Training in progress, step 40

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:704f03dd6d6a4a15c353dc922b1cd8801ee537c707ebb81d453328220dd822e4
+oid sha256:d66909332438e143fe70e0c29ee54f40db0d656b2f2c855e1fb48504b5799f5c
 size 151099494

runs/Jun29_09-39-32_bharatsahaiyak-test/events.out.tfevents.1719661023.bharatsahaiyak-test.1854617.6

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1687f609fdcc4d6a59acca2f6a6554d4482f90f26a6f94ad333c7808c79a2fed
+size 6396

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 32.377740303541316,
-  "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-200",
+  "best_metric": 31.703204047217536,
+  "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-80",
   "epoch": 6.349206349206349,
   "eval_steps": 40,
   "global_step": 200,
@@ -10,127 +10,127 @@
   "log_history": [
     {
       "epoch": 0.6349206349206349,
-      "grad_norm": 5.7862091064453125,
+      "grad_norm": 5.663797855377197,
       "learning_rate": 3.75e-05,
-      "loss": 0.3639,
+      "loss": 0.3885,
       "step": 20
     },
     {
       "epoch": 1.2698412698412698,
-      "grad_norm": 4.1368255615234375,
+      "grad_norm": 4.554736137390137,
       "learning_rate": 3.75e-05,
-      "loss": 0.2955,
+      "loss": 0.2894,
       "step": 40
     },
     {
       "epoch": 1.2698412698412698,
-      "eval_loss": 0.23172235488891602,
-      "eval_runtime": 56.6839,
-      "eval_samples_per_second": 1.764,
-      "eval_steps_per_second": 1.764,
-      "eval_wer": 36.00337268128162,
+      "eval_loss": 0.21782219409942627,
+      "eval_runtime": 55.5859,
+      "eval_samples_per_second": 1.799,
+      "eval_steps_per_second": 1.799,
+      "eval_wer": 32.96795952782462,
       "step": 40
     },
     {
       "epoch": 1.9047619047619047,
-      "grad_norm": 4.015283584594727,
+      "grad_norm": 4.489897727966309,
       "learning_rate": 3.75e-05,
-      "loss": 0.2134,
+      "loss": 0.1965,
       "step": 60
     },
     {
       "epoch": 2.5396825396825395,
-      "grad_norm": 3.215930223464966,
+      "grad_norm": 2.790485382080078,
       "learning_rate": 3.75e-05,
-      "loss": 0.1522,
+      "loss": 0.1333,
       "step": 80
     },
     {
       "epoch": 2.5396825396825395,
-      "eval_loss": 0.22003404796123505,
-      "eval_runtime": 55.4903,
-      "eval_samples_per_second": 1.802,
-      "eval_steps_per_second": 1.802,
-      "eval_wer": 34.40134907251265,
+      "eval_loss": 0.2209174931049347,
+      "eval_runtime": 55.2087,
+      "eval_samples_per_second": 1.811,
+      "eval_steps_per_second": 1.811,
+      "eval_wer": 31.703204047217536,
       "step": 80
     },
     {
       "epoch": 3.1746031746031744,
-      "grad_norm": 2.828727960586548,
+      "grad_norm": 2.2434844970703125,
       "learning_rate": 3.75e-05,
-      "loss": 0.1209,
+      "loss": 0.1016,
       "step": 100
     },
     {
       "epoch": 3.8095238095238093,
-      "grad_norm": 2.526444673538208,
+      "grad_norm": 2.5447564125061035,
       "learning_rate": 3.75e-05,
-      "loss": 0.0819,
+      "loss": 0.0599,
       "step": 120
     },
     {
       "epoch": 3.8095238095238093,
-      "eval_loss": 0.22554056346416473,
-      "eval_runtime": 54.7978,
-      "eval_samples_per_second": 1.825,
-      "eval_steps_per_second": 1.825,
-      "eval_wer": 32.54637436762226,
+      "eval_loss": 0.23671233654022217,
+      "eval_runtime": 55.6056,
+      "eval_samples_per_second": 1.798,
+      "eval_steps_per_second": 1.798,
+      "eval_wer": 33.22091062394604,
       "step": 120
     },
     {
       "epoch": 4.444444444444445,
-      "grad_norm": 2.135953664779663,
+      "grad_norm": 2.4546711444854736,
       "learning_rate": 3.75e-05,
-      "loss": 0.061,
+      "loss": 0.0414,
       "step": 140
     },
     {
       "epoch": 5.079365079365079,
-      "grad_norm": 2.1400399208068848,
+      "grad_norm": 1.9287947416305542,
       "learning_rate": 3.75e-05,
-      "loss": 0.0463,
+      "loss": 0.0332,
       "step": 160
     },
     {
       "epoch": 5.079365079365079,
-      "eval_loss": 0.24825386703014374,
-      "eval_runtime": 54.848,
-      "eval_samples_per_second": 1.823,
-      "eval_steps_per_second": 1.823,
-      "eval_wer": 32.79932546374368,
+      "eval_loss": 0.24990424513816833,
+      "eval_runtime": 55.3919,
+      "eval_samples_per_second": 1.805,
+      "eval_steps_per_second": 1.805,
+      "eval_wer": 31.871838111298484,
       "step": 160
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 1.9905014038085938,
+      "grad_norm": 2.413437843322754,
       "learning_rate": 3.75e-05,
-      "loss": 0.0304,
+      "loss": 0.0232,
       "step": 180
     },
     {
       "epoch": 6.349206349206349,
-      "grad_norm": 1.8886462450027466,
+      "grad_norm": 2.503246784210205,
       "learning_rate": 3.75e-05,
-      "loss": 0.0289,
+      "loss": 0.0189,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
-      "eval_loss": 0.2673737406730652,
-      "eval_runtime": 55.0065,
-      "eval_samples_per_second": 1.818,
-      "eval_steps_per_second": 1.818,
-      "eval_wer": 32.377740303541316,
+      "eval_loss": 0.27371078729629517,
+      "eval_runtime": 55.2258,
+      "eval_samples_per_second": 1.811,
+      "eval_steps_per_second": 1.811,
+      "eval_wer": 33.05227655986509,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
       "step": 200,
       "total_flos": 1.5637915828224e+17,
-      "train_loss": 0.13942732244729997,
-      "train_runtime": 1124.5585,
-      "train_samples_per_second": 5.691,
-      "train_steps_per_second": 0.178
+      "train_loss": 0.1285850764811039,
+      "train_runtime": 1129.5385,
+      "train_samples_per_second": 5.666,
+      "train_steps_per_second": 0.177
     }
   ],
   "logging_steps": 20,