Training in progress, step 28000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d88057cfed823ea5d6af511d7d549bd0a28fca10b9c1df2d15ec85aa5ad4c72b
+oid sha256:8dbf5326743cba24796a21cd4be5cacd74765499c868633eae2deae6b8dc9180
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2b8dd0c49acf269a4fd5ec6489dcf34bc83db6b3298bf03e0ca7ab26fc4e757
+oid sha256:4289ec24de7d24b854fc7c0329ad35256addd59868ad0eeac322d0566b3b76aa
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3a8de78e722e39a242378f9e619abe10dce82fb669f6420bd3866281f0426c23
-size 14180
+oid sha256:76c15e95c3ead5985b9f017c695cd826a81bf4424d7190d89dd280fab1b797e6
+size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bc179071d4d6b0c2935c28332a13543209647944abee9bab5ae5846d265bb9bb
+oid sha256:79254bbc05dba933815b89684ae930ec67367dac01fccace4bc1b991bbf870b2
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 18.724559023066487,
+  "epoch": 18.99592944369064,
   "eval_steps": 400,
-  "global_step": 27600,
+  "global_step": 28000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1111,6 +1111,22 @@
       "eval_steps_per_second": 0.834,
       "eval_wer": 0.04375288861500539,
       "step": 27600
+    },
+    {
+      "epoch": 19.0,
+      "grad_norm": 0.289604514837265,
+      "learning_rate": 0.00011129917657822506,
+      "loss": 0.0533,
+      "step": 28000
+    },
+    {
+      "epoch": 19.0,
+      "eval_loss": 0.0264176856726408,
+      "eval_runtime": 210.0905,
+      "eval_samples_per_second": 6.659,
+      "eval_steps_per_second": 0.833,
+      "eval_wer": 0.041287937143737484,
+      "step": 28000
     }
   ],
   "logging_steps": 400,
@@ -1118,7 +1134,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 8.417104754761507e+19,
+  "total_flos": 8.538003628194787e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null