Training in progress, epoch 1

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8954730c3b7d6a8c6e7b661b708ad0ec4eae16384a6c580b4e06bf65bc9efb54
+oid sha256:cc2dba7e61e936bd037054d1492dd4bde61240803556e98c67640e59a0d8b932
 size 721661957

last-checkpoint/pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a48f58a40c5499527def9658b126a9b8a8f44a4b10e1712bcf1b9b7eacbfa648
+oid sha256:676d183e8a62a355505e444a9178c71fda557635d4faebdf8ae33188ae206ef1
 size 377646433

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:79e0b4a432ac67fc9f04f92db872bf3b9b6f61266a39527fd0fb9b587793e21f
-size 14639
+oid sha256:bf4aec27bfeea90738219d1867fadd0338c634aad327d5dd48e5ad5b8e19dff0
+size 14575

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ececadbc278c596bb2722aabbaaf3c5203071232d76d6fc01bb563a1af149dc5
+oid sha256:26b20d0624fd6735c75d67cd192faab8b677bc9f276931f799a0eae2e7880574
 size 557

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24f6f2c57c292c85a26411074b2fc9eb3592d917a44cb76c47f4102bdfd0e56e
+oid sha256:85bf7d42805134dc2780b832028e3ebc4cc5356cc9e3b37fd394c3272d80308c
 size 627

last-checkpoint/trainer_state.json

@@ -1,176 +1,33 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 30.0,
-  "global_step": 9660,
+  "epoch": 1.0,
+  "global_step": 322,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 3.0,
-      "learning_rate": 9.4824016563147e-05,
-      "loss": 0.5422,
-      "step": 966
-    },
-    {
-      "epoch": 3.0,
-      "eval_cer": 0.21298701298701297,
-      "eval_loss": 1.338439702987671,
-      "eval_runtime": 7.7826,
-      "eval_samples_per_second": 26.983,
-      "eval_steps_per_second": 3.469,
-      "eval_wer": 0.07142857142857142,
-      "step": 966
-    },
-    {
-      "epoch": 6.0,
-      "learning_rate": 8.429770077367331e-05,
-      "loss": 0.1889,
-      "step": 1932
-    },
-    {
-      "epoch": 6.0,
-      "eval_cer": 0.2354978354978355,
-      "eval_loss": 1.5368783473968506,
-      "eval_runtime": 8.8989,
-      "eval_samples_per_second": 23.598,
-      "eval_steps_per_second": 3.034,
-      "eval_wer": 0.06666666666666667,
-      "step": 1932
-    },
-    {
-      "epoch": 9.0,
-      "learning_rate": 7.378228179143511e-05,
-      "loss": 0.1415,
-      "step": 2898
-    },
-    {
-      "epoch": 9.0,
-      "eval_cer": 0.23376623376623376,
-      "eval_loss": 1.6211291551589966,
-      "eval_runtime": 7.7985,
-      "eval_samples_per_second": 26.928,
-      "eval_steps_per_second": 3.462,
-      "eval_wer": 0.047619047619047616,
-      "step": 2898
-    },
-    {
-      "epoch": 12.0,
-      "learning_rate": 6.325596600196144e-05,
-      "loss": 0.109,
-      "step": 3864
-    },
-    {
-      "epoch": 12.0,
-      "eval_cer": 0.24848484848484848,
-      "eval_loss": 1.5097322463989258,
-      "eval_runtime": 7.3722,
-      "eval_samples_per_second": 28.486,
-      "eval_steps_per_second": 3.662,
-      "eval_wer": 0.0761904761904762,
-      "step": 3864
-    },
-    {
-      "epoch": 15.0,
-      "learning_rate": 5.272965021248775e-05,
-      "loss": 0.1803,
-      "step": 4830
-    },
-    {
-      "epoch": 15.0,
-      "eval_cer": 0.24935064935064935,
-      "eval_loss": 2.160435676574707,
-      "eval_runtime": 7.8188,
-      "eval_samples_per_second": 26.858,
-      "eval_steps_per_second": 3.453,
-      "eval_wer": 0.12380952380952381,
-      "step": 4830
-    },
-    {
-      "epoch": 18.0,
-      "learning_rate": 4.220333442301406e-05,
-      "loss": 0.1495,
-      "step": 5796
-    },
-    {
-      "epoch": 18.0,
-      "eval_cer": 0.2753246753246753,
-      "eval_loss": 2.211272954940796,
-      "eval_runtime": 7.8373,
-      "eval_samples_per_second": 26.795,
-      "eval_steps_per_second": 3.445,
-      "eval_wer": 0.15714285714285714,
-      "step": 5796
-    },
-    {
-      "epoch": 21.0,
-      "learning_rate": 3.169881224801134e-05,
-      "loss": 0.242,
-      "step": 6762
-    },
-    {
-      "epoch": 21.0,
-      "eval_cer": 0.2831168831168831,
-      "eval_loss": 1.5069493055343628,
-      "eval_runtime": 7.8359,
-      "eval_samples_per_second": 26.8,
-      "eval_steps_per_second": 3.446,
-      "eval_wer": 0.14285714285714285,
-      "step": 6762
-    },
-    {
-      "epoch": 24.0,
-      "learning_rate": 2.117249645853765e-05,
-      "loss": 0.5494,
-      "step": 7728
-    },
-    {
-      "epoch": 24.0,
-      "eval_cer": 0.4562770562770563,
-      "eval_loss": 1.3968360424041748,
-      "eval_runtime": 7.8934,
-      "eval_samples_per_second": 26.605,
-      "eval_steps_per_second": 3.421,
-      "eval_wer": 0.23809523809523808,
-      "step": 7728
-    },
-    {
-      "epoch": 27.0,
-      "learning_rate": 1.0646180669063966e-05,
-      "loss": 0.8963,
-      "step": 8694
-    },
-    {
-      "epoch": 27.0,
-      "eval_cer": 0.9783549783549783,
-      "eval_loss": 2.9235026836395264,
-      "eval_runtime": 7.7684,
-      "eval_samples_per_second": 27.032,
-      "eval_steps_per_second": 3.476,
-      "eval_wer": 0.9571428571428572,
-      "step": 8694
-    },
-    {
-      "epoch": 30.0,
-      "learning_rate": 1.1986487959028006e-07,
-      "loss": 0.5384,
-      "step": 9660
-    },
-    {
-      "epoch": 30.0,
-      "eval_cer": 0.6294372294372295,
-      "eval_loss": 2.833552360534668,
-      "eval_runtime": 7.3279,
-      "eval_samples_per_second": 28.658,
-      "eval_steps_per_second": 3.685,
-      "eval_wer": 0.37142857142857144,
-      "step": 9660
+      "epoch": 1.0,
+      "learning_rate": 9.830028328611899e-05,
+      "loss": 3.4876,
+      "step": 322
+    },
+    {
+      "epoch": 1.0,
+      "eval_cer": 0.9818181818181818,
+      "eval_loss": 3.3877487182617188,
+      "eval_new_wer": 0.9,
+      "eval_old_wer": 1.0,
+      "eval_runtime": 8.8586,
+      "eval_samples_per_second": 23.706,
+      "eval_steps_per_second": 3.048,
+      "step": 322
     }
   ],
-  "max_steps": 9660,
-  "num_train_epochs": 30,
-  "total_flos": 7.271827161624115e+18,
+  "max_steps": 4830,
+  "num_train_epochs": 15,
+  "total_flos": 2.4964363986000384e+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:a30589e0fe0bf5e7c2e5fa26dc43e03bdd8d96561b2e2a1ebd46f30cb3bee76c
-size 3451
+oid sha256:2482c421f2e54815733d79c491d554a4c42d32f13b9839fbaae213086a45cc2f
+size 3387

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a48f58a40c5499527def9658b126a9b8a8f44a4b10e1712bcf1b9b7eacbfa648
+oid sha256:676d183e8a62a355505e444a9178c71fda557635d4faebdf8ae33188ae206ef1
 size 377646433

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a30589e0fe0bf5e7c2e5fa26dc43e03bdd8d96561b2e2a1ebd46f30cb3bee76c
-size 3451
+oid sha256:2482c421f2e54815733d79c491d554a4c42d32f13b9839fbaae213086a45cc2f
+size 3387