Training in progress, step 28400, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8dbf5326743cba24796a21cd4be5cacd74765499c868633eae2deae6b8dc9180
+oid sha256:6232fc3031019ccf927066ee428a2de682063c59bfbfa36005b4a59f91d614b6
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4289ec24de7d24b854fc7c0329ad35256addd59868ad0eeac322d0566b3b76aa
+oid sha256:93a781d7be1174e0a1fb9282b76358acefa16e55f3114a4a9c7954793376fd63
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76c15e95c3ead5985b9f017c695cd826a81bf4424d7190d89dd280fab1b797e6
+oid sha256:df573c8ebef5630b90590bf8e61369a0825dfe9a42993039ba55510b3bb81d53
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79254bbc05dba933815b89684ae930ec67367dac01fccace4bc1b991bbf870b2
+oid sha256:467385a064422d8119ef7edb78ee1eb671b8d69173aea975b423c4489c0404fe
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 18.99592944369064,
+  "epoch": 19.26729986431479,
   "eval_steps": 400,
-  "global_step": 28000,
+  "global_step": 28400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1127,6 +1127,22 @@
       "eval_steps_per_second": 0.833,
       "eval_wer": 0.041287937143737484,
       "step": 28000
+    },
+    {
+      "epoch": 19.27,
+      "grad_norm": 0.4070824086666107,
+      "learning_rate": 0.00010855443732845378,
+      "loss": 0.046,
+      "step": 28400
+    },
+    {
+      "epoch": 19.27,
+      "eval_loss": 0.02758474461734295,
+      "eval_runtime": 209.6786,
+      "eval_samples_per_second": 6.672,
+      "eval_steps_per_second": 0.835,
+      "eval_wer": 0.04121090741026036,
+      "step": 28400
     }
   ],
   "logging_steps": 400,
@@ -1134,7 +1150,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 8.538003628194787e+19,
+  "total_flos": 8.66018018334763e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null