Training in progress, epoch 1

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3c02091ac6b92226695a59577ab7d4dcb9cd98977e07434d7fa01e3945bc7df3
+oid sha256:f725fd71ef00caf2260fb80555a8b5723469dec92b99bbc84e03d849d9b74bd4
 size 721661957

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09c61da1490c623ed4c01c66304040f3e326c2c1ff57ea40ea62eb04deab3d3a
+oid sha256:4d4cb3787d1fb0774c67e00d75488dc6e4149674b801dd96f43c0f9c673138dc
 size 377646433

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9984e3cbd8716939c7fdea32bd39972cea55b1ab6285ee007229d989a1230b1c
+oid sha256:c8afb52420cad66c8a51ec41653557522fbb3d25f06972216fc1b36496b8dedd
 size 14639

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3e7730cfcb24b34ccf70f77b07c8d0d2b73e258077d413002a5407dd82e57b83
+oid sha256:52261608150abcba3c48f2dcd8ff39b806f986872115e36cf28df63486b380c1
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:130febd259da8024c5a724e8fe97e74d600f3893f62eaf38860a523a0936e692
+oid sha256:cee898ce0fddae89e51bcaaf05334599fc0739fe80927e9c1b5ad3d042f919c7
 size 627

last-checkpoint/trainer_state.json

@@ -1,169 +1,33 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 9.0,
-  "global_step": 2844,
+  "epoch": 1.0,
+  "global_step": 316,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 9.842327337330669e-05,
-      "loss": 24.5654,
+      "learning_rate": 9.840106595602932e-05,
+      "loss": 26.6275,
       "step": 316
     },
     {
       "epoch": 1.0,
       "eval_cer": 1.0,
-      "eval_loss": 3.3034040927886963,
+      "eval_loss": 3.2884294986724854,
       "eval_new_wer": 1.0,
       "eval_old_wer": 1.0,
-      "eval_runtime": 5.1223,
-      "eval_samples_per_second": 27.332,
-      "eval_steps_per_second": 3.514,
+      "eval_runtime": 5.0468,
+      "eval_samples_per_second": 27.741,
+      "eval_steps_per_second": 3.567,
       "step": 316
-    },
-    {
-      "epoch": 2.0,
-      "learning_rate": 9.140572951365757e-05,
-      "loss": 3.4232,
-      "step": 632
-    },
-    {
-      "epoch": 2.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.2738046646118164,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.4174,
-      "eval_samples_per_second": 25.843,
-      "eval_steps_per_second": 3.323,
-      "step": 632
-    },
-    {
-      "epoch": 3.0,
-      "learning_rate": 8.438818565400845e-05,
-      "loss": 3.3815,
-      "step": 948
-    },
-    {
-      "epoch": 3.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.2943756580352783,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.7568,
-      "eval_samples_per_second": 24.319,
-      "eval_steps_per_second": 3.127,
-      "step": 948
-    },
-    {
-      "epoch": 4.0,
-      "learning_rate": 7.737064179435932e-05,
-      "loss": 3.3771,
-      "step": 1264
-    },
-    {
-      "epoch": 4.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.3197929859161377,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.2197,
-      "eval_samples_per_second": 26.821,
-      "eval_steps_per_second": 3.448,
-      "step": 1264
-    },
-    {
-      "epoch": 5.0,
-      "learning_rate": 7.03530979347102e-05,
-      "loss": 3.3742,
-      "step": 1580
-    },
-    {
-      "epoch": 5.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.31402850151062,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.1197,
-      "eval_samples_per_second": 27.345,
-      "eval_steps_per_second": 3.516,
-      "step": 1580
-    },
-    {
-      "epoch": 6.0,
-      "learning_rate": 6.333555407506108e-05,
-      "loss": 3.3683,
-      "step": 1896
-    },
-    {
-      "epoch": 6.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.2977089881896973,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.4977,
-      "eval_samples_per_second": 25.465,
-      "eval_steps_per_second": 3.274,
-      "step": 1896
-    },
-    {
-      "epoch": 7.0,
-      "learning_rate": 5.631801021541196e-05,
-      "loss": 3.3472,
-      "step": 2212
-    },
-    {
-      "epoch": 7.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.3024919033050537,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.4129,
-      "eval_samples_per_second": 25.864,
-      "eval_steps_per_second": 3.325,
-      "step": 2212
-    },
-    {
-      "epoch": 8.0,
-      "learning_rate": 4.9300466355762824e-05,
-      "loss": 3.315,
-      "step": 2528
-    },
-    {
-      "epoch": 8.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.3284974098205566,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.2334,
-      "eval_samples_per_second": 26.751,
-      "eval_steps_per_second": 3.439,
-      "step": 2528
-    },
-    {
-      "epoch": 9.0,
-      "learning_rate": 4.2282922496113704e-05,
-      "loss": 3.2856,
-      "step": 2844
-    },
-    {
-      "epoch": 9.0,
-      "eval_cer": 1.0,
-      "eval_loss": 3.2858822345733643,
-      "eval_new_wer": 1.0,
-      "eval_old_wer": 1.0,
-      "eval_runtime": 5.3674,
-      "eval_samples_per_second": 26.083,
-      "eval_steps_per_second": 3.354,
-      "step": 2844
     }
   ],
   "max_steps": 4740,
   "num_train_epochs": 15,
-  "total_flos": 2.104156079215534e+18,
+  "total_flos": 2.336192011118592e+17,
   "trial_name": null,
   "trial_params": null
 }

last-checkpoint/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c6b9420bb6a966eb97bcd5f710a858bd94b0bf4baee2011fe4f71ce2c6c55bd
+oid sha256:0a98ea4cc0a3e85c5ac6067cc39934b4325207661835f058515305611e4d4f88
 size 3387

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:09c61da1490c623ed4c01c66304040f3e326c2c1ff57ea40ea62eb04deab3d3a
+oid sha256:4d4cb3787d1fb0774c67e00d75488dc6e4149674b801dd96f43c0f9c673138dc
 size 377646433

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c6b9420bb6a966eb97bcd5f710a858bd94b0bf4baee2011fe4f71ce2c6c55bd
+oid sha256:0a98ea4cc0a3e85c5ac6067cc39934b4325207661835f058515305611e4d4f88
 size 3387