Training in progress, step 40

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:693d49ced50265cbb5903f02bab37d536606e9b6f2b7792cb8d77781ef564040
+oid sha256:537ac3362dbc63d6e031264d209ab42955124de5508e3d5644f3077de2bfc50d
 size 151099494

runs/Jun29_09-39-32_bharatsahaiyak-test/events.out.tfevents.1719662186.bharatsahaiyak-test.1854617.7

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:08321b45077baef003b5edad2a8a345b35fb9bd3f64ea79e55c05edd6382da10
+size 6396

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 31.703204047217536,
-  "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-80",
+  "best_metric": 29.34232715008432,
+  "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-120",
   "epoch": 6.349206349206349,
   "eval_steps": 40,
   "global_step": 200,
@@ -10,127 +10,127 @@
   "log_history": [
     {
       "epoch": 0.6349206349206349,
-      "grad_norm": 5.663797855377197,
+      "grad_norm": 6.46198034286499,
       "learning_rate": 3.75e-05,
-      "loss": 0.3885,
+      "loss": 0.3431,
       "step": 20
     },
     {
       "epoch": 1.2698412698412698,
-      "grad_norm": 4.554736137390137,
+      "grad_norm": 3.6580495834350586,
       "learning_rate": 3.75e-05,
-      "loss": 0.2894,
+      "loss": 0.257,
       "step": 40
     },
     {
       "epoch": 1.2698412698412698,
-      "eval_loss": 0.21782219409942627,
-      "eval_runtime": 55.5859,
-      "eval_samples_per_second": 1.799,
-      "eval_steps_per_second": 1.799,
-      "eval_wer": 32.96795952782462,
+      "eval_loss": 0.19398407638072968,
+      "eval_runtime": 55.7217,
+      "eval_samples_per_second": 1.795,
+      "eval_steps_per_second": 1.795,
+      "eval_wer": 30.94435075885329,
       "step": 40
     },
     {
       "epoch": 1.9047619047619047,
-      "grad_norm": 4.489897727966309,
+      "grad_norm": 4.121974468231201,
       "learning_rate": 3.75e-05,
-      "loss": 0.1965,
+      "loss": 0.1865,
       "step": 60
     },
     {
       "epoch": 2.5396825396825395,
-      "grad_norm": 2.790485382080078,
+      "grad_norm": 3.0512349605560303,
       "learning_rate": 3.75e-05,
-      "loss": 0.1333,
+      "loss": 0.1177,
       "step": 80
     },
     {
       "epoch": 2.5396825396825395,
-      "eval_loss": 0.2209174931049347,
-      "eval_runtime": 55.2087,
-      "eval_samples_per_second": 1.811,
-      "eval_steps_per_second": 1.811,
-      "eval_wer": 31.703204047217536,
+      "eval_loss": 0.1945016235113144,
+      "eval_runtime": 55.3505,
+      "eval_samples_per_second": 1.807,
+      "eval_steps_per_second": 1.807,
+      "eval_wer": 31.197301854974707,
       "step": 80
     },
     {
       "epoch": 3.1746031746031744,
-      "grad_norm": 2.2434844970703125,
+      "grad_norm": 2.3403635025024414,
       "learning_rate": 3.75e-05,
-      "loss": 0.1016,
+      "loss": 0.0903,
       "step": 100
     },
     {
       "epoch": 3.8095238095238093,
-      "grad_norm": 2.5447564125061035,
+      "grad_norm": 2.4136340618133545,
       "learning_rate": 3.75e-05,
-      "loss": 0.0599,
+      "loss": 0.0569,
       "step": 120
     },
     {
       "epoch": 3.8095238095238093,
-      "eval_loss": 0.23671233654022217,
-      "eval_runtime": 55.6056,
-      "eval_samples_per_second": 1.798,
-      "eval_steps_per_second": 1.798,
-      "eval_wer": 33.22091062394604,
+      "eval_loss": 0.20778781175613403,
+      "eval_runtime": 55.9933,
+      "eval_samples_per_second": 1.786,
+      "eval_steps_per_second": 1.786,
+      "eval_wer": 29.34232715008432,
       "step": 120
     },
     {
       "epoch": 4.444444444444445,
-      "grad_norm": 2.4546711444854736,
+      "grad_norm": 1.9897055625915527,
       "learning_rate": 3.75e-05,
-      "loss": 0.0414,
+      "loss": 0.0412,
       "step": 140
     },
     {
       "epoch": 5.079365079365079,
-      "grad_norm": 1.9287947416305542,
+      "grad_norm": 1.213258981704712,
       "learning_rate": 3.75e-05,
-      "loss": 0.0332,
+      "loss": 0.0313,
       "step": 160
     },
     {
       "epoch": 5.079365079365079,
-      "eval_loss": 0.24990424513816833,
-      "eval_runtime": 55.3919,
-      "eval_samples_per_second": 1.805,
-      "eval_steps_per_second": 1.805,
-      "eval_wer": 31.871838111298484,
+      "eval_loss": 0.2349608838558197,
+      "eval_runtime": 56.0059,
+      "eval_samples_per_second": 1.786,
+      "eval_steps_per_second": 1.786,
+      "eval_wer": 31.02866779089376,
       "step": 160
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 2.413437843322754,
+      "grad_norm": 1.2564868927001953,
       "learning_rate": 3.75e-05,
-      "loss": 0.0232,
+      "loss": 0.0198,
       "step": 180
     },
     {
       "epoch": 6.349206349206349,
-      "grad_norm": 2.503246784210205,
+      "grad_norm": 1.6000959873199463,
       "learning_rate": 3.75e-05,
-      "loss": 0.0189,
+      "loss": 0.0186,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
-      "eval_loss": 0.27371078729629517,
-      "eval_runtime": 55.2258,
-      "eval_samples_per_second": 1.811,
-      "eval_steps_per_second": 1.811,
-      "eval_wer": 33.05227655986509,
+      "eval_loss": 0.2441704422235489,
+      "eval_runtime": 55.4225,
+      "eval_samples_per_second": 1.804,
+      "eval_steps_per_second": 1.804,
+      "eval_wer": 30.6070826306914,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
       "step": 200,
       "total_flos": 1.5637915828224e+17,
-      "train_loss": 0.1285850764811039,
-      "train_runtime": 1129.5385,
-      "train_samples_per_second": 5.666,
-      "train_steps_per_second": 0.177
+      "train_loss": 0.11623679384589196,
+      "train_runtime": 1122.2532,
+      "train_samples_per_second": 5.703,
+      "train_steps_per_second": 0.178
     }
   ],
   "logging_steps": 20,