{
"best_metric": 40.74074074074074,
"best_model_checkpoint": "all_lang_models/malayalam_models/whisper-medium-ml_alldata_multigpu/checkpoint-4300",
"epoch": 3.437250199840128,
"global_step": 4300,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 2.2570200577939087e-06,
"loss": 2.8955,
"step": 10
},
{
"epoch": 0.02,
"learning_rate": 5.269504270732932e-06,
"loss": 1.5027,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 6.44164771270531e-06,
"loss": 0.9079,
"step": 30
},
{
"epoch": 0.03,
"learning_rate": 7.1833237074248565e-06,
"loss": 0.5788,
"step": 40
},
{
"epoch": 0.04,
"learning_rate": 7.727115553011708e-06,
"loss": 0.4256,
"step": 50
},
{
"epoch": 0.05,
"learning_rate": 8.156679637224794e-06,
"loss": 0.3371,
"step": 60
},
{
"epoch": 0.06,
"learning_rate": 8.511772784687228e-06,
"loss": 0.284,
"step": 70
},
{
"epoch": 0.06,
"learning_rate": 8.814440832045515e-06,
"loss": 0.236,
"step": 80
},
{
"epoch": 0.07,
"learning_rate": 9.078190720115011e-06,
"loss": 0.2057,
"step": 90
},
{
"epoch": 0.08,
"learning_rate": 9.311900181269438e-06,
"loss": 0.1893,
"step": 100
},
{
"epoch": 0.08,
"eval_loss": 0.1680908203125,
"eval_runtime": 263.3879,
"eval_samples_per_second": 6.709,
"eval_steps_per_second": 0.053,
"eval_wer": 67.26838883701629,
"step": 100
},
{
"epoch": 0.09,
"learning_rate": 9.521717895177596e-06,
"loss": 0.1751,
"step": 110
},
{
"epoch": 0.1,
"learning_rate": 9.712078808748597e-06,
"loss": 0.1651,
"step": 120
},
{
"epoch": 0.1,
"learning_rate": 9.886287204237489e-06,
"loss": 0.1528,
"step": 130
},
{
"epoch": 0.11,
"learning_rate": 9.998445998446e-06,
"loss": 0.1467,
"step": 140
},
{
"epoch": 0.12,
"learning_rate": 9.990675990675992e-06,
"loss": 0.1394,
"step": 150
},
{
"epoch": 0.13,
"learning_rate": 9.982905982905984e-06,
"loss": 0.1352,
"step": 160
},
{
"epoch": 0.14,
"learning_rate": 9.975135975135975e-06,
"loss": 0.1306,
"step": 170
},
{
"epoch": 0.14,
"learning_rate": 9.967365967365968e-06,
"loss": 0.1257,
"step": 180
},
{
"epoch": 0.15,
"learning_rate": 9.95959595959596e-06,
"loss": 0.1251,
"step": 190
},
{
"epoch": 0.16,
"learning_rate": 9.951825951825953e-06,
"loss": 0.1207,
"step": 200
},
{
"epoch": 0.16,
"eval_loss": 0.109130859375,
"eval_runtime": 257.5203,
"eval_samples_per_second": 6.862,
"eval_steps_per_second": 0.054,
"eval_wer": 57.37109658678287,
"step": 200
},
{
"epoch": 0.17,
"learning_rate": 9.944055944055945e-06,
"loss": 0.1169,
"step": 210
},
{
"epoch": 0.18,
"learning_rate": 9.936285936285938e-06,
"loss": 0.1128,
"step": 220
},
{
"epoch": 0.18,
"learning_rate": 9.928515928515929e-06,
"loss": 0.1114,
"step": 230
},
{
"epoch": 0.19,
"learning_rate": 9.920745920745921e-06,
"loss": 0.1105,
"step": 240
},
{
"epoch": 0.2,
"learning_rate": 9.912975912975914e-06,
"loss": 0.1099,
"step": 250
},
{
"epoch": 0.21,
"learning_rate": 9.905205905205906e-06,
"loss": 0.1074,
"step": 260
},
{
"epoch": 0.22,
"learning_rate": 9.897435897435899e-06,
"loss": 0.1039,
"step": 270
},
{
"epoch": 0.22,
"learning_rate": 9.88966588966589e-06,
"loss": 0.1027,
"step": 280
},
{
"epoch": 0.23,
"learning_rate": 9.881895881895884e-06,
"loss": 0.1014,
"step": 290
},
{
"epoch": 0.24,
"learning_rate": 9.874125874125875e-06,
"loss": 0.1023,
"step": 300
},
{
"epoch": 0.24,
"eval_loss": 0.0926513671875,
"eval_runtime": 253.7805,
"eval_samples_per_second": 6.963,
"eval_steps_per_second": 0.055,
"eval_wer": 52.78037140782239,
"step": 300
},
{
"epoch": 0.25,
"learning_rate": 9.866355866355867e-06,
"loss": 0.0984,
"step": 310
},
{
"epoch": 0.26,
"learning_rate": 9.85858585858586e-06,
"loss": 0.1005,
"step": 320
},
{
"epoch": 0.26,
"learning_rate": 9.85081585081585e-06,
"loss": 0.0949,
"step": 330
},
{
"epoch": 0.27,
"learning_rate": 9.843045843045845e-06,
"loss": 0.0942,
"step": 340
},
{
"epoch": 0.28,
"learning_rate": 9.835275835275836e-06,
"loss": 0.0942,
"step": 350
},
{
"epoch": 0.29,
"learning_rate": 9.827505827505828e-06,
"loss": 0.0926,
"step": 360
},
{
"epoch": 0.3,
"learning_rate": 9.81973581973582e-06,
"loss": 0.0903,
"step": 370
},
{
"epoch": 0.3,
"learning_rate": 9.811965811965812e-06,
"loss": 0.0875,
"step": 380
},
{
"epoch": 0.31,
"learning_rate": 9.804195804195806e-06,
"loss": 0.0908,
"step": 390
},
{
"epoch": 0.32,
"learning_rate": 9.796425796425797e-06,
"loss": 0.0927,
"step": 400
},
{
"epoch": 0.32,
"eval_loss": 0.083740234375,
"eval_runtime": 234.5774,
"eval_samples_per_second": 7.533,
"eval_steps_per_second": 0.06,
"eval_wer": 52.137151156758996,
"step": 400
},
{
"epoch": 0.33,
"learning_rate": 9.788655788655789e-06,
"loss": 0.0893,
"step": 410
},
{
"epoch": 0.34,
"learning_rate": 9.780885780885782e-06,
"loss": 0.089,
"step": 420
},
{
"epoch": 0.34,
"learning_rate": 9.773115773115774e-06,
"loss": 0.0865,
"step": 430
},
{
"epoch": 0.35,
"learning_rate": 9.765345765345767e-06,
"loss": 0.0868,
"step": 440
},
{
"epoch": 0.36,
"learning_rate": 9.757575757575758e-06,
"loss": 0.0844,
"step": 450
},
{
"epoch": 0.37,
"learning_rate": 9.74980574980575e-06,
"loss": 0.082,
"step": 460
},
{
"epoch": 0.38,
"learning_rate": 9.742035742035743e-06,
"loss": 0.0824,
"step": 470
},
{
"epoch": 0.38,
"learning_rate": 9.734265734265735e-06,
"loss": 0.0828,
"step": 480
},
{
"epoch": 0.39,
"learning_rate": 9.726495726495728e-06,
"loss": 0.0779,
"step": 490
},
{
"epoch": 0.4,
"learning_rate": 9.71872571872572e-06,
"loss": 0.0787,
"step": 500
},
{
"epoch": 0.4,
"eval_loss": 0.0709228515625,
"eval_runtime": 462.1459,
"eval_samples_per_second": 3.823,
"eval_steps_per_second": 0.03,
"eval_wer": 48.49569457412595,
"step": 500
},
{
"epoch": 0.41,
"learning_rate": 9.710955710955711e-06,
"loss": 0.077,
"step": 510
},
{
"epoch": 0.42,
"learning_rate": 9.703185703185704e-06,
"loss": 0.0751,
"step": 520
},
{
"epoch": 0.42,
"learning_rate": 9.695415695415696e-06,
"loss": 0.0764,
"step": 530
},
{
"epoch": 0.43,
"learning_rate": 9.687645687645689e-06,
"loss": 0.076,
"step": 540
},
{
"epoch": 0.44,
"learning_rate": 9.679875679875681e-06,
"loss": 0.0724,
"step": 550
},
{
"epoch": 0.45,
"learning_rate": 9.672105672105672e-06,
"loss": 0.0719,
"step": 560
},
{
"epoch": 0.46,
"learning_rate": 9.664335664335666e-06,
"loss": 0.0707,
"step": 570
},
{
"epoch": 0.46,
"learning_rate": 9.656565656565657e-06,
"loss": 0.0711,
"step": 580
},
{
"epoch": 0.47,
"learning_rate": 9.64879564879565e-06,
"loss": 0.0672,
"step": 590
},
{
"epoch": 0.48,
"learning_rate": 9.641025641025642e-06,
"loss": 0.0705,
"step": 600
},
{
"epoch": 0.48,
"eval_loss": 0.06500244140625,
"eval_runtime": 298.2114,
"eval_samples_per_second": 5.925,
"eval_steps_per_second": 0.047,
"eval_wer": 49.71470069509285,
"step": 600
},
{
"epoch": 0.49,
"learning_rate": 9.633255633255633e-06,
"loss": 0.0696,
"step": 610
},
{
"epoch": 0.5,
"learning_rate": 9.625485625485627e-06,
"loss": 0.066,
"step": 620
},
{
"epoch": 0.5,
"learning_rate": 9.617715617715618e-06,
"loss": 0.0657,
"step": 630
},
{
"epoch": 0.51,
"learning_rate": 9.60994560994561e-06,
"loss": 0.0646,
"step": 640
},
{
"epoch": 0.52,
"learning_rate": 9.602175602175603e-06,
"loss": 0.0657,
"step": 650
},
{
"epoch": 0.53,
"learning_rate": 9.594405594405596e-06,
"loss": 0.0662,
"step": 660
},
{
"epoch": 0.54,
"learning_rate": 9.586635586635588e-06,
"loss": 0.064,
"step": 670
},
{
"epoch": 0.54,
"learning_rate": 9.578865578865579e-06,
"loss": 0.0663,
"step": 680
},
{
"epoch": 0.55,
"learning_rate": 9.571095571095571e-06,
"loss": 0.0626,
"step": 690
},
{
"epoch": 0.56,
"learning_rate": 9.563325563325564e-06,
"loss": 0.0652,
"step": 700
},
{
"epoch": 0.56,
"eval_loss": 0.05706787109375,
"eval_runtime": 274.4365,
"eval_samples_per_second": 6.439,
"eval_steps_per_second": 0.051,
"eval_wer": 46.70090258325553,
"step": 700
},
{
"epoch": 0.57,
"learning_rate": 9.555555555555556e-06,
"loss": 0.0626,
"step": 710
},
{
"epoch": 0.58,
"learning_rate": 9.547785547785549e-06,
"loss": 0.063,
"step": 720
},
{
"epoch": 0.58,
"learning_rate": 9.54001554001554e-06,
"loss": 0.0631,
"step": 730
},
{
"epoch": 0.59,
"learning_rate": 9.532245532245532e-06,
"loss": 0.0656,
"step": 740
},
{
"epoch": 0.6,
"learning_rate": 9.524475524475525e-06,
"loss": 0.0639,
"step": 750
},
{
"epoch": 0.61,
"learning_rate": 9.516705516705517e-06,
"loss": 0.0605,
"step": 760
},
{
"epoch": 0.62,
"learning_rate": 9.50893550893551e-06,
"loss": 0.0636,
"step": 770
},
{
"epoch": 0.62,
"learning_rate": 9.501165501165502e-06,
"loss": 0.0627,
"step": 780
},
{
"epoch": 0.63,
"learning_rate": 9.493395493395493e-06,
"loss": 0.0608,
"step": 790
},
{
"epoch": 0.64,
"learning_rate": 9.485625485625486e-06,
"loss": 0.0609,
"step": 800
},
{
"epoch": 0.64,
"eval_loss": 0.054962158203125,
"eval_runtime": 249.9581,
"eval_samples_per_second": 7.069,
"eval_steps_per_second": 0.056,
"eval_wer": 46.04730781201369,
"step": 800
},
{
"epoch": 0.65,
"learning_rate": 9.477855477855478e-06,
"loss": 0.0605,
"step": 810
},
{
"epoch": 0.66,
"learning_rate": 9.470085470085471e-06,
"loss": 0.0607,
"step": 820
},
{
"epoch": 0.66,
"learning_rate": 9.462315462315463e-06,
"loss": 0.06,
"step": 830
},
{
"epoch": 0.67,
"learning_rate": 9.454545454545456e-06,
"loss": 0.0613,
"step": 840
},
{
"epoch": 0.68,
"learning_rate": 9.446775446775448e-06,
"loss": 0.0613,
"step": 850
},
{
"epoch": 0.69,
"learning_rate": 9.43900543900544e-06,
"loss": 0.0605,
"step": 860
},
{
"epoch": 0.7,
"learning_rate": 9.431235431235432e-06,
"loss": 0.0608,
"step": 870
},
{
"epoch": 0.7,
"learning_rate": 9.423465423465424e-06,
"loss": 0.0603,
"step": 880
},
{
"epoch": 0.71,
"learning_rate": 9.415695415695417e-06,
"loss": 0.0603,
"step": 890
},
{
"epoch": 0.72,
"learning_rate": 9.40792540792541e-06,
"loss": 0.0596,
"step": 900
},
{
"epoch": 0.72,
"eval_loss": 0.05279541015625,
"eval_runtime": 251.267,
"eval_samples_per_second": 7.032,
"eval_steps_per_second": 0.056,
"eval_wer": 45.71013590621434,
"step": 900
},
{
"epoch": 0.73,
"learning_rate": 9.4001554001554e-06,
"loss": 0.059,
"step": 910
},
{
"epoch": 0.74,
"learning_rate": 9.392385392385393e-06,
"loss": 0.0576,
"step": 920
},
{
"epoch": 0.74,
"learning_rate": 9.384615384615385e-06,
"loss": 0.0569,
"step": 930
},
{
"epoch": 0.75,
"learning_rate": 9.376845376845378e-06,
"loss": 0.0592,
"step": 940
},
{
"epoch": 0.76,
"learning_rate": 9.36907536907537e-06,
"loss": 0.0577,
"step": 950
},
{
"epoch": 0.77,
"learning_rate": 9.361305361305361e-06,
"loss": 0.0588,
"step": 960
},
{
"epoch": 0.78,
"learning_rate": 9.353535353535354e-06,
"loss": 0.0575,
"step": 970
},
{
"epoch": 0.78,
"learning_rate": 9.345765345765346e-06,
"loss": 0.0555,
"step": 980
},
{
"epoch": 0.79,
"learning_rate": 9.337995337995339e-06,
"loss": 0.0562,
"step": 990
},
{
"epoch": 0.8,
"learning_rate": 9.330225330225331e-06,
"loss": 0.0571,
"step": 1000
},
{
"epoch": 0.8,
"eval_loss": 0.0517578125,
"eval_runtime": 253.3077,
"eval_samples_per_second": 6.976,
"eval_steps_per_second": 0.055,
"eval_wer": 45.57526714389459,
"step": 1000
},
{
"epoch": 0.81,
"learning_rate": 9.322455322455322e-06,
"loss": 0.058,
"step": 1010
},
{
"epoch": 0.82,
"learning_rate": 9.314685314685316e-06,
"loss": 0.057,
"step": 1020
},
{
"epoch": 0.82,
"learning_rate": 9.306915306915307e-06,
"loss": 0.0559,
"step": 1030
},
{
"epoch": 0.83,
"learning_rate": 9.2991452991453e-06,
"loss": 0.0574,
"step": 1040
},
{
"epoch": 0.84,
"learning_rate": 9.291375291375292e-06,
"loss": 0.0563,
"step": 1050
},
{
"epoch": 0.85,
"learning_rate": 9.283605283605285e-06,
"loss": 0.0581,
"step": 1060
},
{
"epoch": 0.86,
"learning_rate": 9.275835275835277e-06,
"loss": 0.056,
"step": 1070
},
{
"epoch": 0.86,
"learning_rate": 9.268065268065268e-06,
"loss": 0.0588,
"step": 1080
},
{
"epoch": 0.87,
"learning_rate": 9.26029526029526e-06,
"loss": 0.0556,
"step": 1090
},
{
"epoch": 0.88,
"learning_rate": 9.252525252525253e-06,
"loss": 0.0552,
"step": 1100
},
{
"epoch": 0.88,
"eval_loss": 0.05047607421875,
"eval_runtime": 254.3246,
"eval_samples_per_second": 6.948,
"eval_steps_per_second": 0.055,
"eval_wer": 44.35626102292769,
"step": 1100
},
{
"epoch": 0.89,
"learning_rate": 9.244755244755246e-06,
"loss": 0.057,
"step": 1110
},
{
"epoch": 0.9,
"learning_rate": 9.236985236985238e-06,
"loss": 0.0555,
"step": 1120
},
{
"epoch": 0.9,
"learning_rate": 9.22921522921523e-06,
"loss": 0.0551,
"step": 1130
},
{
"epoch": 0.91,
"learning_rate": 9.221445221445222e-06,
"loss": 0.0555,
"step": 1140
},
{
"epoch": 0.92,
"learning_rate": 9.213675213675214e-06,
"loss": 0.0541,
"step": 1150
},
{
"epoch": 0.93,
"learning_rate": 9.205905205905207e-06,
"loss": 0.0555,
"step": 1160
},
{
"epoch": 0.94,
"learning_rate": 9.1981351981352e-06,
"loss": 0.0538,
"step": 1170
},
{
"epoch": 0.94,
"learning_rate": 9.190365190365192e-06,
"loss": 0.0545,
"step": 1180
},
{
"epoch": 0.95,
"learning_rate": 9.182595182595183e-06,
"loss": 0.0536,
"step": 1190
},
{
"epoch": 0.96,
"learning_rate": 9.174825174825177e-06,
"loss": 0.0523,
"step": 1200
},
{
"epoch": 0.96,
"eval_loss": 0.0498046875,
"eval_runtime": 259.5919,
"eval_samples_per_second": 6.807,
"eval_steps_per_second": 0.054,
"eval_wer": 44.16951965971574,
"step": 1200
},
{
"epoch": 0.97,
"learning_rate": 9.167055167055168e-06,
"loss": 0.055,
"step": 1210
},
{
"epoch": 0.98,
"learning_rate": 9.15928515928516e-06,
"loss": 0.0536,
"step": 1220
},
{
"epoch": 0.98,
"learning_rate": 9.151515151515153e-06,
"loss": 0.0543,
"step": 1230
},
{
"epoch": 0.99,
"learning_rate": 9.143745143745144e-06,
"loss": 0.053,
"step": 1240
},
{
"epoch": 1.0,
"learning_rate": 9.135975135975138e-06,
"loss": 0.0549,
"step": 1250
},
{
"epoch": 1.01,
"learning_rate": 9.128205128205129e-06,
"loss": 0.0502,
"step": 1260
},
{
"epoch": 1.02,
"learning_rate": 9.120435120435121e-06,
"loss": 0.0491,
"step": 1270
},
{
"epoch": 1.02,
"learning_rate": 9.112665112665114e-06,
"loss": 0.0466,
"step": 1280
},
{
"epoch": 1.03,
"learning_rate": 9.104895104895104e-06,
"loss": 0.0479,
"step": 1290
},
{
"epoch": 1.04,
"learning_rate": 9.097125097125099e-06,
"loss": 0.048,
"step": 1300
},
{
"epoch": 1.04,
"eval_loss": 0.04876708984375,
"eval_runtime": 298.9142,
"eval_samples_per_second": 5.911,
"eval_steps_per_second": 0.047,
"eval_wer": 43.83234775391638,
"step": 1300
},
{
"epoch": 1.05,
"learning_rate": 9.08935508935509e-06,
"loss": 0.0478,
"step": 1310
},
{
"epoch": 1.06,
"learning_rate": 9.081585081585082e-06,
"loss": 0.0475,
"step": 1320
},
{
"epoch": 1.06,
"learning_rate": 9.073815073815075e-06,
"loss": 0.0491,
"step": 1330
},
{
"epoch": 1.07,
"learning_rate": 9.066045066045067e-06,
"loss": 0.0489,
"step": 1340
},
{
"epoch": 1.08,
"learning_rate": 9.05827505827506e-06,
"loss": 0.048,
"step": 1350
},
{
"epoch": 1.09,
"learning_rate": 9.050505050505052e-06,
"loss": 0.0483,
"step": 1360
},
{
"epoch": 1.1,
"learning_rate": 9.042735042735043e-06,
"loss": 0.0478,
"step": 1370
},
{
"epoch": 1.1,
"learning_rate": 9.034965034965036e-06,
"loss": 0.0473,
"step": 1380
},
{
"epoch": 1.11,
"learning_rate": 9.027195027195028e-06,
"loss": 0.0473,
"step": 1390
},
{
"epoch": 1.12,
"learning_rate": 9.01942501942502e-06,
"loss": 0.0467,
"step": 1400
},
{
"epoch": 1.12,
"eval_loss": 0.0482177734375,
"eval_runtime": 293.889,
"eval_samples_per_second": 6.012,
"eval_steps_per_second": 0.048,
"eval_wer": 43.30843448490507,
"step": 1400
},
{
"epoch": 1.13,
"learning_rate": 9.011655011655013e-06,
"loss": 0.0466,
"step": 1410
},
{
"epoch": 1.14,
"learning_rate": 9.003885003885004e-06,
"loss": 0.0476,
"step": 1420
},
{
"epoch": 1.14,
"learning_rate": 8.996114996114998e-06,
"loss": 0.0486,
"step": 1430
},
{
"epoch": 1.15,
"learning_rate": 8.988344988344989e-06,
"loss": 0.0472,
"step": 1440
},
{
"epoch": 1.16,
"learning_rate": 8.980574980574982e-06,
"loss": 0.046,
"step": 1450
},
{
"epoch": 1.17,
"learning_rate": 8.972804972804974e-06,
"loss": 0.049,
"step": 1460
},
{
"epoch": 1.18,
"learning_rate": 8.965034965034965e-06,
"loss": 0.047,
"step": 1470
},
{
"epoch": 1.18,
"learning_rate": 8.957264957264959e-06,
"loss": 0.0479,
"step": 1480
},
{
"epoch": 1.19,
"learning_rate": 8.94949494949495e-06,
"loss": 0.0475,
"step": 1490
},
{
"epoch": 1.2,
"learning_rate": 8.941724941724942e-06,
"loss": 0.048,
"step": 1500
},
{
"epoch": 1.2,
"eval_loss": 0.04736328125,
"eval_runtime": 285.6961,
"eval_samples_per_second": 6.185,
"eval_steps_per_second": 0.049,
"eval_wer": 43.215063803299095,
"step": 1500
},
{
"epoch": 1.21,
"learning_rate": 8.933954933954935e-06,
"loss": 0.0468,
"step": 1510
},
{
"epoch": 1.22,
"learning_rate": 8.926184926184926e-06,
"loss": 0.0489,
"step": 1520
},
{
"epoch": 1.22,
"learning_rate": 8.91841491841492e-06,
"loss": 0.0461,
"step": 1530
},
{
"epoch": 1.23,
"learning_rate": 8.910644910644911e-06,
"loss": 0.0485,
"step": 1540
},
{
"epoch": 1.24,
"learning_rate": 8.902874902874903e-06,
"loss": 0.0471,
"step": 1550
},
{
"epoch": 1.25,
"learning_rate": 8.895104895104896e-06,
"loss": 0.0484,
"step": 1560
},
{
"epoch": 1.25,
"learning_rate": 8.887334887334888e-06,
"loss": 0.0472,
"step": 1570
},
{
"epoch": 1.26,
"learning_rate": 8.879564879564881e-06,
"loss": 0.0475,
"step": 1580
},
{
"epoch": 1.27,
"learning_rate": 8.871794871794872e-06,
"loss": 0.0463,
"step": 1590
},
{
"epoch": 1.28,
"learning_rate": 8.864024864024864e-06,
"loss": 0.0469,
"step": 1600
},
{
"epoch": 1.28,
"eval_loss": 0.046783447265625,
"eval_runtime": 283.6885,
"eval_samples_per_second": 6.229,
"eval_steps_per_second": 0.049,
"eval_wer": 43.266936404191306,
"step": 1600
},
{
"epoch": 1.29,
"learning_rate": 8.856254856254857e-06,
"loss": 0.0458,
"step": 1610
},
{
"epoch": 1.29,
"learning_rate": 8.84848484848485e-06,
"loss": 0.0467,
"step": 1620
},
{
"epoch": 1.3,
"learning_rate": 8.840714840714842e-06,
"loss": 0.0469,
"step": 1630
},
{
"epoch": 1.31,
"learning_rate": 8.832944832944835e-06,
"loss": 0.0473,
"step": 1640
},
{
"epoch": 1.32,
"learning_rate": 8.825174825174825e-06,
"loss": 0.0457,
"step": 1650
},
{
"epoch": 1.33,
"learning_rate": 8.817404817404818e-06,
"loss": 0.0461,
"step": 1660
},
{
"epoch": 1.33,
"learning_rate": 8.80963480963481e-06,
"loss": 0.0472,
"step": 1670
},
{
"epoch": 1.34,
"learning_rate": 8.801864801864803e-06,
"loss": 0.0463,
"step": 1680
},
{
"epoch": 1.35,
"learning_rate": 8.794094794094795e-06,
"loss": 0.047,
"step": 1690
},
{
"epoch": 1.36,
"learning_rate": 8.786324786324786e-06,
"loss": 0.0481,
"step": 1700
},
{
"epoch": 1.36,
"eval_loss": 0.046112060546875,
"eval_runtime": 276.946,
"eval_samples_per_second": 6.38,
"eval_steps_per_second": 0.051,
"eval_wer": 43.266936404191306,
"step": 1700
},
{
"epoch": 1.37,
"learning_rate": 8.77855477855478e-06,
"loss": 0.0479,
"step": 1710
},
{
"epoch": 1.37,
"learning_rate": 8.770784770784771e-06,
"loss": 0.048,
"step": 1720
},
{
"epoch": 1.38,
"learning_rate": 8.763014763014764e-06,
"loss": 0.0461,
"step": 1730
},
{
"epoch": 1.39,
"learning_rate": 8.755244755244756e-06,
"loss": 0.0451,
"step": 1740
},
{
"epoch": 1.4,
"learning_rate": 8.747474747474747e-06,
"loss": 0.0471,
"step": 1750
},
{
"epoch": 1.41,
"learning_rate": 8.739704739704741e-06,
"loss": 0.046,
"step": 1760
},
{
"epoch": 1.41,
"learning_rate": 8.731934731934732e-06,
"loss": 0.0466,
"step": 1770
},
{
"epoch": 1.42,
"learning_rate": 8.724164724164725e-06,
"loss": 0.0474,
"step": 1780
},
{
"epoch": 1.43,
"learning_rate": 8.716394716394717e-06,
"loss": 0.0458,
"step": 1790
},
{
"epoch": 1.44,
"learning_rate": 8.708624708624708e-06,
"loss": 0.0463,
"step": 1800
},
{
"epoch": 1.44,
"eval_loss": 0.045867919921875,
"eval_runtime": 262.0176,
"eval_samples_per_second": 6.744,
"eval_steps_per_second": 0.053,
"eval_wer": 43.00238613964104,
"step": 1800
},
{
"epoch": 1.45,
"learning_rate": 8.700854700854702e-06,
"loss": 0.0459,
"step": 1810
},
{
"epoch": 1.45,
"learning_rate": 8.693084693084693e-06,
"loss": 0.0469,
"step": 1820
},
{
"epoch": 1.46,
"learning_rate": 8.685314685314686e-06,
"loss": 0.0468,
"step": 1830
},
{
"epoch": 1.47,
"learning_rate": 8.677544677544678e-06,
"loss": 0.0475,
"step": 1840
},
{
"epoch": 1.48,
"learning_rate": 8.66977466977467e-06,
"loss": 0.0473,
"step": 1850
},
{
"epoch": 1.49,
"learning_rate": 8.662004662004663e-06,
"loss": 0.0462,
"step": 1860
},
{
"epoch": 1.49,
"learning_rate": 8.654234654234654e-06,
"loss": 0.0469,
"step": 1870
},
{
"epoch": 1.5,
"learning_rate": 8.646464646464647e-06,
"loss": 0.045,
"step": 1880
},
{
"epoch": 1.51,
"learning_rate": 8.63869463869464e-06,
"loss": 0.0472,
"step": 1890
},
{
"epoch": 1.52,
"learning_rate": 8.630924630924632e-06,
"loss": 0.0462,
"step": 1900
},
{
"epoch": 1.52,
"eval_loss": 0.045501708984375,
"eval_runtime": 266.2018,
"eval_samples_per_second": 6.638,
"eval_steps_per_second": 0.053,
"eval_wer": 43.23581284365598,
"step": 1900
},
{
"epoch": 1.53,
"learning_rate": 8.623154623154624e-06,
"loss": 0.0458,
"step": 1910
},
{
"epoch": 1.53,
"learning_rate": 8.615384615384617e-06,
"loss": 0.0457,
"step": 1920
},
{
"epoch": 1.54,
"learning_rate": 8.607614607614608e-06,
"loss": 0.046,
"step": 1930
},
{
"epoch": 1.55,
"learning_rate": 8.5998445998446e-06,
"loss": 0.0477,
"step": 1940
},
{
"epoch": 1.56,
"learning_rate": 8.592074592074593e-06,
"loss": 0.0469,
"step": 1950
},
{
"epoch": 1.57,
"learning_rate": 8.584304584304585e-06,
"loss": 0.0475,
"step": 1960
},
{
"epoch": 1.57,
"learning_rate": 8.576534576534578e-06,
"loss": 0.0461,
"step": 1970
},
{
"epoch": 1.58,
"learning_rate": 8.568764568764569e-06,
"loss": 0.0459,
"step": 1980
},
{
"epoch": 1.59,
"learning_rate": 8.560994560994563e-06,
"loss": 0.0463,
"step": 1990
},
{
"epoch": 1.6,
"learning_rate": 8.553224553224554e-06,
"loss": 0.0467,
"step": 2000
},
{
"epoch": 1.6,
"eval_loss": 0.04486083984375,
"eval_runtime": 263.6022,
"eval_samples_per_second": 6.703,
"eval_steps_per_second": 0.053,
"eval_wer": 43.147629422139225,
"step": 2000
},
{
"epoch": 1.61,
"learning_rate": 8.545454545454546e-06,
"loss": 0.0452,
"step": 2010
},
{
"epoch": 1.61,
"learning_rate": 8.537684537684539e-06,
"loss": 0.0459,
"step": 2020
},
{
"epoch": 1.62,
"learning_rate": 8.52991452991453e-06,
"loss": 0.0461,
"step": 2030
},
{
"epoch": 1.63,
"learning_rate": 8.522144522144524e-06,
"loss": 0.0463,
"step": 2040
},
{
"epoch": 1.64,
"learning_rate": 8.514374514374515e-06,
"loss": 0.0465,
"step": 2050
},
{
"epoch": 1.65,
"learning_rate": 8.506604506604507e-06,
"loss": 0.0459,
"step": 2060
},
{
"epoch": 1.65,
"learning_rate": 8.4988344988345e-06,
"loss": 0.0452,
"step": 2070
},
{
"epoch": 1.66,
"learning_rate": 8.49106449106449e-06,
"loss": 0.0444,
"step": 2080
},
{
"epoch": 1.67,
"learning_rate": 8.483294483294485e-06,
"loss": 0.0462,
"step": 2090
},
{
"epoch": 1.68,
"learning_rate": 8.475524475524476e-06,
"loss": 0.0448,
"step": 2100
},
{
"epoch": 1.68,
"eval_loss": 0.044464111328125,
"eval_runtime": 218.3005,
"eval_samples_per_second": 8.094,
"eval_steps_per_second": 0.064,
"eval_wer": 42.08942836393817,
"step": 2100
},
{
"epoch": 1.69,
"learning_rate": 8.467754467754468e-06,
"loss": 0.0459,
"step": 2110
},
{
"epoch": 1.69,
"learning_rate": 8.45998445998446e-06,
"loss": 0.0463,
"step": 2120
},
{
"epoch": 1.7,
"learning_rate": 8.452214452214453e-06,
"loss": 0.0461,
"step": 2130
},
{
"epoch": 1.71,
"learning_rate": 8.444444444444446e-06,
"loss": 0.0457,
"step": 2140
},
{
"epoch": 1.72,
"learning_rate": 8.436674436674436e-06,
"loss": 0.0462,
"step": 2150
},
{
"epoch": 1.73,
"learning_rate": 8.428904428904429e-06,
"loss": 0.044,
"step": 2160
},
{
"epoch": 1.73,
"learning_rate": 8.421134421134422e-06,
"loss": 0.0453,
"step": 2170
},
{
"epoch": 1.74,
"learning_rate": 8.413364413364414e-06,
"loss": 0.0469,
"step": 2180
},
{
"epoch": 1.75,
"learning_rate": 8.405594405594407e-06,
"loss": 0.0451,
"step": 2190
},
{
"epoch": 1.76,
"learning_rate": 8.397824397824399e-06,
"loss": 0.0444,
"step": 2200
},
{
"epoch": 1.76,
"eval_loss": 0.0439453125,
"eval_runtime": 243.1614,
"eval_samples_per_second": 7.267,
"eval_steps_per_second": 0.058,
"eval_wer": 42.19317356572259,
"step": 2200
},
{
"epoch": 1.77,
"learning_rate": 8.39005439005439e-06,
"loss": 0.0446,
"step": 2210
},
{
"epoch": 1.77,
"learning_rate": 8.382284382284382e-06,
"loss": 0.0442,
"step": 2220
},
{
"epoch": 1.78,
"learning_rate": 8.374514374514375e-06,
"loss": 0.0441,
"step": 2230
},
{
"epoch": 1.79,
"learning_rate": 8.366744366744368e-06,
"loss": 0.0446,
"step": 2240
},
{
"epoch": 1.8,
"learning_rate": 8.35897435897436e-06,
"loss": 0.046,
"step": 2250
},
{
"epoch": 1.81,
"learning_rate": 8.351204351204351e-06,
"loss": 0.0465,
"step": 2260
},
{
"epoch": 1.81,
"learning_rate": 8.343434343434345e-06,
"loss": 0.0441,
"step": 2270
},
{
"epoch": 1.82,
"learning_rate": 8.335664335664336e-06,
"loss": 0.045,
"step": 2280
},
{
"epoch": 1.83,
"learning_rate": 8.327894327894329e-06,
"loss": 0.0459,
"step": 2290
},
{
"epoch": 1.84,
"learning_rate": 8.320124320124321e-06,
"loss": 0.0431,
"step": 2300
},
{
"epoch": 1.84,
"eval_loss": 0.043853759765625,
"eval_runtime": 237.3655,
"eval_samples_per_second": 7.444,
"eval_steps_per_second": 0.059,
"eval_wer": 42.16723726527648,
"step": 2300
},
{
"epoch": 1.85,
"learning_rate": 8.312354312354312e-06,
"loss": 0.0425,
"step": 2310
},
{
"epoch": 1.85,
"learning_rate": 8.304584304584306e-06,
"loss": 0.0442,
"step": 2320
},
{
"epoch": 1.86,
"learning_rate": 8.296814296814297e-06,
"loss": 0.0427,
"step": 2330
},
{
"epoch": 1.87,
"learning_rate": 8.28904428904429e-06,
"loss": 0.0471,
"step": 2340
},
{
"epoch": 1.88,
"learning_rate": 8.281274281274282e-06,
"loss": 0.044,
"step": 2350
},
{
"epoch": 1.89,
"learning_rate": 8.273504273504273e-06,
"loss": 0.0461,
"step": 2360
},
{
"epoch": 1.89,
"learning_rate": 8.265734265734267e-06,
"loss": 0.0445,
"step": 2370
},
{
"epoch": 1.9,
"learning_rate": 8.257964257964258e-06,
"loss": 0.0457,
"step": 2380
},
{
"epoch": 1.91,
"learning_rate": 8.25019425019425e-06,
"loss": 0.0441,
"step": 2390
},
{
"epoch": 1.92,
"learning_rate": 8.242424242424243e-06,
"loss": 0.0438,
"step": 2400
},
{
"epoch": 1.92,
"eval_loss": 0.0433349609375,
"eval_runtime": 253.2781,
"eval_samples_per_second": 6.977,
"eval_steps_per_second": 0.055,
"eval_wer": 42.592592592592595,
"step": 2400
},
{
"epoch": 1.93,
"learning_rate": 8.234654234654235e-06,
"loss": 0.044,
"step": 2410
},
{
"epoch": 1.93,
"learning_rate": 8.226884226884228e-06,
"loss": 0.0445,
"step": 2420
},
{
"epoch": 1.94,
"learning_rate": 8.219114219114219e-06,
"loss": 0.0441,
"step": 2430
},
{
"epoch": 1.95,
"learning_rate": 8.211344211344211e-06,
"loss": 0.0447,
"step": 2440
},
{
"epoch": 1.96,
"learning_rate": 8.203574203574204e-06,
"loss": 0.0443,
"step": 2450
},
{
"epoch": 1.97,
"learning_rate": 8.195804195804196e-06,
"loss": 0.0457,
"step": 2460
},
{
"epoch": 1.97,
"learning_rate": 8.188034188034189e-06,
"loss": 0.0449,
"step": 2470
},
{
"epoch": 1.98,
"learning_rate": 8.180264180264181e-06,
"loss": 0.0434,
"step": 2480
},
{
"epoch": 1.99,
"learning_rate": 8.172494172494172e-06,
"loss": 0.0438,
"step": 2490
},
{
"epoch": 2.0,
"learning_rate": 8.164724164724165e-06,
"loss": 0.0447,
"step": 2500
},
{
"epoch": 2.0,
"eval_loss": 0.043243408203125,
"eval_runtime": 238.2415,
"eval_samples_per_second": 7.417,
"eval_steps_per_second": 0.059,
"eval_wer": 42.82083203651831,
"step": 2500
},
{
"epoch": 2.01,
"learning_rate": 8.156954156954157e-06,
"loss": 0.04,
"step": 2510
},
{
"epoch": 2.01,
"learning_rate": 8.14918414918415e-06,
"loss": 0.0392,
"step": 2520
},
{
"epoch": 2.02,
"learning_rate": 8.141414141414142e-06,
"loss": 0.0372,
"step": 2530
},
{
"epoch": 2.03,
"learning_rate": 8.133644133644133e-06,
"loss": 0.0395,
"step": 2540
},
{
"epoch": 2.04,
"learning_rate": 8.125874125874127e-06,
"loss": 0.0393,
"step": 2550
},
{
"epoch": 2.05,
"learning_rate": 8.118104118104118e-06,
"loss": 0.0376,
"step": 2560
},
{
"epoch": 2.05,
"learning_rate": 8.11033411033411e-06,
"loss": 0.0375,
"step": 2570
},
{
"epoch": 2.06,
"learning_rate": 8.102564102564103e-06,
"loss": 0.0385,
"step": 2580
},
{
"epoch": 2.07,
"learning_rate": 8.094794094794096e-06,
"loss": 0.0389,
"step": 2590
},
{
"epoch": 2.08,
"learning_rate": 8.087024087024088e-06,
"loss": 0.0372,
"step": 2600
},
{
"epoch": 2.08,
"eval_loss": 0.04345703125,
"eval_runtime": 441.8133,
"eval_samples_per_second": 3.999,
"eval_steps_per_second": 0.032,
"eval_wer": 41.91824878099388,
"step": 2600
},
{
"epoch": 2.09,
"learning_rate": 8.07925407925408e-06,
"loss": 0.0387,
"step": 2610
},
{
"epoch": 2.09,
"learning_rate": 8.071484071484072e-06,
"loss": 0.0373,
"step": 2620
},
{
"epoch": 2.1,
"learning_rate": 8.063714063714064e-06,
"loss": 0.0388,
"step": 2630
},
{
"epoch": 2.11,
"learning_rate": 8.055944055944057e-06,
"loss": 0.038,
"step": 2640
},
{
"epoch": 2.12,
"learning_rate": 8.04817404817405e-06,
"loss": 0.0378,
"step": 2650
},
{
"epoch": 2.13,
"learning_rate": 8.04040404040404e-06,
"loss": 0.0402,
"step": 2660
},
{
"epoch": 2.13,
"learning_rate": 8.032634032634033e-06,
"loss": 0.0379,
"step": 2670
},
{
"epoch": 2.14,
"learning_rate": 8.024864024864025e-06,
"loss": 0.0372,
"step": 2680
},
{
"epoch": 2.15,
"learning_rate": 8.017094017094018e-06,
"loss": 0.0385,
"step": 2690
},
{
"epoch": 2.16,
"learning_rate": 8.00932400932401e-06,
"loss": 0.0395,
"step": 2700
},
{
"epoch": 2.16,
"eval_loss": 0.043304443359375,
"eval_runtime": 501.1941,
"eval_samples_per_second": 3.526,
"eval_steps_per_second": 0.028,
"eval_wer": 41.78856727876335,
"step": 2700
},
{
"epoch": 2.17,
"learning_rate": 8.001554001554003e-06,
"loss": 0.0395,
"step": 2710
},
{
"epoch": 2.17,
"learning_rate": 7.993783993783994e-06,
"loss": 0.0381,
"step": 2720
},
{
"epoch": 2.18,
"learning_rate": 7.987567987567988e-06,
"loss": 0.2206,
"step": 2730
},
{
"epoch": 2.19,
"learning_rate": 7.97979797979798e-06,
"loss": 0.0797,
"step": 2740
},
{
"epoch": 2.2,
"learning_rate": 7.972027972027973e-06,
"loss": 0.04,
"step": 2750
},
{
"epoch": 2.21,
"learning_rate": 7.964257964257965e-06,
"loss": 0.0398,
"step": 2760
},
{
"epoch": 2.21,
"learning_rate": 7.956487956487958e-06,
"loss": 0.039,
"step": 2770
},
{
"epoch": 2.22,
"learning_rate": 7.948717948717949e-06,
"loss": 0.0404,
"step": 2780
},
{
"epoch": 2.23,
"learning_rate": 7.940947940947941e-06,
"loss": 0.0419,
"step": 2790
},
{
"epoch": 2.24,
"learning_rate": 7.933177933177934e-06,
"loss": 0.0407,
"step": 2800
},
{
"epoch": 2.24,
"eval_loss": 0.04669189453125,
"eval_runtime": 553.3408,
"eval_samples_per_second": 3.193,
"eval_steps_per_second": 0.025,
"eval_wer": 42.80527025625065,
"step": 2800
},
{
"epoch": 2.25,
"learning_rate": 7.925407925407926e-06,
"loss": 0.0435,
"step": 2810
},
{
"epoch": 2.25,
"learning_rate": 7.917637917637919e-06,
"loss": 0.042,
"step": 2820
},
{
"epoch": 2.26,
"learning_rate": 7.909867909867911e-06,
"loss": 0.0402,
"step": 2830
},
{
"epoch": 2.27,
"learning_rate": 7.902097902097902e-06,
"loss": 0.0418,
"step": 2840
},
{
"epoch": 2.28,
"learning_rate": 7.894327894327895e-06,
"loss": 0.0392,
"step": 2850
},
{
"epoch": 2.29,
"learning_rate": 7.886557886557887e-06,
"loss": 0.0396,
"step": 2860
},
{
"epoch": 2.29,
"learning_rate": 7.87878787878788e-06,
"loss": 0.039,
"step": 2870
},
{
"epoch": 2.3,
"learning_rate": 7.871017871017872e-06,
"loss": 0.0391,
"step": 2880
},
{
"epoch": 2.31,
"learning_rate": 7.863247863247863e-06,
"loss": 0.0385,
"step": 2890
},
{
"epoch": 2.32,
"learning_rate": 7.855477855477857e-06,
"loss": 0.038,
"step": 2900
},
{
"epoch": 2.32,
"eval_loss": 0.04351806640625,
"eval_runtime": 553.1279,
"eval_samples_per_second": 3.195,
"eval_steps_per_second": 0.025,
"eval_wer": 41.95974686170765,
"step": 2900
},
{
"epoch": 2.33,
"learning_rate": 7.847707847707848e-06,
"loss": 0.0393,
"step": 2910
},
{
"epoch": 2.33,
"learning_rate": 7.83993783993784e-06,
"loss": 0.0395,
"step": 2920
},
{
"epoch": 2.34,
"learning_rate": 7.832167832167833e-06,
"loss": 0.0396,
"step": 2930
},
{
"epoch": 2.35,
"learning_rate": 7.824397824397824e-06,
"loss": 0.0386,
"step": 2940
},
{
"epoch": 2.36,
"learning_rate": 7.816627816627818e-06,
"loss": 0.04,
"step": 2950
},
{
"epoch": 2.37,
"learning_rate": 7.808857808857809e-06,
"loss": 0.0389,
"step": 2960
},
{
"epoch": 2.37,
"learning_rate": 7.801087801087802e-06,
"loss": 0.0395,
"step": 2970
},
{
"epoch": 2.38,
"learning_rate": 7.793317793317794e-06,
"loss": 0.0396,
"step": 2980
},
{
"epoch": 2.39,
"learning_rate": 7.785547785547785e-06,
"loss": 0.039,
"step": 2990
},
{
"epoch": 2.4,
"learning_rate": 7.77777777777778e-06,
"loss": 0.039,
"step": 3000
},
{
"epoch": 2.4,
"eval_loss": 0.04388427734375,
"eval_runtime": 560.9888,
"eval_samples_per_second": 3.15,
"eval_steps_per_second": 0.025,
"eval_wer": 42.151675485008816,
"step": 3000
},
{
"epoch": 2.41,
"learning_rate": 7.77000777000777e-06,
"loss": 0.0405,
"step": 3010
},
{
"epoch": 2.41,
"learning_rate": 7.762237762237763e-06,
"loss": 0.04,
"step": 3020
},
{
"epoch": 2.42,
"learning_rate": 7.754467754467755e-06,
"loss": 0.0387,
"step": 3030
},
{
"epoch": 2.43,
"learning_rate": 7.746697746697748e-06,
"loss": 0.0393,
"step": 3040
},
{
"epoch": 2.44,
"learning_rate": 7.73892773892774e-06,
"loss": 0.0409,
"step": 3050
},
{
"epoch": 2.45,
"learning_rate": 7.731157731157731e-06,
"loss": 0.0399,
"step": 3060
},
{
"epoch": 2.45,
"learning_rate": 7.723387723387723e-06,
"loss": 0.0397,
"step": 3070
},
{
"epoch": 2.46,
"learning_rate": 7.715617715617716e-06,
"loss": 0.046,
"step": 3080
},
{
"epoch": 2.47,
"learning_rate": 7.707847707847709e-06,
"loss": 0.0422,
"step": 3090
},
{
"epoch": 2.48,
"learning_rate": 7.700077700077701e-06,
"loss": 0.0403,
"step": 3100
},
{
"epoch": 2.48,
"eval_loss": 0.04559326171875,
"eval_runtime": 551.8046,
"eval_samples_per_second": 3.202,
"eval_steps_per_second": 0.025,
"eval_wer": 42.74821039526922,
"step": 3100
},
{
"epoch": 2.49,
"learning_rate": 7.692307692307694e-06,
"loss": 0.0439,
"step": 3110
},
{
"epoch": 2.49,
"learning_rate": 7.684537684537684e-06,
"loss": 0.0431,
"step": 3120
},
{
"epoch": 2.5,
"learning_rate": 7.676767676767677e-06,
"loss": 0.0408,
"step": 3130
},
{
"epoch": 2.51,
"learning_rate": 7.66899766899767e-06,
"loss": 0.0392,
"step": 3140
},
{
"epoch": 2.52,
"learning_rate": 7.661227661227662e-06,
"loss": 0.0404,
"step": 3150
},
{
"epoch": 2.53,
"learning_rate": 7.653457653457655e-06,
"loss": 0.0405,
"step": 3160
},
{
"epoch": 2.53,
"learning_rate": 7.645687645687645e-06,
"loss": 0.0406,
"step": 3170
},
{
"epoch": 2.54,
"learning_rate": 7.63791763791764e-06,
"loss": 0.0393,
"step": 3180
},
{
"epoch": 2.55,
"learning_rate": 7.63014763014763e-06,
"loss": 0.0389,
"step": 3190
},
{
"epoch": 2.56,
"learning_rate": 7.622377622377622e-06,
"loss": 0.0394,
"step": 3200
},
{
"epoch": 2.56,
"eval_loss": 0.042999267578125,
"eval_runtime": 561.6586,
"eval_samples_per_second": 3.146,
"eval_steps_per_second": 0.025,
"eval_wer": 41.68482207697894,
"step": 3200
},
{
"epoch": 2.57,
"learning_rate": 7.6146076146076155e-06,
"loss": 0.039,
"step": 3210
},
{
"epoch": 2.57,
"learning_rate": 7.606837606837607e-06,
"loss": 0.041,
"step": 3220
},
{
"epoch": 2.58,
"learning_rate": 7.5990675990676e-06,
"loss": 0.041,
"step": 3230
},
{
"epoch": 2.59,
"learning_rate": 7.591297591297591e-06,
"loss": 0.0425,
"step": 3240
},
{
"epoch": 2.6,
"learning_rate": 7.583527583527585e-06,
"loss": 0.0407,
"step": 3250
},
{
"epoch": 2.61,
"learning_rate": 7.5757575757575764e-06,
"loss": 0.0408,
"step": 3260
},
{
"epoch": 2.61,
"learning_rate": 7.567987567987568e-06,
"loss": 0.0412,
"step": 3270
},
{
"epoch": 2.62,
"learning_rate": 7.560217560217561e-06,
"loss": 0.0408,
"step": 3280
},
{
"epoch": 2.63,
"learning_rate": 7.552447552447552e-06,
"loss": 0.0389,
"step": 3290
},
{
"epoch": 2.64,
"learning_rate": 7.544677544677546e-06,
"loss": 0.0414,
"step": 3300
},
{
"epoch": 2.64,
"eval_loss": 0.043243408203125,
"eval_runtime": 559.314,
"eval_samples_per_second": 3.159,
"eval_steps_per_second": 0.025,
"eval_wer": 42.10499014420583,
"step": 3300
},
{
"epoch": 2.65,
"learning_rate": 7.536907536907537e-06,
"loss": 0.0402,
"step": 3310
},
{
"epoch": 2.65,
"learning_rate": 7.52913752913753e-06,
"loss": 0.0404,
"step": 3320
},
{
"epoch": 2.66,
"learning_rate": 7.521367521367522e-06,
"loss": 0.0394,
"step": 3330
},
{
"epoch": 2.67,
"learning_rate": 7.513597513597513e-06,
"loss": 0.0424,
"step": 3340
},
{
"epoch": 2.68,
"learning_rate": 7.505827505827507e-06,
"loss": 0.0404,
"step": 3350
},
{
"epoch": 2.69,
"learning_rate": 7.498057498057498e-06,
"loss": 0.0388,
"step": 3360
},
{
"epoch": 2.69,
"learning_rate": 7.490287490287491e-06,
"loss": 0.0393,
"step": 3370
},
{
"epoch": 2.7,
"learning_rate": 7.4825174825174825e-06,
"loss": 0.0383,
"step": 3380
},
{
"epoch": 2.71,
"learning_rate": 7.474747474747476e-06,
"loss": 0.0377,
"step": 3390
},
{
"epoch": 2.72,
"learning_rate": 7.466977466977468e-06,
"loss": 0.0379,
"step": 3400
},
{
"epoch": 2.72,
"eval_loss": 0.04315185546875,
"eval_runtime": 561.7746,
"eval_samples_per_second": 3.145,
"eval_steps_per_second": 0.025,
"eval_wer": 42.24504616661479,
"step": 3400
},
{
"epoch": 2.73,
"learning_rate": 7.45920745920746e-06,
"loss": 0.0395,
"step": 3410
},
{
"epoch": 2.73,
"learning_rate": 7.451437451437452e-06,
"loss": 0.0383,
"step": 3420
},
{
"epoch": 2.74,
"learning_rate": 7.4436674436674435e-06,
"loss": 0.038,
"step": 3430
},
{
"epoch": 2.75,
"learning_rate": 7.435897435897437e-06,
"loss": 0.0389,
"step": 3440
},
{
"epoch": 2.76,
"learning_rate": 7.4281274281274285e-06,
"loss": 0.0407,
"step": 3450
},
{
"epoch": 2.77,
"learning_rate": 7.420357420357421e-06,
"loss": 0.0391,
"step": 3460
},
{
"epoch": 2.77,
"learning_rate": 7.412587412587413e-06,
"loss": 0.0392,
"step": 3470
},
{
"epoch": 2.78,
"learning_rate": 7.404817404817406e-06,
"loss": 0.0393,
"step": 3480
},
{
"epoch": 2.79,
"learning_rate": 7.397047397047398e-06,
"loss": 0.039,
"step": 3490
},
{
"epoch": 2.8,
"learning_rate": 7.3892773892773895e-06,
"loss": 0.039,
"step": 3500
},
{
"epoch": 2.8,
"eval_loss": 0.04229736328125,
"eval_runtime": 502.2833,
"eval_samples_per_second": 3.518,
"eval_steps_per_second": 0.028,
"eval_wer": 41.326901130822705,
"step": 3500
},
{
"epoch": 2.81,
"learning_rate": 7.381507381507382e-06,
"loss": 0.0396,
"step": 3510
},
{
"epoch": 2.81,
"learning_rate": 7.373737373737374e-06,
"loss": 0.0399,
"step": 3520
},
{
"epoch": 2.82,
"learning_rate": 7.365967365967367e-06,
"loss": 0.0387,
"step": 3530
},
{
"epoch": 2.83,
"learning_rate": 7.358197358197359e-06,
"loss": 0.0375,
"step": 3540
},
{
"epoch": 2.84,
"learning_rate": 7.350427350427351e-06,
"loss": 0.0407,
"step": 3550
},
{
"epoch": 2.85,
"learning_rate": 7.342657342657343e-06,
"loss": 0.0391,
"step": 3560
},
{
"epoch": 2.85,
"learning_rate": 7.3348873348873355e-06,
"loss": 0.0395,
"step": 3570
},
{
"epoch": 2.86,
"learning_rate": 7.327117327117328e-06,
"loss": 0.0389,
"step": 3580
},
{
"epoch": 2.87,
"learning_rate": 7.31934731934732e-06,
"loss": 0.0391,
"step": 3590
},
{
"epoch": 2.88,
"learning_rate": 7.311577311577312e-06,
"loss": 0.0404,
"step": 3600
},
{
"epoch": 2.88,
"eval_loss": 0.04168701171875,
"eval_runtime": 240.8671,
"eval_samples_per_second": 7.336,
"eval_steps_per_second": 0.058,
"eval_wer": 41.26465400975205,
"step": 3600
},
{
"epoch": 2.89,
"learning_rate": 7.303807303807304e-06,
"loss": 0.0393,
"step": 3610
},
{
"epoch": 2.89,
"learning_rate": 7.296037296037297e-06,
"loss": 0.0408,
"step": 3620
},
{
"epoch": 2.9,
"learning_rate": 7.288267288267289e-06,
"loss": 0.0381,
"step": 3630
},
{
"epoch": 2.91,
"learning_rate": 7.280497280497281e-06,
"loss": 0.0382,
"step": 3640
},
{
"epoch": 2.92,
"learning_rate": 7.272727272727273e-06,
"loss": 0.0392,
"step": 3650
},
{
"epoch": 2.93,
"learning_rate": 7.264957264957266e-06,
"loss": 0.038,
"step": 3660
},
{
"epoch": 2.93,
"learning_rate": 7.257187257187258e-06,
"loss": 0.0391,
"step": 3670
},
{
"epoch": 2.94,
"learning_rate": 7.24941724941725e-06,
"loss": 0.0394,
"step": 3680
},
{
"epoch": 2.95,
"learning_rate": 7.2416472416472424e-06,
"loss": 0.039,
"step": 3690
},
{
"epoch": 2.96,
"learning_rate": 7.233877233877234e-06,
"loss": 0.038,
"step": 3700
},
{
"epoch": 2.96,
"eval_loss": 0.04193115234375,
"eval_runtime": 227.7374,
"eval_samples_per_second": 7.759,
"eval_steps_per_second": 0.061,
"eval_wer": 41.90787426081543,
"step": 3700
},
{
"epoch": 2.97,
"learning_rate": 7.226107226107227e-06,
"loss": 0.0385,
"step": 3710
},
{
"epoch": 2.97,
"learning_rate": 7.218337218337219e-06,
"loss": 0.0388,
"step": 3720
},
{
"epoch": 2.98,
"learning_rate": 7.210567210567211e-06,
"loss": 0.0381,
"step": 3730
},
{
"epoch": 2.99,
"learning_rate": 7.202797202797203e-06,
"loss": 0.0384,
"step": 3740
},
{
"epoch": 3.0,
"learning_rate": 7.195027195027196e-06,
"loss": 0.0395,
"step": 3750
},
{
"epoch": 3.01,
"learning_rate": 7.1872571872571884e-06,
"loss": 0.0343,
"step": 3760
},
{
"epoch": 3.01,
"learning_rate": 7.17948717948718e-06,
"loss": 0.0345,
"step": 3770
},
{
"epoch": 3.02,
"learning_rate": 7.171717171717172e-06,
"loss": 0.0342,
"step": 3780
},
{
"epoch": 3.03,
"learning_rate": 7.163947163947164e-06,
"loss": 0.0335,
"step": 3790
},
{
"epoch": 3.04,
"learning_rate": 7.156177156177157e-06,
"loss": 0.0346,
"step": 3800
},
{
"epoch": 3.04,
"eval_loss": 0.042877197265625,
"eval_runtime": 545.6142,
"eval_samples_per_second": 3.239,
"eval_steps_per_second": 0.026,
"eval_wer": 41.89231248054777,
"step": 3800
},
{
"epoch": 3.05,
"learning_rate": 7.148407148407149e-06,
"loss": 0.0344,
"step": 3810
},
{
"epoch": 3.05,
"learning_rate": 7.140637140637141e-06,
"loss": 0.0341,
"step": 3820
},
{
"epoch": 3.06,
"learning_rate": 7.132867132867134e-06,
"loss": 0.0329,
"step": 3830
},
{
"epoch": 3.07,
"learning_rate": 7.125097125097126e-06,
"loss": 0.0355,
"step": 3840
},
{
"epoch": 3.08,
"learning_rate": 7.117327117327118e-06,
"loss": 0.034,
"step": 3850
},
{
"epoch": 3.09,
"learning_rate": 7.10955710955711e-06,
"loss": 0.0327,
"step": 3860
},
{
"epoch": 3.09,
"learning_rate": 7.101787101787102e-06,
"loss": 0.0338,
"step": 3870
},
{
"epoch": 3.1,
"learning_rate": 7.0940170940170945e-06,
"loss": 0.0337,
"step": 3880
},
{
"epoch": 3.11,
"learning_rate": 7.086247086247087e-06,
"loss": 0.0331,
"step": 3890
},
{
"epoch": 3.12,
"learning_rate": 7.07847707847708e-06,
"loss": 0.0351,
"step": 3900
},
{
"epoch": 3.12,
"eval_loss": 0.04302978515625,
"eval_runtime": 528.551,
"eval_samples_per_second": 3.343,
"eval_steps_per_second": 0.026,
"eval_wer": 41.2283431891275,
"step": 3900
},
{
"epoch": 3.13,
"learning_rate": 7.070707070707071e-06,
"loss": 0.0338,
"step": 3910
},
{
"epoch": 3.13,
"learning_rate": 7.062937062937063e-06,
"loss": 0.0334,
"step": 3920
},
{
"epoch": 3.14,
"learning_rate": 7.055167055167056e-06,
"loss": 0.034,
"step": 3930
},
{
"epoch": 3.15,
"learning_rate": 7.047397047397048e-06,
"loss": 0.0336,
"step": 3940
},
{
"epoch": 3.16,
"learning_rate": 7.0396270396270405e-06,
"loss": 0.0339,
"step": 3950
},
{
"epoch": 3.17,
"learning_rate": 7.031857031857032e-06,
"loss": 0.0338,
"step": 3960
},
{
"epoch": 3.17,
"learning_rate": 7.024087024087025e-06,
"loss": 0.0354,
"step": 3970
},
{
"epoch": 3.18,
"learning_rate": 7.016317016317017e-06,
"loss": 0.0344,
"step": 3980
},
{
"epoch": 3.19,
"learning_rate": 7.008547008547009e-06,
"loss": 0.0324,
"step": 3990
},
{
"epoch": 3.2,
"learning_rate": 7.0007770007770015e-06,
"loss": 0.0352,
"step": 4000
},
{
"epoch": 3.2,
"eval_loss": 0.042999267578125,
"eval_runtime": 544.2571,
"eval_samples_per_second": 3.247,
"eval_steps_per_second": 0.026,
"eval_wer": 41.99605768233219,
"step": 4000
},
{
"epoch": 3.21,
"learning_rate": 6.993006993006993e-06,
"loss": 0.0331,
"step": 4010
},
{
"epoch": 3.21,
"learning_rate": 6.9852369852369865e-06,
"loss": 0.0328,
"step": 4020
},
{
"epoch": 3.22,
"learning_rate": 6.977466977466978e-06,
"loss": 0.0355,
"step": 4030
},
{
"epoch": 3.23,
"learning_rate": 6.969696969696971e-06,
"loss": 0.0345,
"step": 4040
},
{
"epoch": 3.24,
"learning_rate": 6.9619269619269624e-06,
"loss": 0.0336,
"step": 4050
},
{
"epoch": 3.25,
"learning_rate": 6.954156954156954e-06,
"loss": 0.0338,
"step": 4060
},
{
"epoch": 3.25,
"learning_rate": 6.9463869463869475e-06,
"loss": 0.0339,
"step": 4070
},
{
"epoch": 3.26,
"learning_rate": 6.938616938616939e-06,
"loss": 0.0342,
"step": 4080
},
{
"epoch": 3.27,
"learning_rate": 6.930846930846932e-06,
"loss": 0.0341,
"step": 4090
},
{
"epoch": 3.28,
"learning_rate": 6.923076923076923e-06,
"loss": 0.0333,
"step": 4100
},
{
"epoch": 3.28,
"eval_loss": 0.042816162109375,
"eval_runtime": 542.5061,
"eval_samples_per_second": 3.257,
"eval_steps_per_second": 0.026,
"eval_wer": 42.250233426704014,
"step": 4100
},
{
"epoch": 3.29,
"learning_rate": 6.915306915306917e-06,
"loss": 0.0343,
"step": 4110
},
{
"epoch": 3.29,
"learning_rate": 6.9075369075369084e-06,
"loss": 0.0346,
"step": 4120
},
{
"epoch": 3.3,
"learning_rate": 6.8997668997669e-06,
"loss": 0.0355,
"step": 4130
},
{
"epoch": 3.31,
"learning_rate": 6.891996891996893e-06,
"loss": 0.0346,
"step": 4140
},
{
"epoch": 3.32,
"learning_rate": 6.884226884226884e-06,
"loss": 0.0342,
"step": 4150
},
{
"epoch": 3.33,
"learning_rate": 6.876456876456878e-06,
"loss": 0.0351,
"step": 4160
},
{
"epoch": 3.33,
"learning_rate": 6.868686868686869e-06,
"loss": 0.0356,
"step": 4170
},
{
"epoch": 3.34,
"learning_rate": 6.860916860916862e-06,
"loss": 0.0352,
"step": 4180
},
{
"epoch": 3.35,
"learning_rate": 6.853146853146854e-06,
"loss": 0.0347,
"step": 4190
},
{
"epoch": 3.36,
"learning_rate": 6.845376845376845e-06,
"loss": 0.0338,
"step": 4200
},
{
"epoch": 3.36,
"eval_loss": 0.043304443359375,
"eval_runtime": 485.8042,
"eval_samples_per_second": 3.637,
"eval_steps_per_second": 0.029,
"eval_wer": 40.91192032368503,
"step": 4200
},
{
"epoch": 3.37,
"learning_rate": 6.837606837606839e-06,
"loss": 0.0342,
"step": 4210
},
{
"epoch": 3.37,
"learning_rate": 6.82983682983683e-06,
"loss": 0.0333,
"step": 4220
},
{
"epoch": 3.38,
"learning_rate": 6.822066822066823e-06,
"loss": 0.0344,
"step": 4230
},
{
"epoch": 3.39,
"learning_rate": 6.8142968142968145e-06,
"loss": 0.0339,
"step": 4240
},
{
"epoch": 3.4,
"learning_rate": 6.806526806526808e-06,
"loss": 0.0353,
"step": 4250
},
{
"epoch": 3.41,
"learning_rate": 6.7987567987568e-06,
"loss": 0.0347,
"step": 4260
},
{
"epoch": 3.41,
"learning_rate": 6.790986790986791e-06,
"loss": 0.0335,
"step": 4270
},
{
"epoch": 3.42,
"learning_rate": 6.783216783216784e-06,
"loss": 0.0336,
"step": 4280
},
{
"epoch": 3.43,
"learning_rate": 6.7754467754467755e-06,
"loss": 0.0325,
"step": 4290
},
{
"epoch": 3.44,
"learning_rate": 6.767676767676769e-06,
"loss": 0.0346,
"step": 4300
},
{
"epoch": 3.44,
"eval_loss": 0.04229736328125,
"eval_runtime": 415.5987,
"eval_samples_per_second": 4.252,
"eval_steps_per_second": 0.034,
"eval_wer": 40.74074074074074,
"step": 4300
}
],
"max_steps": 13000,
"num_train_epochs": 11,
"total_flos": 1.1231512173833817e+21,
"trial_name": null,
"trial_params": null
}