Training in progress, step 20800, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:01af806bfb43537f473716c6921c56c3f1b6cff5b24f430ec50b63d6432ff62c
+oid sha256:f2ccbaa3366fc91a9caa865d514bdc5bc7272bb243e4623552358daf6c381842
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f54746b0feae2e036d16930d4e673363c77f9f9aafe635895c1098194f55d6ef
+oid sha256:bd4cc3504193d269c1271fcbb4b8624ac22d68154afdcd94625f3fb1b458ee8d
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a7ec026adbf9bcbd8a783c266184ac6a8dfb47b9e7d5a7687187770c89b84faf
+oid sha256:2f7bbcf5ff7ad61af1c3ef61aa10d73d1160259d8b5010c2a4340676fa439df1
 size 14308

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05c969971a5565a3fd0897b347349de9a9a330022bc9b3755b121094b8004da8
+oid sha256:af549fa9698688de4d4da8eafe34f4994b7fb3147b788357d90bc61bb8d36a8a
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 13.83989145183175,
+  "epoch": 14.111261872455902,
   "eval_steps": 400,
-  "global_step": 20400,
+  "global_step": 20800,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -823,6 +823,22 @@
       "eval_steps_per_second": 0.83,
       "eval_wer": 0.0625481435834232,
       "step": 20400
+    },
+    {
+      "epoch": 14.11,
+      "grad_norm": 0.3025607466697693,
+      "learning_rate": 0.00016070448307410794,
+      "loss": 0.0737,
+      "step": 20800
+    },
+    {
+      "epoch": 14.11,
+      "eval_loss": 0.03549855947494507,
+      "eval_runtime": 209.8898,
+      "eval_samples_per_second": 6.665,
+      "eval_steps_per_second": 0.834,
+      "eval_wer": 0.060314281312586655,
+      "step": 20800
     }
   ],
   "logging_steps": 400,
@@ -830,7 +846,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 6.2228407342062576e+19,
+  "total_flos": 6.345721739092566e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null