Training in progress, step 19600, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6352187cf385397f40eed62cf701dec875c7e3375118d91430a850443ab1f27d
+oid sha256:61f07e30ba6c3e4d46423b96f0cbcbd82106b6bb256c1c12e811925d96d026b5
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0cfa6e921875d63852f96db976fdc599ce3aea4b10926ddb0746660a5a59dc4
+oid sha256:67dae5c921c1d0e3ec0ef2ca745ddb83ae850f785c418cb799e8d39e9977effb
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7a7a0041b53de4a505be9cc74a61653f340f65bd90c8c4022f66bf9eaf4ab421
+oid sha256:aed632af97046c1419a3b3bfb31bf5b26a7fe13c4e9fee7877e803a6d9b9ab78
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:46e26e61dfa7a5f81fbf493f586e17fb54189c196dc287ec04a698210453fdf9
+oid sha256:ed90542fa85abfeb1fc1d8ba1917d59ae062284951ce0a75c17787d5046ce253
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 13.025780189959294,
+  "epoch": 13.297150610583447,
   "eval_steps": 400,
-  "global_step": 19200,
+  "global_step": 19600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -775,6 +775,22 @@
       "eval_steps_per_second": 0.831,
       "eval_wer": 0.057695270374364505,
       "step": 19200
+    },
+    {
+      "epoch": 13.3,
+      "grad_norm": 0.4078648090362549,
+      "learning_rate": 0.00016893870082342174,
+      "loss": 0.0779,
+      "step": 19600
+    },
+    {
+      "epoch": 13.3,
+      "eval_loss": 0.03577824681997299,
+      "eval_runtime": 210.5433,
+      "eval_samples_per_second": 6.645,
+      "eval_steps_per_second": 0.831,
+      "eval_wer": 0.06208596518256047,
+      "step": 19600
     }
   ],
   "logging_steps": 400,
@@ -782,7 +798,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 5.85537722383966e+19,
+  "total_flos": 5.978374840640582e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null