End of training

all_results.json

@@ -1,14 +1,14 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 14.633834838867188,
-    "eval_runtime": 13.0798,
+    "epoch": 100.0,
+    "eval_loss": 0.535879373550415,
+    "eval_runtime": 13.4178,
     "eval_samples": 301,
-    "eval_samples_per_second": 23.013,
-    "eval_steps_per_second": 0.765,
-    "eval_wer": 1.0,
-    "train_loss": 15.28215859153054,
-    "train_runtime": 50.1491,
+    "eval_samples_per_second": 22.433,
+    "eval_steps_per_second": 0.745,
+    "eval_wer": 0.6191872643485546,
+    "train_loss": 1.9645296894420277,
+    "train_runtime": 5564.2506,
     "train_samples": 704,
-    "train_samples_per_second": 14.038,
-    "train_steps_per_second": 0.439
+    "train_samples_per_second": 12.652,
+    "train_steps_per_second": 0.395
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 14.633834838867188,
-    "eval_runtime": 13.0798,
+    "epoch": 100.0,
+    "eval_loss": 0.535879373550415,
+    "eval_runtime": 13.4178,
     "eval_samples": 301,
-    "eval_samples_per_second": 23.013,
-    "eval_steps_per_second": 0.765,
-    "eval_wer": 1.0
+    "eval_samples_per_second": 22.433,
+    "eval_steps_per_second": 0.745,
+    "eval_wer": 0.6191872643485546
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:048b851ca4920043c2b10c6c2c9a342537084ee9653667ef3732a7b7a03750db
+oid sha256:78b042e9bf1f74a24f8ca542b89658c28b5ae650b6d32eaaa99d0607f8b042e2
 size 1262108145

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "train_loss": 15.28215859153054,
-    "train_runtime": 50.1491,
+    "epoch": 100.0,
+    "train_loss": 1.9645296894420277,
+    "train_runtime": 5564.2506,
     "train_samples": 704,
-    "train_samples_per_second": 14.038,
-    "train_steps_per_second": 0.439
+    "train_samples_per_second": 12.652,
+    "train_steps_per_second": 0.395
 }

trainer_state.json

@@ -1,25 +1,193 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.0,
-  "global_step": 22,
+  "epoch": 100.0,
+  "global_step": 2200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 1.0,
-      "step": 22,
-      "total_flos": 1.5056241143943168e+17,
-      "train_loss": 15.28215859153054,
-      "train_runtime": 50.1491,
-      "train_samples_per_second": 14.038,
-      "train_steps_per_second": 0.439
+      "epoch": 4.55,
+      "learning_rate": 3.4299999999999993e-05,
+      "loss": 9.7952,
+      "step": 100
+    },
+    {
+      "epoch": 9.09,
+      "learning_rate": 6.929999999999999e-05,
+      "loss": 3.6097,
+      "step": 200
+    },
+    {
+      "epoch": 13.64,
+      "learning_rate": 6.657e-05,
+      "loss": 3.0981,
+      "step": 300
+    },
+    {
+      "epoch": 18.18,
+      "learning_rate": 6.306999999999999e-05,
+      "loss": 3.0036,
+      "step": 400
+    },
+    {
+      "epoch": 22.73,
+      "learning_rate": 5.9569999999999994e-05,
+      "loss": 2.8617,
+      "step": 500
+    },
+    {
+      "epoch": 22.73,
+      "eval_loss": 2.626376152038574,
+      "eval_runtime": 13.7667,
+      "eval_samples_per_second": 21.864,
+      "eval_steps_per_second": 0.726,
+      "eval_wer": 1.0012568077084205,
+      "step": 500
+    },
+    {
+      "epoch": 27.27,
+      "learning_rate": 5.607e-05,
+      "loss": 2.4115,
+      "step": 600
+    },
+    {
+      "epoch": 31.82,
+      "learning_rate": 5.256999999999999e-05,
+      "loss": 1.8202,
+      "step": 700
+    },
+    {
+      "epoch": 36.36,
+      "learning_rate": 4.906999999999999e-05,
+      "loss": 1.5225,
+      "step": 800
+    },
+    {
+      "epoch": 40.91,
+      "learning_rate": 4.557e-05,
+      "loss": 1.3681,
+      "step": 900
+    },
+    {
+      "epoch": 45.45,
+      "learning_rate": 4.2069999999999995e-05,
+      "loss": 1.2716,
+      "step": 1000
+    },
+    {
+      "epoch": 45.45,
+      "eval_loss": 0.6217501163482666,
+      "eval_runtime": 13.948,
+      "eval_samples_per_second": 21.58,
+      "eval_steps_per_second": 0.717,
+      "eval_wer": 0.6941767909509845,
+      "step": 1000
+    },
+    {
+      "epoch": 50.0,
+      "learning_rate": 3.857e-05,
+      "loss": 1.2079,
+      "step": 1100
+    },
+    {
+      "epoch": 54.55,
+      "learning_rate": 3.5069999999999995e-05,
+      "loss": 1.154,
+      "step": 1200
+    },
+    {
+      "epoch": 59.09,
+      "learning_rate": 3.157e-05,
+      "loss": 1.1124,
+      "step": 1300
+    },
+    {
+      "epoch": 63.64,
+      "learning_rate": 2.807e-05,
+      "loss": 1.087,
+      "step": 1400
+    },
+    {
+      "epoch": 68.18,
+      "learning_rate": 2.4569999999999997e-05,
+      "loss": 1.049,
+      "step": 1500
+    },
+    {
+      "epoch": 68.18,
+      "eval_loss": 0.544226348400116,
+      "eval_runtime": 13.3563,
+      "eval_samples_per_second": 22.536,
+      "eval_steps_per_second": 0.749,
+      "eval_wer": 0.6367825722664432,
+      "step": 1500
+    },
+    {
+      "epoch": 72.73,
+      "learning_rate": 2.1069999999999996e-05,
+      "loss": 1.023,
+      "step": 1600
+    },
+    {
+      "epoch": 77.27,
+      "learning_rate": 1.757e-05,
+      "loss": 1.0046,
+      "step": 1700
+    },
+    {
+      "epoch": 81.82,
+      "learning_rate": 1.4069999999999999e-05,
+      "loss": 0.9863,
+      "step": 1800
+    },
+    {
+      "epoch": 86.36,
+      "learning_rate": 1.0569999999999999e-05,
+      "loss": 0.9772,
+      "step": 1900
+    },
+    {
+      "epoch": 90.91,
+      "learning_rate": 7.07e-06,
+      "loss": 0.9632,
+      "step": 2000
+    },
+    {
+      "epoch": 90.91,
+      "eval_loss": 0.5364273190498352,
+      "eval_runtime": 13.6559,
+      "eval_samples_per_second": 22.042,
+      "eval_steps_per_second": 0.732,
+      "eval_wer": 0.6242144951822372,
+      "step": 2000
+    },
+    {
+      "epoch": 95.45,
+      "learning_rate": 3.5699999999999993e-06,
+      "loss": 0.9443,
+      "step": 2100
+    },
+    {
+      "epoch": 100.0,
+      "learning_rate": 6.999999999999999e-08,
+      "loss": 0.9485,
+      "step": 2200
+    },
+    {
+      "epoch": 100.0,
+      "step": 2200,
+      "total_flos": 1.4967759850398351e+19,
+      "train_loss": 1.9645296894420277,
+      "train_runtime": 5564.2506,
+      "train_samples_per_second": 12.652,
+      "train_steps_per_second": 0.395
     }
   ],
-  "max_steps": 22,
-  "num_train_epochs": 1,
-  "total_flos": 1.5056241143943168e+17,
+  "max_steps": 2200,
+  "num_train_epochs": 100,
+  "total_flos": 1.4967759850398351e+19,
   "trial_name": null,
   "trial_params": null
 }