Training in progress, step 22000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ae3ae3d3f6eed21e23e402626f9b5d36d0eaf722fe8b6b3668eaa433af56d43d
+oid sha256:a394aeeea4a2de10f768b61832c5745c1c11d2e1d8a9cc5baec1c529bf6fdc87
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d63d4faba976ec1e6b945518fe24c679be5a3f42c8e71fb2eaa42392c14dcd4b
+oid sha256:84a2d925065461d86e9e9afc058809c9dd1ad9bcb825ca34061365af47a11542
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dd6fe92ded35c583b056abd78291bd133b79c7076da34fc990f7b299e212c88d
+oid sha256:8f692015617e705ff861686b4082c2da1240af93c73d6ae630cfff58a945aab7
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:47997405a3602f8cade0562119b11b6490f526c5b0319017e96e10ae9857494a
+oid sha256:e33c5010c37401e20c1de34fe9285c339740cd9c43f7069d973e83e97e3419e1
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 14.654002713704207,
+  "epoch": 14.925373134328359,
   "eval_steps": 400,
-  "global_step": 21600,
+  "global_step": 22000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -871,6 +871,22 @@
       "eval_steps_per_second": 0.825,
       "eval_wer": 0.057464181173933135,
       "step": 21600
+    },
+    {
+      "epoch": 14.93,
+      "grad_norm": 0.19861555099487305,
+      "learning_rate": 0.00015247026532479413,
+      "loss": 0.0722,
+      "step": 22000
+    },
+    {
+      "epoch": 14.93,
+      "eval_loss": 0.03097483143210411,
+      "eval_runtime": 209.3416,
+      "eval_samples_per_second": 6.683,
+      "eval_steps_per_second": 0.836,
+      "eval_wer": 0.05754121090741026,
+      "step": 22000
     }
   ],
   "logging_steps": 400,
@@ -878,7 +894,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 6.58884399297287e+19,
+  "total_flos": 6.710427073339886e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null