33fe410d7c39aed7512ee64bb9c07ee05e603db567ade52fed1bf5077
+oid sha256:fb36538bca0e8f82a230d205fcda169c895fd4416ff12f18badea8938ea79e34
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3fefc995528f5d0eb5c6c7d885b8cddef853e490bf4df60e860a20196ce82f7a
+oid sha256:04fb24a5e33627ec28e9d8236874cf729cbb5859a1714c36c632683bea019abf
 size 1064

last-checkpoint/trainer_state.json

@@ -2,9 +2,9 @@
   "best_global_step": 16750,
   "best_metric": 1.4708431959152222,
   "best_model_checkpoint": "./ar-diffusion-checkpoints/checkpoint-15500",
-  "epoch": 1.346050303822783,
+  "epoch": 1.3845088839320052,
   "eval_steps": 250,
-  "global_step": 17500,
+  "global_step": 18000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -3018,6 +3018,92 @@
       "eval_samples_per_second": 55.814,
       "eval_steps_per_second": 13.954,
       "step": 17500
+    },
+    {
+      "epoch": 1.349896161833705,
+      "grad_norm": 1.6970813274383545,
+      "learning_rate": 5.0243741038932386e-05,
+      "loss": 1.4442,
+      "step": 17550
+    },
+    {
+      "epoch": 1.3537420198446273,
+      "grad_norm": 1.0942292213439941,
+      "learning_rate": 4.998639755891327e-05,
+      "loss": 1.4769,
+      "step": 17600
+    },
+    {
+      "epoch": 1.3575878778555497,
+      "grad_norm": 1.720035195350647,
+      "learning_rate": 4.972905407889416e-05,
+      "loss": 1.4519,
+      "step": 17650
+    },
+    {
+      "epoch": 1.3614337358664719,
+      "grad_norm": 0.8887185454368591,
+      "learning_rate": 4.947171059887503e-05,
+      "loss": 1.4201,
+      "step": 17700
+    },
+    {
+      "epoch": 1.365279593877394,
+      "grad_norm": 1.9557030200958252,
+      "learning_rate": 4.921436711885592e-05,
+      "loss": 1.4848,
+      "step": 17750
+    },
+    {
+      "epoch": 1.365279593877394,
+      "eval_loss": 1.476893424987793,
+      "eval_runtime": 17.9988,
+      "eval_samples_per_second": 55.559,
+      "eval_steps_per_second": 13.89,
+      "step": 17750
+    },
+    {
+      "epoch": 1.3691254518883162,
+      "grad_norm": 1.471414566040039,
+      "learning_rate": 4.8957023638836804e-05,
+      "loss": 1.4541,
+      "step": 17800
+    },
+    {
+      "epoch": 1.3729713098992384,
+      "grad_norm": 1.350690484046936,
+      "learning_rate": 4.8699680158817684e-05,
+      "loss": 1.3954,
+      "step": 17850
+    },
+    {
+      "epoch": 1.3768171679101608,
+      "grad_norm": 0.7363431453704834,
+      "learning_rate": 4.844233667879857e-05,
+      "loss": 1.4919,
+      "step": 17900
+    },
+    {
+      "epoch": 1.380663025921083,
+      "grad_norm": 1.8820909261703491,
+      "learning_rate": 4.818499319877946e-05,
+      "loss": 1.4177,
+      "step": 17950
+    },
+    {
+      "epoch": 1.3845088839320052,
+      "grad_norm": 0.8440986275672913,
+      "learning_rate": 4.792764971876033e-05,
+      "loss": 1.3995,
+      "step": 18000
+    },
+    {
+      "epoch": 1.3845088839320052,
+      "eval_loss": 1.4794726371765137,
+      "eval_runtime": 17.9989,
+      "eval_samples_per_second": 55.559,
+      "eval_steps_per_second": 13.89,
+      "step": 18000
     }
   ],
   "logging_steps": 50,