Training in progress, step 21600, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:16dbb1a04572952218422ea39cf7570a42446d5f087786783ad33828643993bc
+oid sha256:ae3ae3d3f6eed21e23e402626f9b5d36d0eaf722fe8b6b3668eaa433af56d43d
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a216eb1303cd5b71afad6f5a91b0b5e3768ab04fd074af690869ca5bedb8926f
+oid sha256:d63d4faba976ec1e6b945518fe24c679be5a3f42c8e71fb2eaa42392c14dcd4b
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7ad8ec50c3c1c91dac2f5655527d8594f2033e004db09ad65dd9a85d7afb91a4
+oid sha256:dd6fe92ded35c583b056abd78291bd133b79c7076da34fc990f7b299e212c88d
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a3f2e6e05de398f17a37f5b322d92a7ed3bd1f1f712be1a3b88b12b3511e8906
+oid sha256:47997405a3602f8cade0562119b11b6490f526c5b0319017e96e10ae9857494a
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 14.382632293080054,
+  "epoch": 14.654002713704207,
   "eval_steps": 400,
-  "global_step": 21200,
+  "global_step": 21600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -855,6 +855,22 @@
       "eval_steps_per_second": 0.834,
       "eval_wer": 0.06262517331690032,
       "step": 21200
+    },
+    {
+      "epoch": 14.65,
+      "grad_norm": 0.1968027800321579,
+      "learning_rate": 0.00015521500457456541,
+      "loss": 0.0715,
+      "step": 21600
+    },
+    {
+      "epoch": 14.65,
+      "eval_loss": 0.03143769130110741,
+      "eval_runtime": 212.2,
+      "eval_samples_per_second": 6.593,
+      "eval_steps_per_second": 0.825,
+      "eval_wer": 0.057464181173933135,
+      "step": 21600
     }
   ],
   "logging_steps": 400,
@@ -862,7 +878,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 6.467377957667312e+19,
+  "total_flos": 6.58884399297287e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null