Training in progress, step 10

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:daeea7778cdeda935a2e90f442518eb60512721cd10bce0921a80c47aa8f2543
+oid sha256:6449d1e6c9c423fb8849934a1549ffae7f8f17cfec3201f6fc3aa2a077782fa6
 size 151061672

runs/Jun25_21-12-28_bharatsahaiyak-test/events.out.tfevents.1719349949.bharatsahaiyak-test.990312.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:edd4300438a7ad67abf772860cada59cd4255716e7200d824dbda51d7c17f72f
-size 6070
+oid sha256:29182ce5fbab9e6fcf4da9cfaf764d721c25d1412d041930b309a3de5b0a6875
+size 6382

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:37c1d71d4573028fde34dd156869c75306de202ae193d1b60195dcab72289f8d
+oid sha256:76aa8e1d1d3d0d745a1e6d6e9ca6f5ddf9f945a3ba698d9644e4cbe35394d28f
 size 5240

whisper-tiny-hi-checkpoint-1/trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 69.98899144720129,
-  "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-120",
+  "best_metric": 58.31498729889923,
+  "best_model_checkpoint": "./whisper-tiny-hi2_test/checkpoint-80",
   "epoch": 6.25,
   "eval_steps": 40,
   "global_step": 200,
@@ -10,126 +10,126 @@
   "log_history": [
     {
       "epoch": 0.625,
-      "grad_norm": 10.95101547241211,
-      "learning_rate": 9.200000000000002e-06,
-      "loss": 2.3747,
+      "grad_norm": 8.544885635375977,
+      "learning_rate": 3.75e-05,
+      "loss": 1.8228,
       "step": 20
     },
     {
       "epoch": 1.25,
-      "grad_norm": 7.747081279754639,
-      "learning_rate": 8.2e-06,
-      "loss": 1.2169,
+      "grad_norm": 6.32118558883667,
+      "learning_rate": 3.75e-05,
+      "loss": 0.6803,
       "step": 40
     },
     {
       "epoch": 1.25,
-      "eval_loss": 1.0360029935836792,
-      "eval_runtime": 574.5339,
-      "eval_samples_per_second": 1.741,
-      "eval_steps_per_second": 1.741,
-      "eval_wer": 91.60809552036582,
+      "eval_loss": 0.6102083325386047,
+      "eval_runtime": 542.5987,
+      "eval_samples_per_second": 1.843,
+      "eval_steps_per_second": 1.843,
+      "eval_wer": 65.94411515664692,
       "step": 40
     },
     {
       "epoch": 1.875,
-      "grad_norm": 7.32460355758667,
-      "learning_rate": 7.2000000000000005e-06,
-      "loss": 0.9058,
+      "grad_norm": 6.4600510597229,
+      "learning_rate": 3.75e-05,
+      "loss": 0.4966,
       "step": 60
     },
     {
       "epoch": 2.5,
-      "grad_norm": 6.835532188415527,
-      "learning_rate": 6.200000000000001e-06,
-      "loss": 0.7302,
+      "grad_norm": 5.194849491119385,
+      "learning_rate": 3.75e-05,
+      "loss": 0.3649,
       "step": 80
     },
     {
       "epoch": 2.5,
-      "eval_loss": 0.7368654608726501,
-      "eval_runtime": 549.562,
-      "eval_samples_per_second": 1.82,
-      "eval_steps_per_second": 1.82,
-      "eval_wer": 79.80353967313066,
+      "eval_loss": 0.5036975741386414,
+      "eval_runtime": 540.6408,
+      "eval_samples_per_second": 1.85,
+      "eval_steps_per_second": 1.85,
+      "eval_wer": 58.31498729889923,
       "step": 80
     },
     {
       "epoch": 3.125,
-      "grad_norm": 6.408384323120117,
-      "learning_rate": 5.2e-06,
-      "loss": 0.6453,
+      "grad_norm": 4.111001014709473,
+      "learning_rate": 3.75e-05,
+      "loss": 0.3075,
       "step": 100
     },
     {
       "epoch": 3.75,
-      "grad_norm": 5.809150695800781,
-      "learning_rate": 4.2000000000000004e-06,
-      "loss": 0.5917,
+      "grad_norm": 4.358007431030273,
+      "learning_rate": 3.75e-05,
+      "loss": 0.2296,
       "step": 120
     },
     {
       "epoch": 3.75,
-      "eval_loss": 0.6504533886909485,
-      "eval_runtime": 548.8688,
-      "eval_samples_per_second": 1.822,
-      "eval_steps_per_second": 1.822,
-      "eval_wer": 69.98899144720129,
+      "eval_loss": 0.48966631293296814,
+      "eval_runtime": 538.04,
+      "eval_samples_per_second": 1.859,
+      "eval_steps_per_second": 1.859,
+      "eval_wer": 66.71464860287891,
       "step": 120
     },
     {
       "epoch": 4.375,
-      "grad_norm": 5.724637508392334,
-      "learning_rate": 3.2000000000000003e-06,
-      "loss": 0.5356,
+      "grad_norm": 4.111566066741943,
+      "learning_rate": 3.75e-05,
+      "loss": 0.1811,
       "step": 140
     },
     {
       "epoch": 5.0,
-      "grad_norm": 10.124310493469238,
-      "learning_rate": 2.2e-06,
-      "loss": 0.5156,
+      "grad_norm": 6.702174186706543,
+      "learning_rate": 3.75e-05,
+      "loss": 0.1564,
       "step": 160
     },
     {
       "epoch": 5.0,
-      "eval_loss": 0.6157090663909912,
-      "eval_runtime": 549.7895,
-      "eval_samples_per_second": 1.819,
-      "eval_steps_per_second": 1.819,
-      "eval_wer": 70.4293335591498,
+      "eval_loss": 0.5130774974822998,
+      "eval_runtime": 529.7879,
+      "eval_samples_per_second": 1.888,
+      "eval_steps_per_second": 1.888,
+      "eval_wer": 65.34292972057578,
       "step": 160
     },
     {
       "epoch": 5.625,
-      "grad_norm": 4.947811126708984,
-      "learning_rate": 1.2000000000000002e-06,
-      "loss": 0.4986,
+      "grad_norm": 3.0882856845855713,
+      "learning_rate": 3.75e-05,
+      "loss": 0.101,
       "step": 180
     },
     {
       "epoch": 6.25,
-      "grad_norm": 4.541861057281494,
-      "learning_rate": 2.0000000000000002e-07,
-      "loss": 0.481,
+      "grad_norm": 4.524318695068359,
+      "learning_rate": 3.75e-05,
+      "loss": 0.0905,
       "step": 200
     },
     {
       "epoch": 6.25,
-      "eval_loss": 0.6050636768341064,
-      "eval_runtime": 552.964,
-      "eval_samples_per_second": 1.808,
-      "eval_steps_per_second": 1.808,
-      "eval_wer": 70.03980015242611,
+      "eval_loss": 0.5503659248352051,
+      "eval_runtime": 587.5545,
+      "eval_samples_per_second": 1.702,
+      "eval_steps_per_second": 1.702,
+      "eval_wer": 63.903471634208294,
       "step": 200
     },
     {
       "epoch": 6.25,
       "step": 200,
       "total_flos": 1.5401574531072e+17,
-      "train_loss": 0.8495243072509766,
-      "train_runtime": 3592.94,
-      "train_samples_per_second": 1.781,
+      "train_loss": 0.44307032763957976,
+      "train_runtime": 3561.8571,
+      "train_samples_per_second": 1.797,
       "train_steps_per_second": 0.056
     }
   ],