Training in progress, step 29600, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9fe283633a388ba2d022ff9f5bd22a4410a9ce61cf5ab36a933f8316c404dc0d
+oid sha256:3e63f877c4506681a490774017678ec3ddaf77e75d1e84f86025ada6590efbd8
 size 1262135480

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4ffc119225525035a7f335e5b8efd08e194472d357b13852524d4f4eb0578b80
+oid sha256:4ff96e9d8d8bd2f41f071244c4ff2909a15f3a277d70d0f48e2c1231181b87b5
 size 2490815798

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1d853691023356ee049c7f9520a152041301e582ba9494903ecaa091cb4963f6
+oid sha256:32ab7f10ca6b13468e41de24a05121a51b38477fe4c9a87064b918ea752578ac
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:76c2f43626453dc6c68d43c01f88ecb4cc62eda02fd56285615f23696dcf1415
+oid sha256:940d3836b6775a577a1a3b78f4ff4621e538b7a2a81b57676d7aecf40e93b414
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 19.810040705563093,
+  "epoch": 20.081411126187245,
   "eval_steps": 400,
-  "global_step": 29200,
+  "global_step": 29600,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -1175,6 +1175,22 @@
       "eval_steps_per_second": 0.832,
       "eval_wer": 0.04167308581112309,
       "step": 29200
+    },
+    {
+      "epoch": 20.08,
+      "grad_norm": 7.543890953063965,
+      "learning_rate": 0.00010032021957913997,
+      "loss": 0.0483,
+      "step": 29600
+    },
+    {
+      "epoch": 20.08,
+      "eval_loss": 0.025978881865739822,
+      "eval_runtime": 210.6554,
+      "eval_samples_per_second": 6.641,
+      "eval_steps_per_second": 0.831,
+      "eval_wer": 0.0395932830072408,
+      "step": 29600
     }
   ],
   "logging_steps": 400,
@@ -1182,7 +1198,7 @@
   "num_input_tokens_seen": 0,
   "num_train_epochs": 30,
   "save_steps": 400,
-  "total_flos": 8.903221408478038e+19,
+  "total_flos": 9.027987196010655e+19,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null