Training in progress, step 27200, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:454556d34765dde3ef0274c5c6b8eb640baa6e6df29bfda1cfc6dee223b8e007
+oid sha256:3f45da09880b955d067e0e83ff636f04f3f476029c545b739733d115bc2b56df
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0ddd8ed542961bb752300cafd51f0407b631cdc7c5454a6c0ade8d6fe545927a
+oid sha256:d12ee1adf29afb1cd124c4828bb5e8a3d010ecd5cb382013efdc00ad6d87c0be
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd885655941c277ae79cac4e6aa317e40edeeb51d003a9c14bd5afa883c299e2
+oid sha256:e39a8570cf9bb277229c0af9010650d1b7d87179a3c9ea7ea62202b3d40a00b5
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a79b8f6ac5adc78ea938a1602b91550b06839c8391ae94f652aa3c8e2540ea8c
+oid sha256:be52170b0c7201f7d765090cb342725a52adbb4895d759580d167c5f09a8a61c
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 18.181818181818183,
+  "epoch": 18.453188602442335,
   "eval_steps": 400,
-  "global_step": 26800,
+  "global_step": 27200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1079,6 +1079,22 @@
       "eval_steps_per_second": 0.837,
       "eval_wer": 0.045986750885841934,
       "step": 26800
+    },
+    {
+      "epoch": 18.45,
+      "grad_norm": 1.2717292308807373,
+      "learning_rate": 0.0001167886550777676,
+      "loss": 0.0544,
+      "step": 27200
+    },
+    {
+      "epoch": 18.45,
+      "eval_loss": 0.02826772816479206,
+      "eval_runtime": 208.8935,
+      "eval_samples_per_second": 6.697,
+      "eval_steps_per_second": 0.838,
+      "eval_wer": 0.04644892928670467,
+      "step": 27200
     }
   ],
   "logging_steps": 400,
@@ -1086,7 +1102,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 8.175079702729595e+19,
+  "total_flos": 8.295970690164207e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null