Training in progress, step 3500, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cb935ab771f33e20a7b505537eb78c6d9f0917f5a90dac3d0266df181fcefd2
+oid sha256:211c3e081cd4d45139c1b37769c82a39d49594bd6cb32f70c0f4dfa85757fa43
 size 966995080

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:be80de71962a2bed0e9e9eceb9370db35d5115beb28a844300a0d8836493df5a
+oid sha256:0c6b9f0ce0751785007dca8a03a72091fb50b8a1cf30251390dddfa7806b03fb
 size 1228890682

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b7d89181cfd314a69a1aa1b9ba842a99dedf4dca5b395632f38021ef8424361f
+oid sha256:8e3d8253d03fcee7ed2b170bbb1632a1a5520714c44e4406d8dab09e403df5b0
 size 14244

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7b5cc1ed35f48cc59384f56c8de34e461d08b0f64957aa14f6f8878a25fc4550
+oid sha256:2daaa0057b288ba4d4f56331182174ee76136e14690065d8b128321e93809fdf
 size 988

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c7d6af335768dfd8ff7047730bbedefbd3d6088c1613499f72ef42fc1e0161ba
+oid sha256:c382cdf70b3eb758a0dda9a8be2ad82136bc5362f13a6d2d16779614d18daa02
 size 1064

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 3000,
-  "best_metric": 44.32302984611291,
-  "best_model_checkpoint": "./whisper-small-warsh-quran-v3/checkpoint-3000",
-  "epoch": 0.6288648988575621,
+  "best_global_step": 3500,
+  "best_metric": 40.41509601312697,
+  "best_model_checkpoint": "./whisper-small-warsh-quran-v3/checkpoint-3500",
+  "epoch": 0.7336757153338225,
   "eval_steps": 500,
-  "global_step": 3000,
+  "global_step": 3500,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -902,6 +902,155 @@
       "eval_steps_per_second": 0.333,
       "eval_wer": 44.32302984611291,
       "step": 3000
+    },
+    {
+      "epoch": 0.6341054396813751,
+      "grad_norm": 3.61225962638855,
+      "learning_rate": 5.180869565217391e-06,
+      "loss": 0.2671,
+      "step": 3025
+    },
+    {
+      "epoch": 0.6393459805051881,
+      "grad_norm": 6.848639011383057,
+      "learning_rate": 5.137391304347826e-06,
+      "loss": 0.2305,
+      "step": 3050
+    },
+    {
+      "epoch": 0.6445865213290012,
+      "grad_norm": 5.066479206085205,
+      "learning_rate": 5.0939130434782615e-06,
+      "loss": 0.2075,
+      "step": 3075
+    },
+    {
+      "epoch": 0.6498270621528142,
+      "grad_norm": 10.204203605651855,
+      "learning_rate": 5.050434782608697e-06,
+      "loss": 0.2618,
+      "step": 3100
+    },
+    {
+      "epoch": 0.6550676029766271,
+      "grad_norm": 4.369320869445801,
+      "learning_rate": 5.006956521739132e-06,
+      "loss": 0.3662,
+      "step": 3125
+    },
+    {
+      "epoch": 0.6603081438004402,
+      "grad_norm": 5.38928747177124,
+      "learning_rate": 4.963478260869566e-06,
+      "loss": 0.3384,
+      "step": 3150
+    },
+    {
+      "epoch": 0.6655486846242532,
+      "grad_norm": 13.416193962097168,
+      "learning_rate": 4.92e-06,
+      "loss": 0.3091,
+      "step": 3175
+    },
+    {
+      "epoch": 0.6707892254480663,
+      "grad_norm": 3.561743974685669,
+      "learning_rate": 4.8765217391304355e-06,
+      "loss": 0.2343,
+      "step": 3200
+    },
+    {
+      "epoch": 0.6760297662718793,
+      "grad_norm": 15.610285758972168,
+      "learning_rate": 4.83304347826087e-06,
+      "loss": 0.356,
+      "step": 3225
+    },
+    {
+      "epoch": 0.6812703070956923,
+      "grad_norm": 7.896173477172852,
+      "learning_rate": 4.789565217391305e-06,
+      "loss": 0.2121,
+      "step": 3250
+    },
+    {
+      "epoch": 0.6865108479195053,
+      "grad_norm": 6.182321548461914,
+      "learning_rate": 4.746086956521739e-06,
+      "loss": 0.2922,
+      "step": 3275
+    },
+    {
+      "epoch": 0.6917513887433183,
+      "grad_norm": 14.184751510620117,
+      "learning_rate": 4.702608695652174e-06,
+      "loss": 0.3107,
+      "step": 3300
+    },
+    {
+      "epoch": 0.6969919295671313,
+      "grad_norm": 3.7635350227355957,
+      "learning_rate": 4.6591304347826085e-06,
+      "loss": 0.2339,
+      "step": 3325
+    },
+    {
+      "epoch": 0.7022324703909444,
+      "grad_norm": 4.777770519256592,
+      "learning_rate": 4.615652173913044e-06,
+      "loss": 0.2221,
+      "step": 3350
+    },
+    {
+      "epoch": 0.7074730112147574,
+      "grad_norm": 6.8489556312561035,
+      "learning_rate": 4.572173913043479e-06,
+      "loss": 0.3584,
+      "step": 3375
+    },
+    {
+      "epoch": 0.7127135520385703,
+      "grad_norm": 8.557856559753418,
+      "learning_rate": 4.528695652173913e-06,
+      "loss": 0.3055,
+      "step": 3400
+    },
+    {
+      "epoch": 0.7179540928623834,
+      "grad_norm": 6.584381580352783,
+      "learning_rate": 4.485217391304348e-06,
+      "loss": 0.2357,
+      "step": 3425
+    },
+    {
+      "epoch": 0.7231946336861964,
+      "grad_norm": 4.871785640716553,
+      "learning_rate": 4.4417391304347825e-06,
+      "loss": 0.2408,
+      "step": 3450
+    },
+    {
+      "epoch": 0.7284351745100094,
+      "grad_norm": 8.363597869873047,
+      "learning_rate": 4.398260869565218e-06,
+      "loss": 0.2942,
+      "step": 3475
+    },
+    {
+      "epoch": 0.7336757153338225,
+      "grad_norm": 8.755147933959961,
+      "learning_rate": 4.354782608695653e-06,
+      "loss": 0.2587,
+      "step": 3500
+    },
+    {
+      "epoch": 0.7336757153338225,
+      "eval_loss": 0.271077036857605,
+      "eval_runtime": 7162.8947,
+      "eval_samples_per_second": 1.332,
+      "eval_steps_per_second": 0.333,
+      "eval_wer": 40.41509601312697,
+      "step": 3500
     }
   ],
   "logging_steps": 25,
@@ -921,7 +1070,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.92604960768e+18,
+  "total_flos": 8.08039120896e+18,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null