Upload 10 files

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d6b0967b5ab414903ec82cb4eb256dc82f207ea195239a335fbb7c0f1e434f7c
+oid sha256:6f5204db756e34f799719358f88672c8b4c6026f91eebda145d29f6b5a859364
 size 377851056

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ba2d5edcb13e7e96adbb0ca65cf4d6241e0e76c2d3114a8ecf24fc189d00c1bb
+oid sha256:5c63dd48407ceec67dee01431e0a481bb387803d9ef4f6a5541ecfead1b611e8
 size 377899102

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1e731ad63e753d284193b80f173a2ff1be45026693bc2996a76b495f4e45667
+oid sha256:b3913ae9493cc4baad846c1b707d069986928d1245d876521c5d6546dda5464f
 size 14244

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:da279824d75da67ee43e4d57e1cc5b4735e62b87cb30e4ce9cc7d11a432023a9
+oid sha256:ea297f7ab56af8b17d03b96627bb16862b6bf5fc7088f8cd1f0a0f4093d4464c
 size 1064

trainer_state.json

@@ -1,198 +1,298 @@
 {
-  "best_metric": 0.051292333751916885,
-  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-4500",
-  "epoch": 10.0,
+  "best_metric": 0.05392587557435036,
+  "best_model_checkpoint": "./wav2vec2-base-demo/checkpoint-7000",
+  "epoch": 14.553014553014552,
   "eval_steps": 500,
-  "global_step": 4810,
+  "global_step": 7000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 1.0395010395010396,
-      "grad_norm": 1.8444721698760986,
+      "grad_norm": 2.4276885986328125,
       "learning_rate": 4.99e-05,
-      "loss": 0.1188,
+      "loss": 0.0935,
       "step": 500
     },
     {
       "epoch": 1.0395010395010396,
-      "eval_cer": 0.035194462663447716,
-      "eval_loss": 0.06642936915159225,
-      "eval_mer": 0.10481941437848924,
-      "eval_runtime": 34.4969,
-      "eval_samples_per_second": 44.555,
-      "eval_steps_per_second": 5.595,
-      "eval_wer": 0.10539580707984877,
-      "eval_wil": 0.1892106514423464,
-      "eval_wip": 0.8107893485576536,
+      "eval_cer": 0.03704485933935964,
+      "eval_loss": 0.07600321620702744,
+      "eval_mer": 0.10765687967760507,
+      "eval_runtime": 34.3532,
+      "eval_samples_per_second": 44.741,
+      "eval_steps_per_second": 5.618,
+      "eval_wer": 0.1080737444373808,
+      "eval_wil": 0.19400502379268658,
+      "eval_wip": 0.8059949762073134,
       "step": 500
     },
     {
       "epoch": 2.079002079002079,
-      "grad_norm": 1.2975250482559204,
-      "learning_rate": 4.421113689095128e-05,
-      "loss": 0.1151,
+      "grad_norm": 1.2353624105453491,
+      "learning_rate": 4.8671291355389544e-05,
+      "loss": 0.0914,
       "step": 1000
     },
     {
       "epoch": 2.079002079002079,
-      "eval_cer": 0.034538578545611855,
-      "eval_loss": 0.062206532806158066,
-      "eval_mer": 0.10494778291388461,
-      "eval_runtime": 34.1611,
-      "eval_samples_per_second": 44.993,
-      "eval_steps_per_second": 5.65,
-      "eval_wer": 0.10533852674991408,
-      "eval_wil": 0.19021462347071394,
-      "eval_wip": 0.8097853765292861,
+      "eval_cer": 0.037467263664779926,
+      "eval_loss": 0.08513512462377548,
+      "eval_mer": 0.1085740644640489,
+      "eval_runtime": 34.2139,
+      "eval_samples_per_second": 44.923,
+      "eval_steps_per_second": 5.641,
+      "eval_wer": 0.10882505923828238,
+      "eval_wil": 0.19587239193616368,
+      "eval_wip": 0.8041276080638363,
       "step": 1000
     },
     {
       "epoch": 3.1185031185031185,
-      "grad_norm": 0.5322809815406799,
-      "learning_rate": 3.8410672853828304e-05,
-      "loss": 0.1235,
+      "grad_norm": 0.4545057415962219,
+      "learning_rate": 4.733724653148346e-05,
+      "loss": 0.1007,
       "step": 1500
     },
     {
       "epoch": 3.1185031185031185,
-      "eval_cer": 0.03377105457793159,
-      "eval_loss": 0.062230244278907776,
-      "eval_mer": 0.10482995141468991,
-      "eval_runtime": 34.2894,
-      "eval_samples_per_second": 44.824,
-      "eval_steps_per_second": 5.629,
-      "eval_wer": 0.10505212510024058,
-      "eval_wil": 0.19100713135160063,
-      "eval_wip": 0.8089928686483994,
+      "eval_cer": 0.03667877559066205,
+      "eval_loss": 0.08273813873529434,
+      "eval_mer": 0.10847477196628565,
+      "eval_runtime": 34.1503,
+      "eval_samples_per_second": 45.007,
+      "eval_steps_per_second": 5.651,
+      "eval_wer": 0.10859388545338959,
+      "eval_wil": 0.1960486347087158,
+      "eval_wip": 0.8039513652912842,
       "step": 1500
     },
     {
       "epoch": 4.158004158004158,
-      "grad_norm": 1.327663779258728,
-      "learning_rate": 3.262180974477959e-05,
-      "loss": 0.1068,
+      "grad_norm": 0.7860220074653625,
+      "learning_rate": 4.600586979722519e-05,
+      "loss": 0.0883,
       "step": 2000
     },
     {
       "epoch": 4.158004158004158,
-      "eval_cer": 0.03463626341422571,
-      "eval_loss": 0.06272279471158981,
-      "eval_mer": 0.10423917384606607,
-      "eval_runtime": 34.2637,
-      "eval_samples_per_second": 44.858,
-      "eval_steps_per_second": 5.633,
-      "eval_wer": 0.10465116279069768,
-      "eval_wil": 0.18912080601859016,
-      "eval_wip": 0.8108791939814098,
+      "eval_cer": 0.03734054236715384,
+      "eval_loss": 0.08130443096160889,
+      "eval_mer": 0.10836359444348377,
+      "eval_runtime": 34.1136,
+      "eval_samples_per_second": 45.055,
+      "eval_steps_per_second": 5.658,
+      "eval_wer": 0.10865167889961279,
+      "eval_wil": 0.19535745081720401,
+      "eval_wip": 0.804642549182796,
       "step": 2000
     },
     {
       "epoch": 5.197505197505198,
-      "grad_norm": 1.7840451002120972,
-      "learning_rate": 2.6821345707656613e-05,
-      "loss": 0.1123,
+      "grad_norm": 5.241793632507324,
+      "learning_rate": 4.46718249733191e-05,
+      "loss": 0.0988,
       "step": 2500
     },
     {
       "epoch": 5.197505197505198,
-      "eval_cer": 0.033366360122245635,
-      "eval_loss": 0.06801871210336685,
-      "eval_mer": 0.10333599908602764,
-      "eval_runtime": 34.1023,
-      "eval_samples_per_second": 45.07,
-      "eval_steps_per_second": 5.659,
-      "eval_wer": 0.10362011685187307,
-      "eval_wil": 0.18807839647659041,
-      "eval_wip": 0.8119216035234096,
+      "eval_cer": 0.03660837486975867,
+      "eval_loss": 0.07141255587339401,
+      "eval_mer": 0.10702128887093983,
+      "eval_runtime": 34.0472,
+      "eval_samples_per_second": 45.143,
+      "eval_steps_per_second": 5.669,
+      "eval_wer": 0.10720684274403283,
+      "eval_wil": 0.1936232791597472,
+      "eval_wip": 0.8063767208402528,
       "step": 2500
     },
     {
       "epoch": 6.237006237006237,
-      "grad_norm": 0.5479459762573242,
-      "learning_rate": 2.1032482598607887e-05,
-      "loss": 0.1066,
+      "grad_norm": 1.3025474548339844,
+      "learning_rate": 4.334044823906083e-05,
+      "loss": 0.0962,
       "step": 3000
     },
     {
       "epoch": 6.237006237006237,
-      "eval_cer": 0.03377105457793159,
-      "eval_loss": 0.0692114531993866,
-      "eval_mer": 0.10309631065746706,
-      "eval_runtime": 34.3278,
-      "eval_samples_per_second": 44.774,
-      "eval_steps_per_second": 5.622,
-      "eval_wer": 0.10356283652193837,
-      "eval_wil": 0.18640279212307265,
-      "eval_wip": 0.8135972078769274,
+      "eval_cer": 0.035636844921291995,
+      "eval_loss": 0.06766052544116974,
+      "eval_mer": 0.1056455798396863,
+      "eval_runtime": 34.1329,
+      "eval_samples_per_second": 45.03,
+      "eval_steps_per_second": 5.654,
+      "eval_wer": 0.10587759348089927,
+      "eval_wil": 0.1915670539174873,
+      "eval_wip": 0.8084329460825127,
       "step": 3000
     },
     {
       "epoch": 7.276507276507276,
-      "grad_norm": 2.529324769973755,
-      "learning_rate": 1.5243619489559166e-05,
-      "loss": 0.0985,
+      "grad_norm": 1.1820029020309448,
+      "learning_rate": 4.200907150480256e-05,
+      "loss": 0.0951,
       "step": 3500
     },
     {
       "epoch": 7.276507276507276,
-      "eval_cer": 0.032780250910562525,
-      "eval_loss": 0.0544947050511837,
-      "eval_mer": 0.1027412906910337,
-      "eval_runtime": 34.2034,
-      "eval_samples_per_second": 44.937,
-      "eval_steps_per_second": 5.643,
-      "eval_wer": 0.10304731355252607,
-      "eval_wil": 0.1872840097439975,
-      "eval_wip": 0.8127159902560025,
+      "eval_cer": 0.03551012362366591,
+      "eval_loss": 0.06142089143395424,
+      "eval_mer": 0.10593122547888299,
+      "eval_runtime": 34.0865,
+      "eval_samples_per_second": 45.091,
+      "eval_steps_per_second": 5.662,
+      "eval_wer": 0.10610876726579206,
+      "eval_wil": 0.19198195094690818,
+      "eval_wip": 0.8080180490530918,
       "step": 3500
     },
     {
       "epoch": 8.316008316008316,
-      "grad_norm": 2.9378857612609863,
-      "learning_rate": 9.454756380510442e-06,
-      "loss": 0.1145,
+      "grad_norm": 3.633211851119995,
+      "learning_rate": 4.0677694770544295e-05,
+      "loss": 0.1161,
       "step": 4000
     },
     {
       "epoch": 8.316008316008316,
-      "eval_cer": 0.03286398079794583,
-      "eval_loss": 0.05142713338136673,
-      "eval_mer": 0.10264465642314503,
-      "eval_runtime": 34.3124,
-      "eval_samples_per_second": 44.794,
-      "eval_steps_per_second": 5.625,
-      "eval_wer": 0.10293275289265666,
-      "eval_wil": 0.18734075205582912,
-      "eval_wip": 0.8126592479441709,
+      "eval_cer": 0.03551012362366591,
+      "eval_loss": 0.06067756190896034,
+      "eval_mer": 0.10610691424946658,
+      "eval_runtime": 34.0596,
+      "eval_samples_per_second": 45.127,
+      "eval_steps_per_second": 5.667,
+      "eval_wer": 0.10633994105068485,
+      "eval_wil": 0.19244783032392676,
+      "eval_wip": 0.8075521696760732,
       "step": 4000
     },
     {
       "epoch": 9.355509355509355,
-      "grad_norm": 1.8327735662460327,
-      "learning_rate": 3.654292343387471e-06,
-      "loss": 0.1128,
+      "grad_norm": 1.8992377519607544,
+      "learning_rate": 3.934364994663821e-05,
+      "loss": 0.1202,
       "step": 4500
     },
     {
       "epoch": 9.355509355509355,
-      "eval_cer": 0.03269652102317922,
-      "eval_loss": 0.051292333751916885,
-      "eval_mer": 0.10197669104204753,
-      "eval_runtime": 34.3592,
-      "eval_samples_per_second": 44.733,
-      "eval_steps_per_second": 5.617,
-      "eval_wer": 0.10224538893344026,
-      "eval_wil": 0.1861754557735349,
-      "eval_wip": 0.8138245442264651,
+      "eval_cer": 0.036453493283771225,
+      "eval_loss": 0.07183138281106949,
+      "eval_mer": 0.10711813567143516,
+      "eval_runtime": 34.0394,
+      "eval_samples_per_second": 45.154,
+      "eval_steps_per_second": 5.67,
+      "eval_wer": 0.10732242963647923,
+      "eval_wil": 0.19300207421517157,
+      "eval_wip": 0.8069979257848284,
       "step": 4500
+    },
+    {
+      "epoch": 10.395010395010395,
+      "grad_norm": 1.5443813800811768,
+      "learning_rate": 3.800960512273213e-05,
+      "loss": 0.1431,
+      "step": 5000
+    },
+    {
+      "epoch": 10.395010395010395,
+      "eval_cer": 0.036031088958350936,
+      "eval_loss": 0.0653519257903099,
+      "eval_mer": 0.10687726984492996,
+      "eval_runtime": 34.4227,
+      "eval_samples_per_second": 44.651,
+      "eval_steps_per_second": 5.607,
+      "eval_wer": 0.10714904929780963,
+      "eval_wil": 0.19299958643561255,
+      "eval_wip": 0.8070004135643875,
+      "step": 5000
+    },
+    {
+      "epoch": 11.434511434511435,
+      "grad_norm": 0.9764054417610168,
+      "learning_rate": 3.667556029882604e-05,
+      "loss": 0.1356,
+      "step": 5500
+    },
+    {
+      "epoch": 11.434511434511435,
+      "eval_cer": 0.03546788319112388,
+      "eval_loss": 0.07734616100788116,
+      "eval_mer": 0.10516307479543621,
+      "eval_runtime": 34.1476,
+      "eval_samples_per_second": 45.01,
+      "eval_steps_per_second": 5.652,
+      "eval_wer": 0.10547303935733687,
+      "eval_wil": 0.18957490448529868,
+      "eval_wip": 0.8104250955147013,
+      "step": 5500
+    },
+    {
+      "epoch": 12.474012474012474,
+      "grad_norm": 1.104174017906189,
+      "learning_rate": 3.5341515474919964e-05,
+      "loss": 0.1561,
+      "step": 6000
+    },
+    {
+      "epoch": 12.474012474012474,
+      "eval_cer": 0.035862127228182814,
+      "eval_loss": 0.06988665461540222,
+      "eval_mer": 0.10506015773415463,
+      "eval_runtime": 34.0048,
+      "eval_samples_per_second": 45.2,
+      "eval_steps_per_second": 5.676,
+      "eval_wer": 0.10547303935733687,
+      "eval_wil": 0.1890727356961288,
+      "eval_wip": 0.8109272643038712,
+      "step": 6000
+    },
+    {
+      "epoch": 13.513513513513514,
+      "grad_norm": 4.416908264160156,
+      "learning_rate": 3.400747065101387e-05,
+      "loss": 0.134,
+      "step": 6500
+    },
+    {
+      "epoch": 13.513513513513514,
+      "eval_cer": 0.0349750781448002,
+      "eval_loss": 0.0634101927280426,
+      "eval_mer": 0.10508709193678625,
+      "eval_runtime": 34.1003,
+      "eval_samples_per_second": 45.073,
+      "eval_steps_per_second": 5.66,
+      "eval_wer": 0.10529965901866728,
+      "eval_wil": 0.19050757507724403,
+      "eval_wip": 0.809492424922756,
+      "step": 6500
+    },
+    {
+      "epoch": 14.553014553014552,
+      "grad_norm": 0.7748957276344299,
+      "learning_rate": 3.267342582710779e-05,
+      "loss": 0.1304,
+      "step": 7000
+    },
+    {
+      "epoch": 14.553014553014552,
+      "eval_cer": 0.03490467742389682,
+      "eval_loss": 0.05392587557435036,
+      "eval_mer": 0.10498097544102386,
+      "eval_runtime": 34.1635,
+      "eval_samples_per_second": 44.99,
+      "eval_steps_per_second": 5.649,
+      "eval_wer": 0.10524186557244408,
+      "eval_wil": 0.18975677659084123,
+      "eval_wip": 0.8102432234091588,
+      "step": 7000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 4810,
+  "max_steps": 19240,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 40,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -201,12 +301,12 @@
         "should_evaluate": false,
         "should_log": false,
         "should_save": true,
-        "should_training_stop": true
+        "should_training_stop": false
       },
       "attributes": {}
     }
   },
-  "total_flos": 3.07474151587841e+18,
+  "total_flos": 4.4824764428392996e+18,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adb2dde227c901afcc4b8ad055a16de87aa03de53761c737e130e66fa8b7dd9a
+oid sha256:775696d645371d4025f4adc46331697961085adc3d0812695e8af16d5dc92a47
 size 5176