Training in progress, step 40

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:28d00ca3bfb780e4dc72d46545fa78a6ba7b4ff64ff7fca8ceca22f284d3373c
+oid sha256:ec43cf6eef1f5aa761a37e943c182de501d2bc55b4d9d8424a011f69ffa73a1e
 size 151099494

runs/Jun29_09-39-32_bharatsahaiyak-test/events.out.tfevents.1719657502.bharatsahaiyak-test.1854617.3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80ae0d0376797cbf383586bd9fc07951e66d4f11e270768f2186a245c1741852
+size 6396

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 41.31534569983137,
-  "best_model_checkpoint": "./whisper-tiny-hindi2_test/checkpoint-160",
+  "best_metric": 37.77403035413153,
+  "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": 5.291396141052246,
+      "grad_norm": 6.332822322845459,
       "learning_rate": 3.75e-05,
-      "loss": 0.486,
+      "loss": 0.4561,
       "step": 20
     },
     {
       "epoch": 1.2698412698412698,
-      "grad_norm": 4.965972423553467,
+      "grad_norm": 4.458395957946777,
       "learning_rate": 3.75e-05,
-      "loss": 0.3869,
+      "loss": 0.3384,
       "step": 40
     },
     {
       "epoch": 1.2698412698412698,
-      "eval_loss": 0.35420987010002136,
-      "eval_runtime": 54.9702,
-      "eval_samples_per_second": 1.819,
-      "eval_steps_per_second": 1.819,
-      "eval_wer": 46.964586846543,
+      "eval_loss": 0.300150990486145,
+      "eval_runtime": 55.4195,
+      "eval_samples_per_second": 1.804,
+      "eval_steps_per_second": 1.804,
+      "eval_wer": 39.62900505902192,
       "step": 40
     },
     {
       "epoch": 1.9047619047619047,
-      "grad_norm": 5.148536682128906,
+      "grad_norm": 3.924806594848633,
       "learning_rate": 3.75e-05,
-      "loss": 0.3043,
+      "loss": 0.2515,
       "step": 60
     },
     {
       "epoch": 2.5396825396825395,
-      "grad_norm": 4.049417972564697,
+      "grad_norm": 3.3504045009613037,
       "learning_rate": 3.75e-05,
-      "loss": 0.2215,
+      "loss": 0.1811,
       "step": 80
     },
     {
       "epoch": 2.5396825396825395,
-      "eval_loss": 0.3464047908782959,
-      "eval_runtime": 55.3552,
-      "eval_samples_per_second": 1.807,
-      "eval_steps_per_second": 1.807,
-      "eval_wer": 45.27824620573356,
+      "eval_loss": 0.2907748520374298,
+      "eval_runtime": 55.3083,
+      "eval_samples_per_second": 1.808,
+      "eval_steps_per_second": 1.808,
+      "eval_wer": 37.77403035413153,
       "step": 80
     },
     {
       "epoch": 3.1746031746031744,
-      "grad_norm": 3.4034841060638428,
+      "grad_norm": 3.752678155899048,
       "learning_rate": 3.75e-05,
-      "loss": 0.1888,
+      "loss": 0.1512,
       "step": 100
     },
     {
       "epoch": 3.8095238095238093,
-      "grad_norm": 4.5857930183410645,
+      "grad_norm": 3.3088107109069824,
       "learning_rate": 3.75e-05,
-      "loss": 0.1394,
+      "loss": 0.0989,
       "step": 120
     },
     {
       "epoch": 3.8095238095238093,
-      "eval_loss": 0.34948885440826416,
-      "eval_runtime": 54.9671,
-      "eval_samples_per_second": 1.819,
-      "eval_steps_per_second": 1.819,
-      "eval_wer": 43.00168634064081,
+      "eval_loss": 0.3248939514160156,
+      "eval_runtime": 55.1423,
+      "eval_samples_per_second": 1.813,
+      "eval_steps_per_second": 1.813,
+      "eval_wer": 39.79763912310287,
       "step": 120
     },
     {
       "epoch": 4.444444444444445,
-      "grad_norm": 3.3065757751464844,
+      "grad_norm": 3.1989450454711914,
       "learning_rate": 3.75e-05,
-      "loss": 0.0954,
+      "loss": 0.0738,
       "step": 140
     },
     {
       "epoch": 5.079365079365079,
-      "grad_norm": 2.241292953491211,
+      "grad_norm": 2.052924871444702,
       "learning_rate": 3.75e-05,
-      "loss": 0.0857,
+      "loss": 0.0623,
       "step": 160
     },
     {
       "epoch": 5.079365079365079,
-      "eval_loss": 0.37522462010383606,
-      "eval_runtime": 55.4528,
-      "eval_samples_per_second": 1.803,
-      "eval_steps_per_second": 1.803,
-      "eval_wer": 41.31534569983137,
+      "eval_loss": 0.33577826619148254,
+      "eval_runtime": 56.8542,
+      "eval_samples_per_second": 1.759,
+      "eval_steps_per_second": 1.759,
+      "eval_wer": 45.362563237774026,
       "step": 160
     },
     {
       "epoch": 5.714285714285714,
-      "grad_norm": 2.4962286949157715,
+      "grad_norm": 1.9767564535140991,
       "learning_rate": 3.75e-05,
-      "loss": 0.055,
+      "loss": 0.0377,
       "step": 180
     },
     {
       "epoch": 6.349206349206349,
-      "grad_norm": 2.652437210083008,
+      "grad_norm": 3.543325185775757,
       "learning_rate": 3.75e-05,
-      "loss": 0.0434,
+      "loss": 0.0344,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
-      "eval_loss": 0.40884342789649963,
-      "eval_runtime": 55.4835,
-      "eval_samples_per_second": 1.802,
-      "eval_steps_per_second": 1.802,
-      "eval_wer": 43.76053962900506,
+      "eval_loss": 0.3560102880001068,
+      "eval_runtime": 55.9482,
+      "eval_samples_per_second": 1.787,
+      "eval_steps_per_second": 1.787,
+      "eval_wer": 40.80944350758853,
       "step": 200
     },
     {
       "epoch": 6.349206349206349,
       "step": 200,
       "total_flos": 1.5637915828224e+17,
-      "train_loss": 0.2006339970231056,
-      "train_runtime": 1120.101,
-      "train_samples_per_second": 5.714,
-      "train_steps_per_second": 0.179
+      "train_loss": 0.16853376895189284,
+      "train_runtime": 1128.2867,
+      "train_samples_per_second": 5.672,
+      "train_steps_per_second": 0.177
     }
   ],
   "logging_steps": 20,