|
{ |
|
"best_metric": 0.6712434887886047, |
|
"best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v6-1/checkpoint-2590", |
|
"epoch": 99.99290780141844, |
|
"global_step": 3500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.28, |
|
"learning_rate": 6.666666666666667e-05, |
|
"loss": 55.7655, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.0001866666666666667, |
|
"loss": 58.6473, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 0.00032, |
|
"loss": 17.2739, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"eval_loss": 2.9041128158569336, |
|
"eval_runtime": 8.8924, |
|
"eval_samples_per_second": 7.085, |
|
"eval_steps_per_second": 1.799, |
|
"eval_wer": 1.0, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 0.00039953890489913545, |
|
"loss": 2.8972, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 0.0003983861671469741, |
|
"loss": 2.082, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 0.0003972334293948127, |
|
"loss": 2.1557, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 0.00039608069164265134, |
|
"loss": 1.8424, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"eval_loss": 3.505493402481079, |
|
"eval_runtime": 9.0402, |
|
"eval_samples_per_second": 6.969, |
|
"eval_steps_per_second": 1.77, |
|
"eval_wer": 1.0, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.28, |
|
"learning_rate": 0.0003949279538904899, |
|
"loss": 1.8014, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"learning_rate": 0.00039377521613832855, |
|
"loss": 1.6721, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"learning_rate": 0.0003926224783861672, |
|
"loss": 1.7092, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"eval_loss": 2.0045998096466064, |
|
"eval_runtime": 9.0436, |
|
"eval_samples_per_second": 6.966, |
|
"eval_steps_per_second": 1.769, |
|
"eval_wer": 1.0, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 0.00039146974063400576, |
|
"loss": 1.6035, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 3.43, |
|
"learning_rate": 0.0003903170028818444, |
|
"loss": 1.6196, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"learning_rate": 0.000389164265129683, |
|
"loss": 1.587, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"learning_rate": 0.00038801152737752165, |
|
"loss": 1.5022, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"eval_loss": 1.9661929607391357, |
|
"eval_runtime": 9.0602, |
|
"eval_samples_per_second": 6.954, |
|
"eval_steps_per_second": 1.766, |
|
"eval_wer": 0.9675242513707296, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 4.28, |
|
"learning_rate": 0.00038685878962536023, |
|
"loss": 1.486, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"learning_rate": 0.00038570605187319886, |
|
"loss": 1.382, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 4.85, |
|
"learning_rate": 0.0003845533141210375, |
|
"loss": 1.2964, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 4.99, |
|
"eval_loss": 1.9017366170883179, |
|
"eval_runtime": 8.7445, |
|
"eval_samples_per_second": 7.204, |
|
"eval_steps_per_second": 1.83, |
|
"eval_wer": 0.5504006748207507, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 5.14, |
|
"learning_rate": 0.0003834005763688761, |
|
"loss": 1.3537, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 0.0003822478386167147, |
|
"loss": 1.1033, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 5.71, |
|
"learning_rate": 0.00038109510086455333, |
|
"loss": 1.2447, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 5.99, |
|
"learning_rate": 0.00037994236311239196, |
|
"loss": 1.1235, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 5.99, |
|
"eval_loss": 1.9874823093414307, |
|
"eval_runtime": 8.6787, |
|
"eval_samples_per_second": 7.259, |
|
"eval_steps_per_second": 1.844, |
|
"eval_wer": 0.4643610291016449, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 6.28, |
|
"learning_rate": 0.00037878962536023054, |
|
"loss": 1.1092, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 6.57, |
|
"learning_rate": 0.00037763688760806917, |
|
"loss": 1.0707, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 6.85, |
|
"learning_rate": 0.0003764841498559078, |
|
"loss": 1.1056, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"eval_loss": 1.8791475296020508, |
|
"eval_runtime": 8.7777, |
|
"eval_samples_per_second": 7.177, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.47617039223956137, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"learning_rate": 0.00037533141210374643, |
|
"loss": 1.0885, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 7.43, |
|
"learning_rate": 0.000374178674351585, |
|
"loss": 0.9235, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 7.71, |
|
"learning_rate": 0.00037302593659942364, |
|
"loss": 1.0039, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 7.99, |
|
"learning_rate": 0.0003718731988472623, |
|
"loss": 0.8907, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 7.99, |
|
"eval_loss": 1.4811413288116455, |
|
"eval_runtime": 8.7992, |
|
"eval_samples_per_second": 7.16, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.467313369886124, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"learning_rate": 0.0003707204610951009, |
|
"loss": 1.0557, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 8.57, |
|
"learning_rate": 0.0003695677233429395, |
|
"loss": 0.9808, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 8.85, |
|
"learning_rate": 0.0003684149855907781, |
|
"loss": 0.8605, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 8.99, |
|
"eval_loss": 1.9113553762435913, |
|
"eval_runtime": 8.737, |
|
"eval_samples_per_second": 7.211, |
|
"eval_steps_per_second": 1.831, |
|
"eval_wer": 0.4479122733024041, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 9.14, |
|
"learning_rate": 0.00036726224783861674, |
|
"loss": 0.8891, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 9.43, |
|
"learning_rate": 0.0003661095100864553, |
|
"loss": 0.9639, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 9.71, |
|
"learning_rate": 0.00036495677233429395, |
|
"loss": 0.8316, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 9.99, |
|
"learning_rate": 0.0003638040345821326, |
|
"loss": 0.8498, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 9.99, |
|
"eval_loss": 1.210748553276062, |
|
"eval_runtime": 8.7375, |
|
"eval_samples_per_second": 7.21, |
|
"eval_steps_per_second": 1.831, |
|
"eval_wer": 0.47279628848587096, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 10.28, |
|
"learning_rate": 0.0003626512968299712, |
|
"loss": 0.9453, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 10.57, |
|
"learning_rate": 0.0003614985590778098, |
|
"loss": 0.8499, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 10.85, |
|
"learning_rate": 0.0003603458213256484, |
|
"loss": 0.7205, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 10.99, |
|
"eval_loss": 1.574381947517395, |
|
"eval_runtime": 8.7195, |
|
"eval_samples_per_second": 7.225, |
|
"eval_steps_per_second": 1.835, |
|
"eval_wer": 0.45255166596372837, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 11.14, |
|
"learning_rate": 0.00035919308357348705, |
|
"loss": 1.0183, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 11.43, |
|
"learning_rate": 0.0003580403458213257, |
|
"loss": 0.7708, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 11.71, |
|
"learning_rate": 0.00035688760806916426, |
|
"loss": 0.8779, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"learning_rate": 0.0003557348703170029, |
|
"loss": 0.8417, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"eval_loss": 1.468916654586792, |
|
"eval_runtime": 8.7774, |
|
"eval_samples_per_second": 7.178, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.453395191902151, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 12.28, |
|
"learning_rate": 0.0003545821325648415, |
|
"loss": 0.8197, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 12.57, |
|
"learning_rate": 0.0003534293948126801, |
|
"loss": 0.7148, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"learning_rate": 0.00035227665706051873, |
|
"loss": 0.7734, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 12.99, |
|
"eval_loss": 1.3531099557876587, |
|
"eval_runtime": 8.7009, |
|
"eval_samples_per_second": 7.241, |
|
"eval_steps_per_second": 1.839, |
|
"eval_wer": 0.4550822437789962, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 13.14, |
|
"learning_rate": 0.00035112391930835736, |
|
"loss": 0.7908, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 13.43, |
|
"learning_rate": 0.000349971181556196, |
|
"loss": 0.7157, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 13.71, |
|
"learning_rate": 0.00034881844380403457, |
|
"loss": 0.7873, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 13.99, |
|
"learning_rate": 0.00034766570605187326, |
|
"loss": 0.7762, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 13.99, |
|
"eval_loss": 1.2924444675445557, |
|
"eval_runtime": 8.7984, |
|
"eval_samples_per_second": 7.16, |
|
"eval_steps_per_second": 1.819, |
|
"eval_wer": 0.4664698439477014, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 14.28, |
|
"learning_rate": 0.00034651296829971183, |
|
"loss": 0.8195, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 14.57, |
|
"learning_rate": 0.00034536023054755047, |
|
"loss": 0.7457, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 14.85, |
|
"learning_rate": 0.00034420749279538904, |
|
"loss": 0.6812, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 14.99, |
|
"eval_loss": 1.0827070474624634, |
|
"eval_runtime": 8.7923, |
|
"eval_samples_per_second": 7.165, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.40995360607338677, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 15.14, |
|
"learning_rate": 0.0003430547550432277, |
|
"loss": 0.7591, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 15.43, |
|
"learning_rate": 0.0003419020172910663, |
|
"loss": 0.7642, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 0.0003407492795389049, |
|
"loss": 1.27, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 15.99, |
|
"learning_rate": 0.0003395965417867435, |
|
"loss": 0.7245, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 15.99, |
|
"eval_loss": 1.4069534540176392, |
|
"eval_runtime": 8.8121, |
|
"eval_samples_per_second": 7.149, |
|
"eval_steps_per_second": 1.816, |
|
"eval_wer": 0.43525938422606497, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 16.28, |
|
"learning_rate": 0.00033844380403458214, |
|
"loss": 0.7721, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 16.57, |
|
"learning_rate": 0.0003372910662824208, |
|
"loss": 0.6945, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 16.85, |
|
"learning_rate": 0.00033613832853025935, |
|
"loss": 0.6508, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 16.99, |
|
"eval_loss": 1.052022099494934, |
|
"eval_runtime": 8.7064, |
|
"eval_samples_per_second": 7.236, |
|
"eval_steps_per_second": 1.838, |
|
"eval_wer": 0.4086883171657528, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 17.14, |
|
"learning_rate": 0.00033498559077809804, |
|
"loss": 0.735, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 17.43, |
|
"learning_rate": 0.0003338328530259366, |
|
"loss": 0.6526, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 17.71, |
|
"learning_rate": 0.00033268011527377525, |
|
"loss": 0.6955, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 17.99, |
|
"learning_rate": 0.0003315273775216138, |
|
"loss": 0.7144, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 17.99, |
|
"eval_loss": 1.0728662014007568, |
|
"eval_runtime": 8.8004, |
|
"eval_samples_per_second": 7.159, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.42091944327288067, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 18.28, |
|
"learning_rate": 0.0003303746397694525, |
|
"loss": 0.6136, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 18.57, |
|
"learning_rate": 0.0003292219020172911, |
|
"loss": 0.6545, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 18.85, |
|
"learning_rate": 0.00032806916426512966, |
|
"loss": 0.6566, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 18.99, |
|
"eval_loss": 1.1671946048736572, |
|
"eval_runtime": 8.7903, |
|
"eval_samples_per_second": 7.167, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.4053142134120624, |
|
"step": 665 |
|
}, |
|
{ |
|
"epoch": 19.14, |
|
"learning_rate": 0.00032691642651296835, |
|
"loss": 0.7112, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 19.43, |
|
"learning_rate": 0.0003257636887608069, |
|
"loss": 0.6853, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 19.71, |
|
"learning_rate": 0.00032461095100864556, |
|
"loss": 0.7222, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 19.99, |
|
"learning_rate": 0.00032345821325648413, |
|
"loss": 0.5802, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 19.99, |
|
"eval_loss": 1.0128839015960693, |
|
"eval_runtime": 8.823, |
|
"eval_samples_per_second": 7.14, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.4015183466891607, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 20.28, |
|
"learning_rate": 0.0003223054755043228, |
|
"loss": 0.6989, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 20.57, |
|
"learning_rate": 0.0003211527377521614, |
|
"loss": 0.6103, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 20.85, |
|
"learning_rate": 0.00032, |
|
"loss": 0.5924, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 20.99, |
|
"eval_loss": 1.0761842727661133, |
|
"eval_runtime": 8.6827, |
|
"eval_samples_per_second": 7.256, |
|
"eval_steps_per_second": 1.843, |
|
"eval_wer": 0.40067482075073807, |
|
"step": 735 |
|
}, |
|
{ |
|
"epoch": 21.14, |
|
"learning_rate": 0.0003188472622478386, |
|
"loss": 0.641, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 21.43, |
|
"learning_rate": 0.0003176945244956773, |
|
"loss": 0.7465, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 21.71, |
|
"learning_rate": 0.00031654178674351587, |
|
"loss": 0.6215, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 21.99, |
|
"learning_rate": 0.00031538904899135444, |
|
"loss": 0.7051, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 21.99, |
|
"eval_loss": 1.025338053703308, |
|
"eval_runtime": 8.7155, |
|
"eval_samples_per_second": 7.229, |
|
"eval_steps_per_second": 1.836, |
|
"eval_wer": 0.4027836355967946, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 22.28, |
|
"learning_rate": 0.00031423631123919313, |
|
"loss": 0.6773, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 22.57, |
|
"learning_rate": 0.0003130835734870317, |
|
"loss": 0.6137, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 22.85, |
|
"learning_rate": 0.00031193083573487034, |
|
"loss": 0.5669, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 22.99, |
|
"eval_loss": 1.0526078939437866, |
|
"eval_runtime": 8.675, |
|
"eval_samples_per_second": 7.262, |
|
"eval_steps_per_second": 1.844, |
|
"eval_wer": 0.41881062842682415, |
|
"step": 805 |
|
}, |
|
{ |
|
"epoch": 23.14, |
|
"learning_rate": 0.0003107780979827089, |
|
"loss": 0.5976, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 23.43, |
|
"learning_rate": 0.0003096253602305476, |
|
"loss": 0.6745, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 23.71, |
|
"learning_rate": 0.0003084726224783862, |
|
"loss": 0.7289, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 23.99, |
|
"learning_rate": 0.0003073198847262248, |
|
"loss": 0.6209, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 23.99, |
|
"eval_loss": 1.0177326202392578, |
|
"eval_runtime": 8.832, |
|
"eval_samples_per_second": 7.133, |
|
"eval_steps_per_second": 1.812, |
|
"eval_wer": 0.4213412062420919, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 24.28, |
|
"learning_rate": 0.00030616714697406344, |
|
"loss": 0.6383, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 24.57, |
|
"learning_rate": 0.00030501440922190207, |
|
"loss": 0.5804, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 24.85, |
|
"learning_rate": 0.00030386167146974065, |
|
"loss": 0.635, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 24.99, |
|
"eval_loss": 0.929938554763794, |
|
"eval_runtime": 8.8255, |
|
"eval_samples_per_second": 7.138, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.401940109658372, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 25.14, |
|
"learning_rate": 0.0003027089337175792, |
|
"loss": 0.74, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 25.43, |
|
"learning_rate": 0.0003015561959654179, |
|
"loss": 0.6235, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 25.71, |
|
"learning_rate": 0.0003004034582132565, |
|
"loss": 0.6219, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 25.99, |
|
"learning_rate": 0.0002992507204610951, |
|
"loss": 0.5914, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 25.99, |
|
"eval_loss": 1.0058447122573853, |
|
"eval_runtime": 8.6702, |
|
"eval_samples_per_second": 7.266, |
|
"eval_steps_per_second": 1.845, |
|
"eval_wer": 0.4141712357654998, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 26.28, |
|
"learning_rate": 0.00029809798270893375, |
|
"loss": 0.6127, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 26.57, |
|
"learning_rate": 0.0002969452449567724, |
|
"loss": 0.5335, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 26.85, |
|
"learning_rate": 0.00029579250720461096, |
|
"loss": 0.5983, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 26.99, |
|
"eval_loss": 0.9719786643981934, |
|
"eval_runtime": 8.6673, |
|
"eval_samples_per_second": 7.269, |
|
"eval_steps_per_second": 1.846, |
|
"eval_wer": 0.4141712357654998, |
|
"step": 945 |
|
}, |
|
{ |
|
"epoch": 27.14, |
|
"learning_rate": 0.0002946397694524496, |
|
"loss": 0.559, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 27.43, |
|
"learning_rate": 0.0002934870317002882, |
|
"loss": 0.6208, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 27.71, |
|
"learning_rate": 0.00029233429394812685, |
|
"loss": 0.5678, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 27.99, |
|
"learning_rate": 0.0002911815561959654, |
|
"loss": 0.5631, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 27.99, |
|
"eval_loss": 0.8982506990432739, |
|
"eval_runtime": 8.7362, |
|
"eval_samples_per_second": 7.211, |
|
"eval_steps_per_second": 1.831, |
|
"eval_wer": 0.4027836355967946, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 28.28, |
|
"learning_rate": 0.000290028818443804, |
|
"loss": 0.5503, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 28.57, |
|
"learning_rate": 0.0002888760806916427, |
|
"loss": 0.5238, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 28.85, |
|
"learning_rate": 0.00028772334293948127, |
|
"loss": 0.552, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 28.99, |
|
"eval_loss": 0.914803683757782, |
|
"eval_runtime": 8.7166, |
|
"eval_samples_per_second": 7.228, |
|
"eval_steps_per_second": 1.836, |
|
"eval_wer": 0.41838886545761284, |
|
"step": 1015 |
|
}, |
|
{ |
|
"epoch": 29.14, |
|
"learning_rate": 0.0002865706051873199, |
|
"loss": 0.5816, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 29.43, |
|
"learning_rate": 0.00028541786743515853, |
|
"loss": 0.5295, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 29.71, |
|
"learning_rate": 0.00028426512968299716, |
|
"loss": 0.5742, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 29.99, |
|
"learning_rate": 0.00028311239193083574, |
|
"loss": 0.5213, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 29.99, |
|
"eval_loss": 1.0816500186920166, |
|
"eval_runtime": 8.9248, |
|
"eval_samples_per_second": 7.059, |
|
"eval_steps_per_second": 1.793, |
|
"eval_wer": 0.4141712357654998, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 30.28, |
|
"learning_rate": 0.00028195965417867437, |
|
"loss": 0.681, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 30.57, |
|
"learning_rate": 0.000280806916426513, |
|
"loss": 0.5139, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 30.85, |
|
"learning_rate": 0.00027965417867435163, |
|
"loss": 0.5387, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 30.99, |
|
"eval_loss": 0.9432059526443481, |
|
"eval_runtime": 8.8063, |
|
"eval_samples_per_second": 7.154, |
|
"eval_steps_per_second": 1.817, |
|
"eval_wer": 0.41881062842682415, |
|
"step": 1085 |
|
}, |
|
{ |
|
"epoch": 31.14, |
|
"learning_rate": 0.0002785014409221902, |
|
"loss": 0.5541, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 31.43, |
|
"learning_rate": 0.00027734870317002884, |
|
"loss": 0.6006, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 31.71, |
|
"learning_rate": 0.00027619596541786747, |
|
"loss": 0.6104, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 31.99, |
|
"learning_rate": 0.00027504322766570605, |
|
"loss": 0.5276, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 31.99, |
|
"eval_loss": 1.1206682920455933, |
|
"eval_runtime": 8.7507, |
|
"eval_samples_per_second": 7.199, |
|
"eval_steps_per_second": 1.828, |
|
"eval_wer": 0.40067482075073807, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 32.28, |
|
"learning_rate": 0.0002738904899135447, |
|
"loss": 0.511, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 32.57, |
|
"learning_rate": 0.0002727377521613833, |
|
"loss": 0.4948, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 32.85, |
|
"learning_rate": 0.00027158501440922194, |
|
"loss": 0.5778, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 32.99, |
|
"eval_loss": 0.9253532886505127, |
|
"eval_runtime": 8.8188, |
|
"eval_samples_per_second": 7.144, |
|
"eval_steps_per_second": 1.814, |
|
"eval_wer": 0.4150147617039224, |
|
"step": 1155 |
|
}, |
|
{ |
|
"epoch": 33.14, |
|
"learning_rate": 0.0002704322766570605, |
|
"loss": 0.551, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 33.43, |
|
"learning_rate": 0.00026927953890489915, |
|
"loss": 0.5115, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 33.71, |
|
"learning_rate": 0.0002681268011527378, |
|
"loss": 0.6795, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 33.99, |
|
"learning_rate": 0.0002669740634005764, |
|
"loss": 0.5001, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 33.99, |
|
"eval_loss": 1.0393145084381104, |
|
"eval_runtime": 8.783, |
|
"eval_samples_per_second": 7.173, |
|
"eval_steps_per_second": 1.822, |
|
"eval_wer": 0.4192323913960354, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 34.28, |
|
"learning_rate": 0.000265821325648415, |
|
"loss": 0.5146, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 34.57, |
|
"learning_rate": 0.0002646685878962536, |
|
"loss": 0.5165, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 34.85, |
|
"learning_rate": 0.00026351585014409225, |
|
"loss": 0.5329, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 34.99, |
|
"eval_loss": 0.9108927249908447, |
|
"eval_runtime": 8.8479, |
|
"eval_samples_per_second": 7.12, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.39645719105862504, |
|
"step": 1225 |
|
}, |
|
{ |
|
"epoch": 35.14, |
|
"learning_rate": 0.0002623631123919308, |
|
"loss": 0.5703, |
|
"step": 1230 |
|
}, |
|
{ |
|
"epoch": 35.43, |
|
"learning_rate": 0.00026121037463976946, |
|
"loss": 0.5296, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 35.71, |
|
"learning_rate": 0.0002600576368876081, |
|
"loss": 0.5212, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 35.99, |
|
"learning_rate": 0.0002589048991354467, |
|
"loss": 0.5168, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 35.99, |
|
"eval_loss": 0.8983320593833923, |
|
"eval_runtime": 8.8493, |
|
"eval_samples_per_second": 7.119, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.429776465626318, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 36.28, |
|
"learning_rate": 0.0002577521613832853, |
|
"loss": 0.528, |
|
"step": 1270 |
|
}, |
|
{ |
|
"epoch": 36.57, |
|
"learning_rate": 0.00025659942363112393, |
|
"loss": 0.4487, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 36.85, |
|
"learning_rate": 0.00025544668587896256, |
|
"loss": 0.4918, |
|
"step": 1290 |
|
}, |
|
{ |
|
"epoch": 36.99, |
|
"eval_loss": 0.8411929607391357, |
|
"eval_runtime": 8.8153, |
|
"eval_samples_per_second": 7.147, |
|
"eval_steps_per_second": 1.815, |
|
"eval_wer": 0.4086883171657528, |
|
"step": 1295 |
|
}, |
|
{ |
|
"epoch": 37.14, |
|
"learning_rate": 0.0002542939481268012, |
|
"loss": 0.5363, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 37.43, |
|
"learning_rate": 0.00025314121037463977, |
|
"loss": 0.4954, |
|
"step": 1310 |
|
}, |
|
{ |
|
"epoch": 37.71, |
|
"learning_rate": 0.0002519884726224784, |
|
"loss": 0.4878, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 37.99, |
|
"learning_rate": 0.00025083573487031703, |
|
"loss": 0.5651, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 37.99, |
|
"eval_loss": 0.8560371398925781, |
|
"eval_runtime": 9.0193, |
|
"eval_samples_per_second": 6.985, |
|
"eval_steps_per_second": 1.774, |
|
"eval_wer": 0.42176296921130324, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 38.28, |
|
"learning_rate": 0.0002496829971181556, |
|
"loss": 0.5888, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 38.57, |
|
"learning_rate": 0.00024853025936599424, |
|
"loss": 0.485, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 38.85, |
|
"learning_rate": 0.00024737752161383287, |
|
"loss": 0.438, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 38.99, |
|
"eval_loss": 0.8555735349655151, |
|
"eval_runtime": 8.8578, |
|
"eval_samples_per_second": 7.112, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.4171235765499789, |
|
"step": 1365 |
|
}, |
|
{ |
|
"epoch": 39.14, |
|
"learning_rate": 0.0002462247838616715, |
|
"loss": 0.5556, |
|
"step": 1370 |
|
}, |
|
{ |
|
"epoch": 39.43, |
|
"learning_rate": 0.0002450720461095101, |
|
"loss": 0.4795, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 39.71, |
|
"learning_rate": 0.00024391930835734874, |
|
"loss": 0.4785, |
|
"step": 1390 |
|
}, |
|
{ |
|
"epoch": 39.99, |
|
"learning_rate": 0.00024276657060518734, |
|
"loss": 0.4808, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 39.99, |
|
"eval_loss": 0.8319878578186035, |
|
"eval_runtime": 8.7692, |
|
"eval_samples_per_second": 7.184, |
|
"eval_steps_per_second": 1.825, |
|
"eval_wer": 0.4175453395191902, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 40.28, |
|
"learning_rate": 0.00024161383285302597, |
|
"loss": 0.5304, |
|
"step": 1410 |
|
}, |
|
{ |
|
"epoch": 40.57, |
|
"learning_rate": 0.00024046109510086458, |
|
"loss": 0.4896, |
|
"step": 1420 |
|
}, |
|
{ |
|
"epoch": 40.85, |
|
"learning_rate": 0.00023930835734870315, |
|
"loss": 0.5372, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 40.99, |
|
"eval_loss": 0.9745492339134216, |
|
"eval_runtime": 8.7927, |
|
"eval_samples_per_second": 7.165, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.39561366512020246, |
|
"step": 1435 |
|
}, |
|
{ |
|
"epoch": 41.14, |
|
"learning_rate": 0.0002381556195965418, |
|
"loss": 0.4811, |
|
"step": 1440 |
|
}, |
|
{ |
|
"epoch": 41.43, |
|
"learning_rate": 0.0002370028818443804, |
|
"loss": 0.4714, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 41.71, |
|
"learning_rate": 0.00023585014409221905, |
|
"loss": 0.5127, |
|
"step": 1460 |
|
}, |
|
{ |
|
"epoch": 41.99, |
|
"learning_rate": 0.00023469740634005762, |
|
"loss": 0.4814, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 41.99, |
|
"eval_loss": 0.8032535910606384, |
|
"eval_runtime": 8.8256, |
|
"eval_samples_per_second": 7.138, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.4120624209194433, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 42.28, |
|
"learning_rate": 0.00023354466858789628, |
|
"loss": 0.4302, |
|
"step": 1480 |
|
}, |
|
{ |
|
"epoch": 42.57, |
|
"learning_rate": 0.00023239193083573489, |
|
"loss": 0.4579, |
|
"step": 1490 |
|
}, |
|
{ |
|
"epoch": 42.85, |
|
"learning_rate": 0.00023123919308357352, |
|
"loss": 0.4416, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 42.99, |
|
"eval_loss": 0.8194773197174072, |
|
"eval_runtime": 8.7558, |
|
"eval_samples_per_second": 7.195, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.39898776887389287, |
|
"step": 1505 |
|
}, |
|
{ |
|
"epoch": 43.14, |
|
"learning_rate": 0.00023008645533141212, |
|
"loss": 0.4628, |
|
"step": 1510 |
|
}, |
|
{ |
|
"epoch": 43.43, |
|
"learning_rate": 0.00022893371757925075, |
|
"loss": 0.4765, |
|
"step": 1520 |
|
}, |
|
{ |
|
"epoch": 43.71, |
|
"learning_rate": 0.00022778097982708936, |
|
"loss": 0.4504, |
|
"step": 1530 |
|
}, |
|
{ |
|
"epoch": 43.99, |
|
"learning_rate": 0.00022662824207492793, |
|
"loss": 0.4958, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 43.99, |
|
"eval_loss": 0.8263903856277466, |
|
"eval_runtime": 8.775, |
|
"eval_samples_per_second": 7.179, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.39561366512020246, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 44.28, |
|
"learning_rate": 0.0002254755043227666, |
|
"loss": 0.4835, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 44.57, |
|
"learning_rate": 0.00022432276657060517, |
|
"loss": 0.4188, |
|
"step": 1560 |
|
}, |
|
{ |
|
"epoch": 44.85, |
|
"learning_rate": 0.00022317002881844383, |
|
"loss": 0.4665, |
|
"step": 1570 |
|
}, |
|
{ |
|
"epoch": 44.99, |
|
"eval_loss": 0.8172122240066528, |
|
"eval_runtime": 8.7583, |
|
"eval_samples_per_second": 7.193, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.4070012652889076, |
|
"step": 1575 |
|
}, |
|
{ |
|
"epoch": 45.14, |
|
"learning_rate": 0.00022201729106628243, |
|
"loss": 0.4708, |
|
"step": 1580 |
|
}, |
|
{ |
|
"epoch": 45.43, |
|
"learning_rate": 0.00022086455331412106, |
|
"loss": 0.4321, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 45.71, |
|
"learning_rate": 0.00021971181556195967, |
|
"loss": 0.5132, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 45.99, |
|
"learning_rate": 0.0002185590778097983, |
|
"loss": 0.4196, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 45.99, |
|
"eval_loss": 0.7970888018608093, |
|
"eval_runtime": 8.8547, |
|
"eval_samples_per_second": 7.115, |
|
"eval_steps_per_second": 1.807, |
|
"eval_wer": 0.39519190215099115, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 46.28, |
|
"learning_rate": 0.0002174063400576369, |
|
"loss": 0.5437, |
|
"step": 1620 |
|
}, |
|
{ |
|
"epoch": 46.57, |
|
"learning_rate": 0.00021625360230547553, |
|
"loss": 0.4058, |
|
"step": 1630 |
|
}, |
|
{ |
|
"epoch": 46.85, |
|
"learning_rate": 0.00021510086455331414, |
|
"loss": 0.4088, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 46.99, |
|
"eval_loss": 0.7417133450508118, |
|
"eval_runtime": 8.7997, |
|
"eval_samples_per_second": 7.159, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.38802193167439897, |
|
"step": 1645 |
|
}, |
|
{ |
|
"epoch": 47.14, |
|
"learning_rate": 0.0002139481268011527, |
|
"loss": 0.4692, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 47.43, |
|
"learning_rate": 0.00021279538904899137, |
|
"loss": 0.4082, |
|
"step": 1660 |
|
}, |
|
{ |
|
"epoch": 47.71, |
|
"learning_rate": 0.00021164265129682998, |
|
"loss": 0.4934, |
|
"step": 1670 |
|
}, |
|
{ |
|
"epoch": 47.99, |
|
"learning_rate": 0.0002104899135446686, |
|
"loss": 0.4308, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 47.99, |
|
"eval_loss": 0.7806001901626587, |
|
"eval_runtime": 8.8422, |
|
"eval_samples_per_second": 7.125, |
|
"eval_steps_per_second": 1.81, |
|
"eval_wer": 0.39308308730493463, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 48.28, |
|
"learning_rate": 0.0002093371757925072, |
|
"loss": 0.4835, |
|
"step": 1690 |
|
}, |
|
{ |
|
"epoch": 48.57, |
|
"learning_rate": 0.00020818443804034584, |
|
"loss": 0.4237, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 48.85, |
|
"learning_rate": 0.00020703170028818445, |
|
"loss": 0.4173, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 48.99, |
|
"eval_loss": 0.7380473613739014, |
|
"eval_runtime": 8.9386, |
|
"eval_samples_per_second": 7.048, |
|
"eval_steps_per_second": 1.79, |
|
"eval_wer": 0.392239561366512, |
|
"step": 1715 |
|
}, |
|
{ |
|
"epoch": 49.14, |
|
"learning_rate": 0.00020587896253602308, |
|
"loss": 0.6076, |
|
"step": 1720 |
|
}, |
|
{ |
|
"epoch": 49.43, |
|
"learning_rate": 0.00020472622478386168, |
|
"loss": 0.4505, |
|
"step": 1730 |
|
}, |
|
{ |
|
"epoch": 49.71, |
|
"learning_rate": 0.0002035734870317003, |
|
"loss": 0.4394, |
|
"step": 1740 |
|
}, |
|
{ |
|
"epoch": 49.99, |
|
"learning_rate": 0.00020242074927953892, |
|
"loss": 0.4653, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 49.99, |
|
"eval_loss": 0.8961731195449829, |
|
"eval_runtime": 8.8706, |
|
"eval_samples_per_second": 7.102, |
|
"eval_steps_per_second": 1.804, |
|
"eval_wer": 0.4027836355967946, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 50.28, |
|
"learning_rate": 0.00020126801152737752, |
|
"loss": 0.4578, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 50.57, |
|
"learning_rate": 0.00020011527377521615, |
|
"loss": 1.0308, |
|
"step": 1770 |
|
}, |
|
{ |
|
"epoch": 50.85, |
|
"learning_rate": 0.00019896253602305476, |
|
"loss": 0.4406, |
|
"step": 1780 |
|
}, |
|
{ |
|
"epoch": 50.99, |
|
"eval_loss": 0.7790281176567078, |
|
"eval_runtime": 8.8636, |
|
"eval_samples_per_second": 7.108, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.39350485027414595, |
|
"step": 1785 |
|
}, |
|
{ |
|
"epoch": 51.14, |
|
"learning_rate": 0.0001978097982708934, |
|
"loss": 0.5174, |
|
"step": 1790 |
|
}, |
|
{ |
|
"epoch": 51.43, |
|
"learning_rate": 0.00019665706051873202, |
|
"loss": 0.489, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 51.71, |
|
"learning_rate": 0.00019550432276657062, |
|
"loss": 0.6726, |
|
"step": 1810 |
|
}, |
|
{ |
|
"epoch": 51.99, |
|
"learning_rate": 0.00019435158501440923, |
|
"loss": 0.4664, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 51.99, |
|
"eval_loss": 0.917263925075531, |
|
"eval_runtime": 8.8897, |
|
"eval_samples_per_second": 7.087, |
|
"eval_steps_per_second": 1.8, |
|
"eval_wer": 0.3892872205820329, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 52.28, |
|
"learning_rate": 0.00019319884726224783, |
|
"loss": 0.4466, |
|
"step": 1830 |
|
}, |
|
{ |
|
"epoch": 52.57, |
|
"learning_rate": 0.00019204610951008646, |
|
"loss": 0.4465, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 52.85, |
|
"learning_rate": 0.00019089337175792507, |
|
"loss": 0.4486, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 52.99, |
|
"eval_loss": 0.8234918713569641, |
|
"eval_runtime": 8.8605, |
|
"eval_samples_per_second": 7.11, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.392239561366512, |
|
"step": 1855 |
|
}, |
|
{ |
|
"epoch": 53.14, |
|
"learning_rate": 0.0001897406340057637, |
|
"loss": 0.4336, |
|
"step": 1860 |
|
}, |
|
{ |
|
"epoch": 53.43, |
|
"learning_rate": 0.0001885878962536023, |
|
"loss": 0.4376, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 53.71, |
|
"learning_rate": 0.00018743515850144093, |
|
"loss": 0.4706, |
|
"step": 1880 |
|
}, |
|
{ |
|
"epoch": 53.99, |
|
"learning_rate": 0.00018628242074927956, |
|
"loss": 0.4137, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 53.99, |
|
"eval_loss": 0.8031628727912903, |
|
"eval_runtime": 8.8634, |
|
"eval_samples_per_second": 7.108, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.3926613243357233, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 54.28, |
|
"learning_rate": 0.00018512968299711817, |
|
"loss": 0.4758, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 54.57, |
|
"learning_rate": 0.0001839769452449568, |
|
"loss": 0.3621, |
|
"step": 1910 |
|
}, |
|
{ |
|
"epoch": 54.85, |
|
"learning_rate": 0.0001828242074927954, |
|
"loss": 0.4402, |
|
"step": 1920 |
|
}, |
|
{ |
|
"epoch": 54.99, |
|
"eval_loss": 0.7658433318138123, |
|
"eval_runtime": 8.8656, |
|
"eval_samples_per_second": 7.106, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.38296077604386336, |
|
"step": 1925 |
|
}, |
|
{ |
|
"epoch": 55.14, |
|
"learning_rate": 0.000181671469740634, |
|
"loss": 0.4614, |
|
"step": 1930 |
|
}, |
|
{ |
|
"epoch": 55.43, |
|
"learning_rate": 0.0001805187319884726, |
|
"loss": 0.4183, |
|
"step": 1940 |
|
}, |
|
{ |
|
"epoch": 55.71, |
|
"learning_rate": 0.00017936599423631124, |
|
"loss": 0.4065, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 55.99, |
|
"learning_rate": 0.00017821325648414985, |
|
"loss": 0.4101, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 55.99, |
|
"eval_loss": 0.8620884418487549, |
|
"eval_runtime": 8.9342, |
|
"eval_samples_per_second": 7.052, |
|
"eval_steps_per_second": 1.791, |
|
"eval_wer": 0.3994095318431042, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 56.28, |
|
"learning_rate": 0.00017706051873198848, |
|
"loss": 0.5536, |
|
"step": 1970 |
|
}, |
|
{ |
|
"epoch": 56.57, |
|
"learning_rate": 0.0001759077809798271, |
|
"loss": 0.4167, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 56.85, |
|
"learning_rate": 0.0001747550432276657, |
|
"loss": 0.5239, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 56.99, |
|
"eval_loss": 0.7903460264205933, |
|
"eval_runtime": 8.918, |
|
"eval_samples_per_second": 7.064, |
|
"eval_steps_per_second": 1.794, |
|
"eval_wer": 0.39561366512020246, |
|
"step": 1995 |
|
}, |
|
{ |
|
"epoch": 57.14, |
|
"learning_rate": 0.00017360230547550434, |
|
"loss": 0.6093, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 57.43, |
|
"learning_rate": 0.00017244956772334295, |
|
"loss": 0.4394, |
|
"step": 2010 |
|
}, |
|
{ |
|
"epoch": 57.71, |
|
"learning_rate": 0.00017129682997118158, |
|
"loss": 0.4504, |
|
"step": 2020 |
|
}, |
|
{ |
|
"epoch": 57.99, |
|
"learning_rate": 0.00017014409221902018, |
|
"loss": 0.4151, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 57.99, |
|
"eval_loss": 0.7849175333976746, |
|
"eval_runtime": 8.9241, |
|
"eval_samples_per_second": 7.06, |
|
"eval_steps_per_second": 1.793, |
|
"eval_wer": 0.3871784057359764, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 58.28, |
|
"learning_rate": 0.0001689913544668588, |
|
"loss": 0.463, |
|
"step": 2040 |
|
}, |
|
{ |
|
"epoch": 58.57, |
|
"learning_rate": 0.00016783861671469742, |
|
"loss": 0.4195, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 58.85, |
|
"learning_rate": 0.00016668587896253602, |
|
"loss": 0.4766, |
|
"step": 2060 |
|
}, |
|
{ |
|
"epoch": 58.99, |
|
"eval_loss": 0.8306421041488647, |
|
"eval_runtime": 9.0122, |
|
"eval_samples_per_second": 6.99, |
|
"eval_steps_per_second": 1.775, |
|
"eval_wer": 0.39181779839730074, |
|
"step": 2065 |
|
}, |
|
{ |
|
"epoch": 59.14, |
|
"learning_rate": 0.00016553314121037465, |
|
"loss": 0.4292, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 59.43, |
|
"learning_rate": 0.00016438040345821326, |
|
"loss": 0.4278, |
|
"step": 2080 |
|
}, |
|
{ |
|
"epoch": 59.71, |
|
"learning_rate": 0.0001632276657060519, |
|
"loss": 0.4586, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 59.99, |
|
"learning_rate": 0.0001620749279538905, |
|
"loss": 0.4882, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 59.99, |
|
"eval_loss": 0.8134064674377441, |
|
"eval_runtime": 8.8415, |
|
"eval_samples_per_second": 7.125, |
|
"eval_steps_per_second": 1.81, |
|
"eval_wer": 0.3926613243357233, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 60.28, |
|
"learning_rate": 0.00016092219020172912, |
|
"loss": 0.3925, |
|
"step": 2110 |
|
}, |
|
{ |
|
"epoch": 60.57, |
|
"learning_rate": 0.00015976945244956773, |
|
"loss": 0.4318, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 60.85, |
|
"learning_rate": 0.00015861671469740636, |
|
"loss": 0.4583, |
|
"step": 2130 |
|
}, |
|
{ |
|
"epoch": 60.99, |
|
"eval_loss": 0.9527338743209839, |
|
"eval_runtime": 8.8501, |
|
"eval_samples_per_second": 7.119, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.3850695908899199, |
|
"step": 2135 |
|
}, |
|
{ |
|
"epoch": 61.14, |
|
"learning_rate": 0.00015746397694524496, |
|
"loss": 0.6257, |
|
"step": 2140 |
|
}, |
|
{ |
|
"epoch": 61.43, |
|
"learning_rate": 0.0001563112391930836, |
|
"loss": 0.3906, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 61.71, |
|
"learning_rate": 0.0001551585014409222, |
|
"loss": 0.468, |
|
"step": 2160 |
|
}, |
|
{ |
|
"epoch": 61.99, |
|
"learning_rate": 0.0001540057636887608, |
|
"loss": 0.4284, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 61.99, |
|
"eval_loss": 0.9743154048919678, |
|
"eval_runtime": 8.9267, |
|
"eval_samples_per_second": 7.057, |
|
"eval_steps_per_second": 1.792, |
|
"eval_wer": 0.3998312948123155, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 62.28, |
|
"learning_rate": 0.00015285302593659943, |
|
"loss": 0.3827, |
|
"step": 2180 |
|
}, |
|
{ |
|
"epoch": 62.57, |
|
"learning_rate": 0.00015170028818443804, |
|
"loss": 0.3974, |
|
"step": 2190 |
|
}, |
|
{ |
|
"epoch": 62.85, |
|
"learning_rate": 0.00015054755043227667, |
|
"loss": 0.46, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 62.99, |
|
"eval_loss": 0.7807419300079346, |
|
"eval_runtime": 8.9409, |
|
"eval_samples_per_second": 7.046, |
|
"eval_steps_per_second": 1.79, |
|
"eval_wer": 0.38296077604386336, |
|
"step": 2205 |
|
}, |
|
{ |
|
"epoch": 63.14, |
|
"learning_rate": 0.00014939481268011527, |
|
"loss": 0.4637, |
|
"step": 2210 |
|
}, |
|
{ |
|
"epoch": 63.43, |
|
"learning_rate": 0.0001482420749279539, |
|
"loss": 0.3948, |
|
"step": 2220 |
|
}, |
|
{ |
|
"epoch": 63.71, |
|
"learning_rate": 0.0001470893371757925, |
|
"loss": 0.4112, |
|
"step": 2230 |
|
}, |
|
{ |
|
"epoch": 63.99, |
|
"learning_rate": 0.00014593659942363114, |
|
"loss": 0.4039, |
|
"step": 2240 |
|
}, |
|
{ |
|
"epoch": 63.99, |
|
"eval_loss": 0.8864094614982605, |
|
"eval_runtime": 8.7598, |
|
"eval_samples_per_second": 7.192, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.3884436946436103, |
|
"step": 2240 |
|
}, |
|
{ |
|
"epoch": 64.28, |
|
"learning_rate": 0.00014478386167146974, |
|
"loss": 0.3935, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 64.57, |
|
"learning_rate": 0.00014363112391930838, |
|
"loss": 0.3677, |
|
"step": 2260 |
|
}, |
|
{ |
|
"epoch": 64.85, |
|
"learning_rate": 0.00014247838616714698, |
|
"loss": 0.3868, |
|
"step": 2270 |
|
}, |
|
{ |
|
"epoch": 64.99, |
|
"eval_loss": 0.7304245233535767, |
|
"eval_runtime": 8.9023, |
|
"eval_samples_per_second": 7.077, |
|
"eval_steps_per_second": 1.797, |
|
"eval_wer": 0.3816954871362294, |
|
"step": 2275 |
|
}, |
|
{ |
|
"epoch": 65.14, |
|
"learning_rate": 0.00014132564841498558, |
|
"loss": 0.4321, |
|
"step": 2280 |
|
}, |
|
{ |
|
"epoch": 65.43, |
|
"learning_rate": 0.00014017291066282421, |
|
"loss": 0.3726, |
|
"step": 2290 |
|
}, |
|
{ |
|
"epoch": 65.71, |
|
"learning_rate": 0.00013902017291066282, |
|
"loss": 0.428, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 65.99, |
|
"learning_rate": 0.00013786743515850145, |
|
"loss": 0.3934, |
|
"step": 2310 |
|
}, |
|
{ |
|
"epoch": 65.99, |
|
"eval_loss": 0.875798761844635, |
|
"eval_runtime": 8.8052, |
|
"eval_samples_per_second": 7.155, |
|
"eval_steps_per_second": 1.817, |
|
"eval_wer": 0.38464782792070856, |
|
"step": 2310 |
|
}, |
|
{ |
|
"epoch": 66.28, |
|
"learning_rate": 0.00013671469740634005, |
|
"loss": 0.3963, |
|
"step": 2320 |
|
}, |
|
{ |
|
"epoch": 66.57, |
|
"learning_rate": 0.00013556195965417869, |
|
"loss": 0.3708, |
|
"step": 2330 |
|
}, |
|
{ |
|
"epoch": 66.85, |
|
"learning_rate": 0.0001344092219020173, |
|
"loss": 0.3776, |
|
"step": 2340 |
|
}, |
|
{ |
|
"epoch": 66.99, |
|
"eval_loss": 0.8156087398529053, |
|
"eval_runtime": 8.9243, |
|
"eval_samples_per_second": 7.059, |
|
"eval_steps_per_second": 1.793, |
|
"eval_wer": 0.3762125685364825, |
|
"step": 2345 |
|
}, |
|
{ |
|
"epoch": 67.14, |
|
"learning_rate": 0.00013325648414985592, |
|
"loss": 0.4032, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 67.43, |
|
"learning_rate": 0.00013210374639769455, |
|
"loss": 0.4047, |
|
"step": 2360 |
|
}, |
|
{ |
|
"epoch": 67.71, |
|
"learning_rate": 0.00013095100864553316, |
|
"loss": 0.4063, |
|
"step": 2370 |
|
}, |
|
{ |
|
"epoch": 67.99, |
|
"learning_rate": 0.00012979827089337176, |
|
"loss": 0.3499, |
|
"step": 2380 |
|
}, |
|
{ |
|
"epoch": 67.99, |
|
"eval_loss": 0.814339816570282, |
|
"eval_runtime": 8.9099, |
|
"eval_samples_per_second": 7.071, |
|
"eval_steps_per_second": 1.796, |
|
"eval_wer": 0.3888654576128216, |
|
"step": 2380 |
|
}, |
|
{ |
|
"epoch": 68.28, |
|
"learning_rate": 0.00012864553314121036, |
|
"loss": 0.4262, |
|
"step": 2390 |
|
}, |
|
{ |
|
"epoch": 68.57, |
|
"learning_rate": 0.000127492795389049, |
|
"loss": 0.4213, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 68.85, |
|
"learning_rate": 0.0001263400576368876, |
|
"loss": 0.4055, |
|
"step": 2410 |
|
}, |
|
{ |
|
"epoch": 68.99, |
|
"eval_loss": 0.7502649426460266, |
|
"eval_runtime": 8.8806, |
|
"eval_samples_per_second": 7.094, |
|
"eval_steps_per_second": 1.802, |
|
"eval_wer": 0.3795866722901729, |
|
"step": 2415 |
|
}, |
|
{ |
|
"epoch": 69.14, |
|
"learning_rate": 0.00012518731988472623, |
|
"loss": 0.4393, |
|
"step": 2420 |
|
}, |
|
{ |
|
"epoch": 69.43, |
|
"learning_rate": 0.00012403458213256483, |
|
"loss": 0.3695, |
|
"step": 2430 |
|
}, |
|
{ |
|
"epoch": 69.71, |
|
"learning_rate": 0.00012288184438040347, |
|
"loss": 0.4452, |
|
"step": 2440 |
|
}, |
|
{ |
|
"epoch": 69.99, |
|
"learning_rate": 0.00012172910662824208, |
|
"loss": 0.3505, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 69.99, |
|
"eval_loss": 0.7138158082962036, |
|
"eval_runtime": 8.8451, |
|
"eval_samples_per_second": 7.123, |
|
"eval_steps_per_second": 1.809, |
|
"eval_wer": 0.38043019822859553, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 70.28, |
|
"learning_rate": 0.0001205763688760807, |
|
"loss": 0.4222, |
|
"step": 2460 |
|
}, |
|
{ |
|
"epoch": 70.57, |
|
"learning_rate": 0.00011942363112391932, |
|
"loss": 0.3144, |
|
"step": 2470 |
|
}, |
|
{ |
|
"epoch": 70.85, |
|
"learning_rate": 0.00011827089337175794, |
|
"loss": 0.3755, |
|
"step": 2480 |
|
}, |
|
{ |
|
"epoch": 70.99, |
|
"eval_loss": 0.8071521520614624, |
|
"eval_runtime": 8.8022, |
|
"eval_samples_per_second": 7.157, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.3800084352593842, |
|
"step": 2485 |
|
}, |
|
{ |
|
"epoch": 71.14, |
|
"learning_rate": 0.00011711815561959654, |
|
"loss": 0.3778, |
|
"step": 2490 |
|
}, |
|
{ |
|
"epoch": 71.43, |
|
"learning_rate": 0.00011596541786743516, |
|
"loss": 0.3831, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 71.71, |
|
"learning_rate": 0.00011481268011527378, |
|
"loss": 0.3552, |
|
"step": 2510 |
|
}, |
|
{ |
|
"epoch": 71.99, |
|
"learning_rate": 0.0001136599423631124, |
|
"loss": 0.3594, |
|
"step": 2520 |
|
}, |
|
{ |
|
"epoch": 71.99, |
|
"eval_loss": 0.7692469358444214, |
|
"eval_runtime": 8.7672, |
|
"eval_samples_per_second": 7.186, |
|
"eval_steps_per_second": 1.825, |
|
"eval_wer": 0.3850695908899199, |
|
"step": 2520 |
|
}, |
|
{ |
|
"epoch": 72.28, |
|
"learning_rate": 0.00011250720461095101, |
|
"loss": 0.4549, |
|
"step": 2530 |
|
}, |
|
{ |
|
"epoch": 72.57, |
|
"learning_rate": 0.00011135446685878963, |
|
"loss": 0.3636, |
|
"step": 2540 |
|
}, |
|
{ |
|
"epoch": 72.85, |
|
"learning_rate": 0.00011020172910662825, |
|
"loss": 0.3167, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 72.99, |
|
"eval_loss": 0.6994994282722473, |
|
"eval_runtime": 8.766, |
|
"eval_samples_per_second": 7.187, |
|
"eval_steps_per_second": 1.825, |
|
"eval_wer": 0.3745255166596373, |
|
"step": 2555 |
|
}, |
|
{ |
|
"epoch": 73.14, |
|
"learning_rate": 0.00010904899135446686, |
|
"loss": 0.3651, |
|
"step": 2560 |
|
}, |
|
{ |
|
"epoch": 73.43, |
|
"learning_rate": 0.00010789625360230548, |
|
"loss": 0.3751, |
|
"step": 2570 |
|
}, |
|
{ |
|
"epoch": 73.71, |
|
"learning_rate": 0.0001067435158501441, |
|
"loss": 0.4196, |
|
"step": 2580 |
|
}, |
|
{ |
|
"epoch": 73.99, |
|
"learning_rate": 0.00010559077809798272, |
|
"loss": 0.3915, |
|
"step": 2590 |
|
}, |
|
{ |
|
"epoch": 73.99, |
|
"eval_loss": 0.6712434887886047, |
|
"eval_runtime": 8.8181, |
|
"eval_samples_per_second": 7.144, |
|
"eval_steps_per_second": 1.814, |
|
"eval_wer": 0.3762125685364825, |
|
"step": 2590 |
|
}, |
|
{ |
|
"epoch": 74.28, |
|
"learning_rate": 0.00010443804034582132, |
|
"loss": 0.3952, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 74.57, |
|
"learning_rate": 0.00010328530259365994, |
|
"loss": 0.3115, |
|
"step": 2610 |
|
}, |
|
{ |
|
"epoch": 74.85, |
|
"learning_rate": 0.00010213256484149856, |
|
"loss": 0.3741, |
|
"step": 2620 |
|
}, |
|
{ |
|
"epoch": 74.99, |
|
"eval_loss": 0.7139262557029724, |
|
"eval_runtime": 8.8329, |
|
"eval_samples_per_second": 7.132, |
|
"eval_steps_per_second": 1.811, |
|
"eval_wer": 0.3800084352593842, |
|
"step": 2625 |
|
}, |
|
{ |
|
"epoch": 75.14, |
|
"learning_rate": 0.00010097982708933717, |
|
"loss": 0.3542, |
|
"step": 2630 |
|
}, |
|
{ |
|
"epoch": 75.43, |
|
"learning_rate": 9.982708933717579e-05, |
|
"loss": 0.4682, |
|
"step": 2640 |
|
}, |
|
{ |
|
"epoch": 75.71, |
|
"learning_rate": 9.867435158501441e-05, |
|
"loss": 0.3378, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 75.99, |
|
"learning_rate": 9.752161383285303e-05, |
|
"loss": 0.3708, |
|
"step": 2660 |
|
}, |
|
{ |
|
"epoch": 75.99, |
|
"eval_loss": 0.7064934968948364, |
|
"eval_runtime": 8.8273, |
|
"eval_samples_per_second": 7.137, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2660 |
|
}, |
|
{ |
|
"epoch": 76.28, |
|
"learning_rate": 9.636887608069164e-05, |
|
"loss": 0.4179, |
|
"step": 2670 |
|
}, |
|
{ |
|
"epoch": 76.57, |
|
"learning_rate": 9.521613832853028e-05, |
|
"loss": 0.3543, |
|
"step": 2680 |
|
}, |
|
{ |
|
"epoch": 76.85, |
|
"learning_rate": 9.406340057636888e-05, |
|
"loss": 0.3731, |
|
"step": 2690 |
|
}, |
|
{ |
|
"epoch": 76.99, |
|
"eval_loss": 0.7315667867660522, |
|
"eval_runtime": 8.8516, |
|
"eval_samples_per_second": 7.117, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.37536904259805987, |
|
"step": 2695 |
|
}, |
|
{ |
|
"epoch": 77.14, |
|
"learning_rate": 9.29106628242075e-05, |
|
"loss": 0.4854, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 77.43, |
|
"learning_rate": 9.175792507204611e-05, |
|
"loss": 0.3493, |
|
"step": 2710 |
|
}, |
|
{ |
|
"epoch": 77.71, |
|
"learning_rate": 9.060518731988473e-05, |
|
"loss": 0.3559, |
|
"step": 2720 |
|
}, |
|
{ |
|
"epoch": 77.99, |
|
"learning_rate": 8.945244956772335e-05, |
|
"loss": 0.3785, |
|
"step": 2730 |
|
}, |
|
{ |
|
"epoch": 77.99, |
|
"eval_loss": 0.7071495056152344, |
|
"eval_runtime": 8.8836, |
|
"eval_samples_per_second": 7.092, |
|
"eval_steps_per_second": 1.801, |
|
"eval_wer": 0.3757908055672712, |
|
"step": 2730 |
|
}, |
|
{ |
|
"epoch": 78.28, |
|
"learning_rate": 8.829971181556195e-05, |
|
"loss": 0.3751, |
|
"step": 2740 |
|
}, |
|
{ |
|
"epoch": 78.57, |
|
"learning_rate": 8.714697406340057e-05, |
|
"loss": 0.3622, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 78.85, |
|
"learning_rate": 8.599423631123919e-05, |
|
"loss": 0.3466, |
|
"step": 2760 |
|
}, |
|
{ |
|
"epoch": 78.99, |
|
"eval_loss": 0.7361500263214111, |
|
"eval_runtime": 9.0867, |
|
"eval_samples_per_second": 6.933, |
|
"eval_steps_per_second": 1.761, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2765 |
|
}, |
|
{ |
|
"epoch": 79.14, |
|
"learning_rate": 8.484149855907782e-05, |
|
"loss": 0.3678, |
|
"step": 2770 |
|
}, |
|
{ |
|
"epoch": 79.43, |
|
"learning_rate": 8.368876080691644e-05, |
|
"loss": 0.3829, |
|
"step": 2780 |
|
}, |
|
{ |
|
"epoch": 79.71, |
|
"learning_rate": 8.253602305475506e-05, |
|
"loss": 0.4725, |
|
"step": 2790 |
|
}, |
|
{ |
|
"epoch": 79.99, |
|
"learning_rate": 8.138328530259366e-05, |
|
"loss": 0.3505, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 79.99, |
|
"eval_loss": 0.6964561343193054, |
|
"eval_runtime": 8.8201, |
|
"eval_samples_per_second": 7.143, |
|
"eval_steps_per_second": 1.814, |
|
"eval_wer": 0.3800084352593842, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 80.28, |
|
"learning_rate": 8.023054755043228e-05, |
|
"loss": 0.3439, |
|
"step": 2810 |
|
}, |
|
{ |
|
"epoch": 80.57, |
|
"learning_rate": 7.90778097982709e-05, |
|
"loss": 0.3645, |
|
"step": 2820 |
|
}, |
|
{ |
|
"epoch": 80.85, |
|
"learning_rate": 7.792507204610951e-05, |
|
"loss": 0.4003, |
|
"step": 2830 |
|
}, |
|
{ |
|
"epoch": 80.99, |
|
"eval_loss": 0.752064049243927, |
|
"eval_runtime": 8.7801, |
|
"eval_samples_per_second": 7.175, |
|
"eval_steps_per_second": 1.822, |
|
"eval_wer": 0.3766343315056938, |
|
"step": 2835 |
|
}, |
|
{ |
|
"epoch": 81.14, |
|
"learning_rate": 7.677233429394813e-05, |
|
"loss": 0.4152, |
|
"step": 2840 |
|
}, |
|
{ |
|
"epoch": 81.43, |
|
"learning_rate": 7.561959654178675e-05, |
|
"loss": 0.3652, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 81.71, |
|
"learning_rate": 7.446685878962537e-05, |
|
"loss": 0.3254, |
|
"step": 2860 |
|
}, |
|
{ |
|
"epoch": 81.99, |
|
"learning_rate": 7.331412103746398e-05, |
|
"loss": 0.3723, |
|
"step": 2870 |
|
}, |
|
{ |
|
"epoch": 81.99, |
|
"eval_loss": 0.7616933584213257, |
|
"eval_runtime": 8.8771, |
|
"eval_samples_per_second": 7.097, |
|
"eval_steps_per_second": 1.802, |
|
"eval_wer": 0.3749472796288486, |
|
"step": 2870 |
|
}, |
|
{ |
|
"epoch": 82.28, |
|
"learning_rate": 7.21613832853026e-05, |
|
"loss": 0.3791, |
|
"step": 2880 |
|
}, |
|
{ |
|
"epoch": 82.57, |
|
"learning_rate": 7.100864553314122e-05, |
|
"loss": 0.3127, |
|
"step": 2890 |
|
}, |
|
{ |
|
"epoch": 82.85, |
|
"learning_rate": 6.985590778097984e-05, |
|
"loss": 0.4029, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 82.99, |
|
"eval_loss": 0.7659004926681519, |
|
"eval_runtime": 8.8596, |
|
"eval_samples_per_second": 7.111, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.38127372416701816, |
|
"step": 2905 |
|
}, |
|
{ |
|
"epoch": 83.14, |
|
"learning_rate": 6.870317002881844e-05, |
|
"loss": 0.3742, |
|
"step": 2910 |
|
}, |
|
{ |
|
"epoch": 83.43, |
|
"learning_rate": 6.755043227665706e-05, |
|
"loss": 0.3072, |
|
"step": 2920 |
|
}, |
|
{ |
|
"epoch": 83.71, |
|
"learning_rate": 6.639769452449568e-05, |
|
"loss": 0.3836, |
|
"step": 2930 |
|
}, |
|
{ |
|
"epoch": 83.99, |
|
"learning_rate": 6.52449567723343e-05, |
|
"loss": 0.3478, |
|
"step": 2940 |
|
}, |
|
{ |
|
"epoch": 83.99, |
|
"eval_loss": 0.707669198513031, |
|
"eval_runtime": 8.9849, |
|
"eval_samples_per_second": 7.012, |
|
"eval_steps_per_second": 1.781, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2940 |
|
}, |
|
{ |
|
"epoch": 84.28, |
|
"learning_rate": 6.409221902017291e-05, |
|
"loss": 0.3467, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 84.57, |
|
"learning_rate": 6.293948126801153e-05, |
|
"loss": 0.324, |
|
"step": 2960 |
|
}, |
|
{ |
|
"epoch": 84.85, |
|
"learning_rate": 6.178674351585015e-05, |
|
"loss": 0.3363, |
|
"step": 2970 |
|
}, |
|
{ |
|
"epoch": 84.99, |
|
"eval_loss": 0.7333377599716187, |
|
"eval_runtime": 8.8448, |
|
"eval_samples_per_second": 7.123, |
|
"eval_steps_per_second": 1.809, |
|
"eval_wer": 0.3787431463517503, |
|
"step": 2975 |
|
}, |
|
{ |
|
"epoch": 85.14, |
|
"learning_rate": 6.0634005763688764e-05, |
|
"loss": 0.343, |
|
"step": 2980 |
|
}, |
|
{ |
|
"epoch": 85.43, |
|
"learning_rate": 5.948126801152738e-05, |
|
"loss": 0.3371, |
|
"step": 2990 |
|
}, |
|
{ |
|
"epoch": 85.71, |
|
"learning_rate": 5.8328530259366e-05, |
|
"loss": 0.3989, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 85.99, |
|
"learning_rate": 5.717579250720462e-05, |
|
"loss": 0.4228, |
|
"step": 3010 |
|
}, |
|
{ |
|
"epoch": 85.99, |
|
"eval_loss": 0.7196153998374939, |
|
"eval_runtime": 8.9305, |
|
"eval_samples_per_second": 7.054, |
|
"eval_steps_per_second": 1.792, |
|
"eval_wer": 0.3745255166596373, |
|
"step": 3010 |
|
}, |
|
{ |
|
"epoch": 86.28, |
|
"learning_rate": 5.602305475504323e-05, |
|
"loss": 0.3931, |
|
"step": 3020 |
|
}, |
|
{ |
|
"epoch": 86.57, |
|
"learning_rate": 5.4870317002881845e-05, |
|
"loss": 0.3359, |
|
"step": 3030 |
|
}, |
|
{ |
|
"epoch": 86.85, |
|
"learning_rate": 5.371757925072046e-05, |
|
"loss": 0.3823, |
|
"step": 3040 |
|
}, |
|
{ |
|
"epoch": 86.99, |
|
"eval_loss": 0.7194885015487671, |
|
"eval_runtime": 8.8773, |
|
"eval_samples_per_second": 7.097, |
|
"eval_steps_per_second": 1.802, |
|
"eval_wer": 0.37536904259805987, |
|
"step": 3045 |
|
}, |
|
{ |
|
"epoch": 87.14, |
|
"learning_rate": 5.256484149855908e-05, |
|
"loss": 0.3635, |
|
"step": 3050 |
|
}, |
|
{ |
|
"epoch": 87.43, |
|
"learning_rate": 5.14121037463977e-05, |
|
"loss": 0.3576, |
|
"step": 3060 |
|
}, |
|
{ |
|
"epoch": 87.71, |
|
"learning_rate": 5.0259365994236316e-05, |
|
"loss": 0.4654, |
|
"step": 3070 |
|
}, |
|
{ |
|
"epoch": 87.99, |
|
"learning_rate": 4.9106628242074933e-05, |
|
"loss": 0.3574, |
|
"step": 3080 |
|
}, |
|
{ |
|
"epoch": 87.99, |
|
"eval_loss": 0.713660478591919, |
|
"eval_runtime": 8.8845, |
|
"eval_samples_per_second": 7.091, |
|
"eval_steps_per_second": 1.801, |
|
"eval_wer": 0.3795866722901729, |
|
"step": 3080 |
|
}, |
|
{ |
|
"epoch": 88.28, |
|
"learning_rate": 4.7953890489913544e-05, |
|
"loss": 0.3677, |
|
"step": 3090 |
|
}, |
|
{ |
|
"epoch": 88.57, |
|
"learning_rate": 4.680115273775216e-05, |
|
"loss": 0.3065, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 88.85, |
|
"learning_rate": 4.564841498559078e-05, |
|
"loss": 0.3371, |
|
"step": 3110 |
|
}, |
|
{ |
|
"epoch": 88.99, |
|
"eval_loss": 0.7164372801780701, |
|
"eval_runtime": 8.865, |
|
"eval_samples_per_second": 7.107, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.3762125685364825, |
|
"step": 3115 |
|
}, |
|
{ |
|
"epoch": 89.14, |
|
"learning_rate": 4.44956772334294e-05, |
|
"loss": 0.6018, |
|
"step": 3120 |
|
}, |
|
{ |
|
"epoch": 89.43, |
|
"learning_rate": 4.3342939481268015e-05, |
|
"loss": 0.3404, |
|
"step": 3130 |
|
}, |
|
{ |
|
"epoch": 89.71, |
|
"learning_rate": 4.219020172910663e-05, |
|
"loss": 0.3659, |
|
"step": 3140 |
|
}, |
|
{ |
|
"epoch": 89.99, |
|
"learning_rate": 4.103746397694524e-05, |
|
"loss": 0.3548, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 89.99, |
|
"eval_loss": 0.7766362428665161, |
|
"eval_runtime": 8.874, |
|
"eval_samples_per_second": 7.099, |
|
"eval_steps_per_second": 1.803, |
|
"eval_wer": 0.37916490932096164, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 90.28, |
|
"learning_rate": 3.988472622478386e-05, |
|
"loss": 0.3393, |
|
"step": 3160 |
|
}, |
|
{ |
|
"epoch": 90.57, |
|
"learning_rate": 3.8731988472622485e-05, |
|
"loss": 0.2986, |
|
"step": 3170 |
|
}, |
|
{ |
|
"epoch": 90.85, |
|
"learning_rate": 3.7579250720461096e-05, |
|
"loss": 0.4042, |
|
"step": 3180 |
|
}, |
|
{ |
|
"epoch": 90.99, |
|
"eval_loss": 0.7587747573852539, |
|
"eval_runtime": 8.7904, |
|
"eval_samples_per_second": 7.167, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.3766343315056938, |
|
"step": 3185 |
|
}, |
|
{ |
|
"epoch": 91.14, |
|
"learning_rate": 3.6426512968299714e-05, |
|
"loss": 0.3223, |
|
"step": 3190 |
|
}, |
|
{ |
|
"epoch": 91.43, |
|
"learning_rate": 3.527377521613833e-05, |
|
"loss": 0.5178, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 91.71, |
|
"learning_rate": 3.412103746397695e-05, |
|
"loss": 0.3521, |
|
"step": 3210 |
|
}, |
|
{ |
|
"epoch": 91.99, |
|
"learning_rate": 3.296829971181557e-05, |
|
"loss": 0.3989, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 91.99, |
|
"eval_loss": 0.7310720086097717, |
|
"eval_runtime": 8.8614, |
|
"eval_samples_per_second": 7.109, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.3774778574441164, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 92.28, |
|
"learning_rate": 3.181556195965418e-05, |
|
"loss": 0.329, |
|
"step": 3230 |
|
}, |
|
{ |
|
"epoch": 92.57, |
|
"learning_rate": 3.0662824207492795e-05, |
|
"loss": 0.3594, |
|
"step": 3240 |
|
}, |
|
{ |
|
"epoch": 92.85, |
|
"learning_rate": 2.9510086455331416e-05, |
|
"loss": 0.3625, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 92.99, |
|
"eval_loss": 0.7475365400314331, |
|
"eval_runtime": 8.8406, |
|
"eval_samples_per_second": 7.126, |
|
"eval_steps_per_second": 1.81, |
|
"eval_wer": 0.3745255166596373, |
|
"step": 3255 |
|
}, |
|
{ |
|
"epoch": 93.14, |
|
"learning_rate": 2.8357348703170027e-05, |
|
"loss": 0.3361, |
|
"step": 3260 |
|
}, |
|
{ |
|
"epoch": 93.43, |
|
"learning_rate": 2.7204610951008648e-05, |
|
"loss": 0.3521, |
|
"step": 3270 |
|
}, |
|
{ |
|
"epoch": 93.71, |
|
"learning_rate": 2.6051873198847266e-05, |
|
"loss": 0.3261, |
|
"step": 3280 |
|
}, |
|
{ |
|
"epoch": 93.99, |
|
"learning_rate": 2.489913544668588e-05, |
|
"loss": 0.3036, |
|
"step": 3290 |
|
}, |
|
{ |
|
"epoch": 93.99, |
|
"eval_loss": 0.7138022184371948, |
|
"eval_runtime": 8.8395, |
|
"eval_samples_per_second": 7.127, |
|
"eval_steps_per_second": 1.81, |
|
"eval_wer": 0.37157317587515815, |
|
"step": 3290 |
|
}, |
|
{ |
|
"epoch": 94.28, |
|
"learning_rate": 2.3746397694524498e-05, |
|
"loss": 0.3412, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 94.57, |
|
"learning_rate": 2.2593659942363112e-05, |
|
"loss": 0.3451, |
|
"step": 3310 |
|
}, |
|
{ |
|
"epoch": 94.85, |
|
"learning_rate": 2.144092219020173e-05, |
|
"loss": 0.5157, |
|
"step": 3320 |
|
}, |
|
{ |
|
"epoch": 94.99, |
|
"eval_loss": 0.7245524525642395, |
|
"eval_runtime": 8.8889, |
|
"eval_samples_per_second": 7.087, |
|
"eval_steps_per_second": 1.8, |
|
"eval_wer": 0.3787431463517503, |
|
"step": 3325 |
|
}, |
|
{ |
|
"epoch": 95.14, |
|
"learning_rate": 2.0288184438040347e-05, |
|
"loss": 0.3089, |
|
"step": 3330 |
|
}, |
|
{ |
|
"epoch": 95.43, |
|
"learning_rate": 1.9135446685878965e-05, |
|
"loss": 0.3115, |
|
"step": 3340 |
|
}, |
|
{ |
|
"epoch": 95.71, |
|
"learning_rate": 1.798270893371758e-05, |
|
"loss": 0.3742, |
|
"step": 3350 |
|
}, |
|
{ |
|
"epoch": 95.99, |
|
"learning_rate": 1.6829971181556197e-05, |
|
"loss": 0.4072, |
|
"step": 3360 |
|
}, |
|
{ |
|
"epoch": 95.99, |
|
"eval_loss": 0.7321768999099731, |
|
"eval_runtime": 8.9645, |
|
"eval_samples_per_second": 7.028, |
|
"eval_steps_per_second": 1.785, |
|
"eval_wer": 0.3762125685364825, |
|
"step": 3360 |
|
}, |
|
{ |
|
"epoch": 96.28, |
|
"learning_rate": 1.5677233429394814e-05, |
|
"loss": 0.3496, |
|
"step": 3370 |
|
}, |
|
{ |
|
"epoch": 96.57, |
|
"learning_rate": 1.452449567723343e-05, |
|
"loss": 0.4089, |
|
"step": 3380 |
|
}, |
|
{ |
|
"epoch": 96.85, |
|
"learning_rate": 1.3371757925072045e-05, |
|
"loss": 0.3406, |
|
"step": 3390 |
|
}, |
|
{ |
|
"epoch": 96.99, |
|
"eval_loss": 0.7133801579475403, |
|
"eval_runtime": 8.9109, |
|
"eval_samples_per_second": 7.07, |
|
"eval_steps_per_second": 1.796, |
|
"eval_wer": 0.3770560944749051, |
|
"step": 3395 |
|
}, |
|
{ |
|
"epoch": 97.14, |
|
"learning_rate": 1.2219020172910664e-05, |
|
"loss": 0.3305, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 97.43, |
|
"learning_rate": 1.106628242074928e-05, |
|
"loss": 0.3521, |
|
"step": 3410 |
|
}, |
|
{ |
|
"epoch": 97.71, |
|
"learning_rate": 9.913544668587897e-06, |
|
"loss": 0.3405, |
|
"step": 3420 |
|
}, |
|
{ |
|
"epoch": 97.99, |
|
"learning_rate": 8.760806916426513e-06, |
|
"loss": 0.2987, |
|
"step": 3430 |
|
}, |
|
{ |
|
"epoch": 97.99, |
|
"eval_loss": 0.69514000415802, |
|
"eval_runtime": 8.9709, |
|
"eval_samples_per_second": 7.023, |
|
"eval_steps_per_second": 1.784, |
|
"eval_wer": 0.37536904259805987, |
|
"step": 3430 |
|
}, |
|
{ |
|
"epoch": 98.28, |
|
"learning_rate": 7.60806916426513e-06, |
|
"loss": 0.3394, |
|
"step": 3440 |
|
}, |
|
{ |
|
"epoch": 98.57, |
|
"learning_rate": 6.455331412103746e-06, |
|
"loss": 0.3566, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 98.85, |
|
"learning_rate": 5.302593659942364e-06, |
|
"loss": 0.3355, |
|
"step": 3460 |
|
}, |
|
{ |
|
"epoch": 98.99, |
|
"eval_loss": 0.7004749178886414, |
|
"eval_runtime": 8.9802, |
|
"eval_samples_per_second": 7.015, |
|
"eval_steps_per_second": 1.782, |
|
"eval_wer": 0.3766343315056938, |
|
"step": 3465 |
|
}, |
|
{ |
|
"epoch": 99.14, |
|
"learning_rate": 4.14985590778098e-06, |
|
"loss": 0.3434, |
|
"step": 3470 |
|
}, |
|
{ |
|
"epoch": 99.43, |
|
"learning_rate": 2.9971181556195965e-06, |
|
"loss": 0.3309, |
|
"step": 3480 |
|
}, |
|
{ |
|
"epoch": 99.71, |
|
"learning_rate": 1.8443804034582135e-06, |
|
"loss": 0.369, |
|
"step": 3490 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"learning_rate": 6.9164265129683e-07, |
|
"loss": 0.341, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"eval_loss": 0.6989557147026062, |
|
"eval_runtime": 8.9073, |
|
"eval_samples_per_second": 7.073, |
|
"eval_steps_per_second": 1.796, |
|
"eval_wer": 0.3770560944749051, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"step": 3500, |
|
"total_flos": 5.739635222597828e+18, |
|
"train_loss": 0.9362126126289367, |
|
"train_runtime": 11890.6212, |
|
"train_samples_per_second": 4.735, |
|
"train_steps_per_second": 0.294 |
|
} |
|
], |
|
"max_steps": 3500, |
|
"num_train_epochs": 100, |
|
"total_flos": 5.739635222597828e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|