|
{ |
|
"best_metric": 0.6822821497917175, |
|
"best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v6/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": 55.7616, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.0001866666666666667, |
|
"loss": 58.6167, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.85, |
|
"learning_rate": 0.00032, |
|
"loss": 17.274, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.99, |
|
"eval_loss": 2.9044528007507324, |
|
"eval_runtime": 8.8761, |
|
"eval_samples_per_second": 7.098, |
|
"eval_steps_per_second": 1.803, |
|
"eval_wer": 1.0, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 1.14, |
|
"learning_rate": 0.00039953890489913545, |
|
"loss": 2.8981, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.43, |
|
"learning_rate": 0.0003983861671469741, |
|
"loss": 2.0826, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 1.71, |
|
"learning_rate": 0.0003972334293948127, |
|
"loss": 2.1571, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"learning_rate": 0.00039608069164265134, |
|
"loss": 1.8443, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 1.99, |
|
"eval_loss": 3.506455659866333, |
|
"eval_runtime": 9.3088, |
|
"eval_samples_per_second": 6.768, |
|
"eval_steps_per_second": 1.719, |
|
"eval_wer": 1.0, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 2.28, |
|
"learning_rate": 0.0003949279538904899, |
|
"loss": 1.8021, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 2.57, |
|
"learning_rate": 0.00039377521613832855, |
|
"loss": 1.6728, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 2.85, |
|
"learning_rate": 0.0003926224783861672, |
|
"loss": 1.709, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"eval_loss": 2.007244348526001, |
|
"eval_runtime": 9.0409, |
|
"eval_samples_per_second": 6.968, |
|
"eval_steps_per_second": 1.77, |
|
"eval_wer": 1.0, |
|
"step": 105 |
|
}, |
|
{ |
|
"epoch": 3.14, |
|
"learning_rate": 0.00039146974063400576, |
|
"loss": 1.6017, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 3.43, |
|
"learning_rate": 0.0003903170028818444, |
|
"loss": 1.6211, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 3.71, |
|
"learning_rate": 0.000389164265129683, |
|
"loss": 1.59, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"learning_rate": 0.00038801152737752165, |
|
"loss": 1.4981, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 3.99, |
|
"eval_loss": 1.9509531259536743, |
|
"eval_runtime": 8.975, |
|
"eval_samples_per_second": 7.019, |
|
"eval_steps_per_second": 1.783, |
|
"eval_wer": 0.9687895402783636, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 4.28, |
|
"learning_rate": 0.00038685878962536023, |
|
"loss": 1.4853, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 4.57, |
|
"learning_rate": 0.00038570605187319886, |
|
"loss": 1.3834, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 4.85, |
|
"learning_rate": 0.0003845533141210375, |
|
"loss": 1.2977, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 4.99, |
|
"eval_loss": 1.886266827583313, |
|
"eval_runtime": 8.7241, |
|
"eval_samples_per_second": 7.221, |
|
"eval_steps_per_second": 1.834, |
|
"eval_wer": 0.5533530156052299, |
|
"step": 175 |
|
}, |
|
{ |
|
"epoch": 5.14, |
|
"learning_rate": 0.0003834005763688761, |
|
"loss": 1.3438, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 5.43, |
|
"learning_rate": 0.0003822478386167147, |
|
"loss": 1.1034, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 5.71, |
|
"learning_rate": 0.00038109510086455333, |
|
"loss": 1.2726, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 5.99, |
|
"learning_rate": 0.00037994236311239196, |
|
"loss": 1.1257, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 5.99, |
|
"eval_loss": 1.9137098789215088, |
|
"eval_runtime": 8.7964, |
|
"eval_samples_per_second": 7.162, |
|
"eval_steps_per_second": 1.819, |
|
"eval_wer": 0.48334036271615355, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 6.28, |
|
"learning_rate": 0.00037878962536023054, |
|
"loss": 1.1016, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 6.57, |
|
"learning_rate": 0.00037763688760806917, |
|
"loss": 1.0337, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 6.85, |
|
"learning_rate": 0.0003764841498559078, |
|
"loss": 1.1218, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 6.99, |
|
"eval_loss": 1.970656156539917, |
|
"eval_runtime": 8.6655, |
|
"eval_samples_per_second": 7.27, |
|
"eval_steps_per_second": 1.846, |
|
"eval_wer": 0.4959932517924926, |
|
"step": 245 |
|
}, |
|
{ |
|
"epoch": 7.14, |
|
"learning_rate": 0.00037533141210374643, |
|
"loss": 1.0968, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 7.43, |
|
"learning_rate": 0.000374178674351585, |
|
"loss": 0.9841, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 7.71, |
|
"learning_rate": 0.00037302593659942364, |
|
"loss": 0.9943, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 7.99, |
|
"learning_rate": 0.0003718731988472623, |
|
"loss": 0.8878, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 7.99, |
|
"eval_loss": 1.4178673028945923, |
|
"eval_runtime": 8.6797, |
|
"eval_samples_per_second": 7.258, |
|
"eval_steps_per_second": 1.843, |
|
"eval_wer": 0.47743568114719526, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"learning_rate": 0.0003707204610951009, |
|
"loss": 0.9877, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 8.57, |
|
"learning_rate": 0.0003695677233429395, |
|
"loss": 0.9138, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 8.85, |
|
"learning_rate": 0.0003684149855907781, |
|
"loss": 0.8562, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 8.99, |
|
"eval_loss": 1.5276190042495728, |
|
"eval_runtime": 8.6932, |
|
"eval_samples_per_second": 7.247, |
|
"eval_steps_per_second": 1.841, |
|
"eval_wer": 0.46351750316322227, |
|
"step": 315 |
|
}, |
|
{ |
|
"epoch": 9.14, |
|
"learning_rate": 0.00036726224783861674, |
|
"loss": 0.871, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 9.43, |
|
"learning_rate": 0.0003661095100864553, |
|
"loss": 0.9228, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 9.71, |
|
"learning_rate": 0.00036495677233429395, |
|
"loss": 0.8503, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 9.99, |
|
"learning_rate": 0.0003638040345821326, |
|
"loss": 1.5769, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 9.99, |
|
"eval_loss": 1.1269807815551758, |
|
"eval_runtime": 8.7329, |
|
"eval_samples_per_second": 7.214, |
|
"eval_steps_per_second": 1.832, |
|
"eval_wer": 0.45086461408688316, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 10.28, |
|
"learning_rate": 0.0003626512968299712, |
|
"loss": 0.9824, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 10.57, |
|
"learning_rate": 0.0003614985590778098, |
|
"loss": 0.7782, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 10.85, |
|
"learning_rate": 0.0003603458213256484, |
|
"loss": 0.796, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 10.99, |
|
"eval_loss": 1.2705838680267334, |
|
"eval_runtime": 8.707, |
|
"eval_samples_per_second": 7.236, |
|
"eval_steps_per_second": 1.838, |
|
"eval_wer": 0.4495993251792493, |
|
"step": 385 |
|
}, |
|
{ |
|
"epoch": 11.14, |
|
"learning_rate": 0.00035919308357348705, |
|
"loss": 1.0476, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 11.43, |
|
"learning_rate": 0.0003580403458213257, |
|
"loss": 0.7974, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 11.71, |
|
"learning_rate": 0.00035688760806916426, |
|
"loss": 0.8962, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"learning_rate": 0.0003557348703170029, |
|
"loss": 0.8776, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 11.99, |
|
"eval_loss": 1.237196683883667, |
|
"eval_runtime": 8.6827, |
|
"eval_samples_per_second": 7.256, |
|
"eval_steps_per_second": 1.843, |
|
"eval_wer": 0.44706874736398144, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 12.28, |
|
"learning_rate": 0.0003545821325648415, |
|
"loss": 0.8403, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 12.57, |
|
"learning_rate": 0.0003534293948126801, |
|
"loss": 0.7371, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 12.85, |
|
"learning_rate": 0.00035227665706051873, |
|
"loss": 0.7417, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 12.99, |
|
"eval_loss": 1.2826257944107056, |
|
"eval_runtime": 8.7629, |
|
"eval_samples_per_second": 7.189, |
|
"eval_steps_per_second": 1.826, |
|
"eval_wer": 0.43821172501054406, |
|
"step": 455 |
|
}, |
|
{ |
|
"epoch": 13.14, |
|
"learning_rate": 0.00035112391930835736, |
|
"loss": 0.8591, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 13.43, |
|
"learning_rate": 0.000349971181556196, |
|
"loss": 0.711, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 13.71, |
|
"learning_rate": 0.00034881844380403457, |
|
"loss": 0.849, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 13.99, |
|
"learning_rate": 0.00034766570605187326, |
|
"loss": 0.8273, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 13.99, |
|
"eval_loss": 1.2424755096435547, |
|
"eval_runtime": 8.7863, |
|
"eval_samples_per_second": 7.17, |
|
"eval_steps_per_second": 1.821, |
|
"eval_wer": 0.4542387178405736, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 14.28, |
|
"learning_rate": 0.00034651296829971183, |
|
"loss": 0.8087, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 14.57, |
|
"learning_rate": 0.00034536023054755047, |
|
"loss": 0.6899, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 14.85, |
|
"learning_rate": 0.00034420749279538904, |
|
"loss": 0.7164, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 14.99, |
|
"eval_loss": 1.141480803489685, |
|
"eval_runtime": 8.7134, |
|
"eval_samples_per_second": 7.23, |
|
"eval_steps_per_second": 1.836, |
|
"eval_wer": 0.4192323913960354, |
|
"step": 525 |
|
}, |
|
{ |
|
"epoch": 15.14, |
|
"learning_rate": 0.0003430547550432277, |
|
"loss": 0.7969, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 15.43, |
|
"learning_rate": 0.0003419020172910663, |
|
"loss": 0.7135, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 15.71, |
|
"learning_rate": 0.0003407492795389049, |
|
"loss": 0.6759, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 15.99, |
|
"learning_rate": 0.0003395965417867435, |
|
"loss": 0.7061, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 15.99, |
|
"eval_loss": 1.2315250635147095, |
|
"eval_runtime": 8.6974, |
|
"eval_samples_per_second": 7.244, |
|
"eval_steps_per_second": 1.84, |
|
"eval_wer": 0.4407423028258119, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 16.28, |
|
"learning_rate": 0.00033844380403458214, |
|
"loss": 0.7548, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 16.57, |
|
"learning_rate": 0.0003372910662824208, |
|
"loss": 0.6956, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 16.85, |
|
"learning_rate": 0.00033613832853025935, |
|
"loss": 0.6553, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 16.99, |
|
"eval_loss": 0.9983352422714233, |
|
"eval_runtime": 8.7086, |
|
"eval_samples_per_second": 7.234, |
|
"eval_steps_per_second": 1.837, |
|
"eval_wer": 0.41121889498102066, |
|
"step": 595 |
|
}, |
|
{ |
|
"epoch": 17.14, |
|
"learning_rate": 0.00033498559077809804, |
|
"loss": 0.6556, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 17.43, |
|
"learning_rate": 0.0003338328530259366, |
|
"loss": 0.6944, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 17.71, |
|
"learning_rate": 0.00033268011527377525, |
|
"loss": 0.7009, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 17.99, |
|
"learning_rate": 0.0003315273775216138, |
|
"loss": 0.7114, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 17.99, |
|
"eval_loss": 1.1509625911712646, |
|
"eval_runtime": 8.7423, |
|
"eval_samples_per_second": 7.206, |
|
"eval_steps_per_second": 1.83, |
|
"eval_wer": 0.43821172501054406, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 18.28, |
|
"learning_rate": 0.0003303746397694525, |
|
"loss": 0.6129, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 18.57, |
|
"learning_rate": 0.0003292219020172911, |
|
"loss": 0.6881, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 18.85, |
|
"learning_rate": 0.00032806916426512966, |
|
"loss": 0.6467, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 18.99, |
|
"eval_loss": 1.061179280281067, |
|
"eval_runtime": 8.7964, |
|
"eval_samples_per_second": 7.162, |
|
"eval_steps_per_second": 1.819, |
|
"eval_wer": 0.4048924504428511, |
|
"step": 665 |
|
}, |
|
{ |
|
"epoch": 19.14, |
|
"learning_rate": 0.00032691642651296835, |
|
"loss": 0.7667, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 19.43, |
|
"learning_rate": 0.0003257636887608069, |
|
"loss": 0.7554, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 19.71, |
|
"learning_rate": 0.00032461095100864556, |
|
"loss": 0.7296, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 19.99, |
|
"learning_rate": 0.00032345821325648413, |
|
"loss": 0.6035, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 19.99, |
|
"eval_loss": 1.035975456237793, |
|
"eval_runtime": 8.8552, |
|
"eval_samples_per_second": 7.115, |
|
"eval_steps_per_second": 1.807, |
|
"eval_wer": 0.41881062842682415, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 20.28, |
|
"learning_rate": 0.0003223054755043228, |
|
"loss": 0.7344, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 20.57, |
|
"learning_rate": 0.0003211527377521614, |
|
"loss": 0.6566, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 20.85, |
|
"learning_rate": 0.00032, |
|
"loss": 0.6058, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 20.99, |
|
"eval_loss": 1.0008372068405151, |
|
"eval_runtime": 8.8627, |
|
"eval_samples_per_second": 7.108, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.4137494727962885, |
|
"step": 735 |
|
}, |
|
{ |
|
"epoch": 21.14, |
|
"learning_rate": 0.0003188472622478386, |
|
"loss": 0.6781, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 21.43, |
|
"learning_rate": 0.0003176945244956773, |
|
"loss": 0.6773, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 21.71, |
|
"learning_rate": 0.00031654178674351587, |
|
"loss": 0.6446, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 21.99, |
|
"learning_rate": 0.00031538904899135444, |
|
"loss": 0.682, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 21.99, |
|
"eval_loss": 1.1947873830795288, |
|
"eval_runtime": 8.751, |
|
"eval_samples_per_second": 7.199, |
|
"eval_steps_per_second": 1.828, |
|
"eval_wer": 0.42091944327288067, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 22.28, |
|
"learning_rate": 0.00031423631123919313, |
|
"loss": 0.6224, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 22.57, |
|
"learning_rate": 0.0003130835734870317, |
|
"loss": 0.5607, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 22.85, |
|
"learning_rate": 0.00031193083573487034, |
|
"loss": 0.566, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 22.99, |
|
"eval_loss": 1.055545449256897, |
|
"eval_runtime": 8.7493, |
|
"eval_samples_per_second": 7.201, |
|
"eval_steps_per_second": 1.829, |
|
"eval_wer": 0.41332770982707717, |
|
"step": 805 |
|
}, |
|
{ |
|
"epoch": 23.14, |
|
"learning_rate": 0.0003107780979827089, |
|
"loss": 0.6624, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 23.43, |
|
"learning_rate": 0.0003096253602305476, |
|
"loss": 0.6408, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 23.71, |
|
"learning_rate": 0.0003084726224783862, |
|
"loss": 0.6352, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 23.99, |
|
"learning_rate": 0.0003073198847262248, |
|
"loss": 0.5952, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 23.99, |
|
"eval_loss": 0.8614811301231384, |
|
"eval_runtime": 8.7821, |
|
"eval_samples_per_second": 7.174, |
|
"eval_steps_per_second": 1.822, |
|
"eval_wer": 0.40953184310417545, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 24.28, |
|
"learning_rate": 0.00030616714697406344, |
|
"loss": 0.7729, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 24.57, |
|
"learning_rate": 0.00030501440922190207, |
|
"loss": 0.5275, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 24.85, |
|
"learning_rate": 0.00030386167146974065, |
|
"loss": 0.5889, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 24.99, |
|
"eval_loss": 1.074023723602295, |
|
"eval_runtime": 8.7596, |
|
"eval_samples_per_second": 7.192, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.4301982285955293, |
|
"step": 875 |
|
}, |
|
{ |
|
"epoch": 25.14, |
|
"learning_rate": 0.0003027089337175792, |
|
"loss": 0.5998, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 25.43, |
|
"learning_rate": 0.0003015561959654179, |
|
"loss": 0.5667, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 25.71, |
|
"learning_rate": 0.0003004034582132565, |
|
"loss": 0.5979, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 25.99, |
|
"learning_rate": 0.0002992507204610951, |
|
"loss": 0.5954, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 25.99, |
|
"eval_loss": 1.14651358127594, |
|
"eval_runtime": 8.78, |
|
"eval_samples_per_second": 7.175, |
|
"eval_steps_per_second": 1.822, |
|
"eval_wer": 0.41670181358076763, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 26.28, |
|
"learning_rate": 0.00029809798270893375, |
|
"loss": 0.6101, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 26.57, |
|
"learning_rate": 0.0002969452449567724, |
|
"loss": 0.5444, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 26.85, |
|
"learning_rate": 0.00029579250720461096, |
|
"loss": 0.5615, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 26.99, |
|
"eval_loss": 0.8980141282081604, |
|
"eval_runtime": 8.7788, |
|
"eval_samples_per_second": 7.176, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.40742302825811894, |
|
"step": 945 |
|
}, |
|
{ |
|
"epoch": 27.14, |
|
"learning_rate": 0.0002946397694524496, |
|
"loss": 0.5682, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 27.43, |
|
"learning_rate": 0.0002934870317002882, |
|
"loss": 0.5287, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 27.71, |
|
"learning_rate": 0.00029233429394812685, |
|
"loss": 0.5751, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 27.99, |
|
"learning_rate": 0.0002911815561959654, |
|
"loss": 0.5385, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 27.99, |
|
"eval_loss": 0.8442803621292114, |
|
"eval_runtime": 8.8658, |
|
"eval_samples_per_second": 7.106, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.40615773935048505, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 28.28, |
|
"learning_rate": 0.000290028818443804, |
|
"loss": 0.5477, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 28.57, |
|
"learning_rate": 0.0002888760806916427, |
|
"loss": 0.553, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 28.85, |
|
"learning_rate": 0.00028772334293948127, |
|
"loss": 0.5097, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 28.99, |
|
"eval_loss": 1.1463680267333984, |
|
"eval_runtime": 8.9294, |
|
"eval_samples_per_second": 7.055, |
|
"eval_steps_per_second": 1.792, |
|
"eval_wer": 0.4048924504428511, |
|
"step": 1015 |
|
}, |
|
{ |
|
"epoch": 29.14, |
|
"learning_rate": 0.0002865706051873199, |
|
"loss": 0.5598, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 29.43, |
|
"learning_rate": 0.00028541786743515853, |
|
"loss": 0.529, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 29.71, |
|
"learning_rate": 0.00028426512968299716, |
|
"loss": 0.6047, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 29.99, |
|
"learning_rate": 0.00028311239193083574, |
|
"loss": 0.5224, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 29.99, |
|
"eval_loss": 1.0213007926940918, |
|
"eval_runtime": 8.848, |
|
"eval_samples_per_second": 7.12, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.40025305778152676, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 30.28, |
|
"learning_rate": 0.00028195965417867437, |
|
"loss": 0.6028, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 30.57, |
|
"learning_rate": 0.000280806916426513, |
|
"loss": 0.529, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 30.85, |
|
"learning_rate": 0.00027965417867435163, |
|
"loss": 0.5226, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 30.99, |
|
"eval_loss": 0.8601366281509399, |
|
"eval_runtime": 8.7916, |
|
"eval_samples_per_second": 7.166, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.40911008013496414, |
|
"step": 1085 |
|
}, |
|
{ |
|
"epoch": 31.14, |
|
"learning_rate": 0.0002785014409221902, |
|
"loss": 0.6275, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 31.43, |
|
"learning_rate": 0.00027734870317002884, |
|
"loss": 0.6156, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 31.71, |
|
"learning_rate": 0.00027619596541786747, |
|
"loss": 0.6076, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 31.99, |
|
"learning_rate": 0.00027504322766570605, |
|
"loss": 0.5303, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 31.99, |
|
"eval_loss": 1.0191397666931152, |
|
"eval_runtime": 8.773, |
|
"eval_samples_per_second": 7.181, |
|
"eval_steps_per_second": 1.824, |
|
"eval_wer": 0.39856600590468155, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 32.28, |
|
"learning_rate": 0.0002738904899135447, |
|
"loss": 0.4981, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 32.57, |
|
"learning_rate": 0.0002727377521613833, |
|
"loss": 0.5606, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 32.85, |
|
"learning_rate": 0.00027158501440922194, |
|
"loss": 0.6457, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 32.99, |
|
"eval_loss": 1.244334101676941, |
|
"eval_runtime": 8.8978, |
|
"eval_samples_per_second": 7.08, |
|
"eval_steps_per_second": 1.798, |
|
"eval_wer": 0.4306199915647406, |
|
"step": 1155 |
|
}, |
|
{ |
|
"epoch": 33.14, |
|
"learning_rate": 0.0002704322766570605, |
|
"loss": 0.5721, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 33.43, |
|
"learning_rate": 0.00026927953890489915, |
|
"loss": 0.516, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 33.71, |
|
"learning_rate": 0.0002681268011527378, |
|
"loss": 0.6213, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 33.99, |
|
"learning_rate": 0.0002669740634005764, |
|
"loss": 0.5305, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 33.99, |
|
"eval_loss": 0.9871523380279541, |
|
"eval_runtime": 8.8265, |
|
"eval_samples_per_second": 7.138, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.4171235765499789, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 34.28, |
|
"learning_rate": 0.000265821325648415, |
|
"loss": 0.5362, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 34.57, |
|
"learning_rate": 0.0002646685878962536, |
|
"loss": 0.5388, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 34.85, |
|
"learning_rate": 0.00026351585014409225, |
|
"loss": 0.5179, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 34.99, |
|
"eval_loss": 1.0433094501495361, |
|
"eval_runtime": 8.8277, |
|
"eval_samples_per_second": 7.137, |
|
"eval_steps_per_second": 1.812, |
|
"eval_wer": 0.39350485027414595, |
|
"step": 1225 |
|
}, |
|
{ |
|
"epoch": 35.14, |
|
"learning_rate": 0.0002623631123919308, |
|
"loss": 0.5198, |
|
"step": 1230 |
|
}, |
|
{ |
|
"epoch": 35.43, |
|
"learning_rate": 0.00026121037463976946, |
|
"loss": 0.6019, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 35.71, |
|
"learning_rate": 0.0002600576368876081, |
|
"loss": 0.5521, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 35.99, |
|
"learning_rate": 0.0002589048991354467, |
|
"loss": 0.471, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 35.99, |
|
"eval_loss": 1.0010788440704346, |
|
"eval_runtime": 8.7661, |
|
"eval_samples_per_second": 7.187, |
|
"eval_steps_per_second": 1.825, |
|
"eval_wer": 0.40742302825811894, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 36.28, |
|
"learning_rate": 0.0002577521613832853, |
|
"loss": 0.5277, |
|
"step": 1270 |
|
}, |
|
{ |
|
"epoch": 36.57, |
|
"learning_rate": 0.00025659942363112393, |
|
"loss": 0.4516, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 36.85, |
|
"learning_rate": 0.00025544668587896256, |
|
"loss": 0.473, |
|
"step": 1290 |
|
}, |
|
{ |
|
"epoch": 36.99, |
|
"eval_loss": 0.8886780738830566, |
|
"eval_runtime": 8.8639, |
|
"eval_samples_per_second": 7.107, |
|
"eval_steps_per_second": 1.805, |
|
"eval_wer": 0.3901307465204555, |
|
"step": 1295 |
|
}, |
|
{ |
|
"epoch": 37.14, |
|
"learning_rate": 0.0002542939481268012, |
|
"loss": 0.6135, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 37.43, |
|
"learning_rate": 0.00025314121037463977, |
|
"loss": 0.5081, |
|
"step": 1310 |
|
}, |
|
{ |
|
"epoch": 37.71, |
|
"learning_rate": 0.0002519884726224784, |
|
"loss": 0.5413, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 37.99, |
|
"learning_rate": 0.00025083573487031703, |
|
"loss": 0.5465, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 37.99, |
|
"eval_loss": 0.8611962199211121, |
|
"eval_runtime": 8.9315, |
|
"eval_samples_per_second": 7.054, |
|
"eval_steps_per_second": 1.791, |
|
"eval_wer": 0.3897089835512442, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 38.28, |
|
"learning_rate": 0.0002496829971181556, |
|
"loss": 0.6413, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 38.57, |
|
"learning_rate": 0.00024853025936599424, |
|
"loss": 0.4731, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 38.85, |
|
"learning_rate": 0.00024737752161383287, |
|
"loss": 0.4584, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 38.99, |
|
"eval_loss": 0.9581005573272705, |
|
"eval_runtime": 8.827, |
|
"eval_samples_per_second": 7.137, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.4070012652889076, |
|
"step": 1365 |
|
}, |
|
{ |
|
"epoch": 39.14, |
|
"learning_rate": 0.0002462247838616715, |
|
"loss": 0.496, |
|
"step": 1370 |
|
}, |
|
{ |
|
"epoch": 39.43, |
|
"learning_rate": 0.0002450720461095101, |
|
"loss": 0.4822, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 39.71, |
|
"learning_rate": 0.00024391930835734874, |
|
"loss": 0.495, |
|
"step": 1390 |
|
}, |
|
{ |
|
"epoch": 39.99, |
|
"learning_rate": 0.00024276657060518734, |
|
"loss": 0.565, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 39.99, |
|
"eval_loss": 1.0734798908233643, |
|
"eval_runtime": 8.8601, |
|
"eval_samples_per_second": 7.111, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.40826655419654156, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 40.28, |
|
"learning_rate": 0.00024161383285302597, |
|
"loss": 0.5564, |
|
"step": 1410 |
|
}, |
|
{ |
|
"epoch": 40.57, |
|
"learning_rate": 0.00024046109510086458, |
|
"loss": 0.4628, |
|
"step": 1420 |
|
}, |
|
{ |
|
"epoch": 40.85, |
|
"learning_rate": 0.00023930835734870315, |
|
"loss": 0.4916, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 40.99, |
|
"eval_loss": 0.8890073299407959, |
|
"eval_runtime": 8.7735, |
|
"eval_samples_per_second": 7.181, |
|
"eval_steps_per_second": 1.824, |
|
"eval_wer": 0.3905525094896668, |
|
"step": 1435 |
|
}, |
|
{ |
|
"epoch": 41.14, |
|
"learning_rate": 0.0002381556195965418, |
|
"loss": 0.4825, |
|
"step": 1440 |
|
}, |
|
{ |
|
"epoch": 41.43, |
|
"learning_rate": 0.0002370028818443804, |
|
"loss": 0.5301, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 41.71, |
|
"learning_rate": 0.00023585014409221905, |
|
"loss": 0.5599, |
|
"step": 1460 |
|
}, |
|
{ |
|
"epoch": 41.99, |
|
"learning_rate": 0.00023469740634005762, |
|
"loss": 0.4643, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 41.99, |
|
"eval_loss": 0.731747567653656, |
|
"eval_runtime": 8.7688, |
|
"eval_samples_per_second": 7.185, |
|
"eval_steps_per_second": 1.825, |
|
"eval_wer": 0.40404892450442853, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 42.28, |
|
"learning_rate": 0.00023354466858789628, |
|
"loss": 0.485, |
|
"step": 1480 |
|
}, |
|
{ |
|
"epoch": 42.57, |
|
"learning_rate": 0.00023239193083573489, |
|
"loss": 0.5415, |
|
"step": 1490 |
|
}, |
|
{ |
|
"epoch": 42.85, |
|
"learning_rate": 0.00023123919308357352, |
|
"loss": 0.4633, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 42.99, |
|
"eval_loss": 0.9383570551872253, |
|
"eval_runtime": 8.7533, |
|
"eval_samples_per_second": 7.197, |
|
"eval_steps_per_second": 1.828, |
|
"eval_wer": 0.4141712357654998, |
|
"step": 1505 |
|
}, |
|
{ |
|
"epoch": 43.14, |
|
"learning_rate": 0.00023008645533141212, |
|
"loss": 0.4621, |
|
"step": 1510 |
|
}, |
|
{ |
|
"epoch": 43.43, |
|
"learning_rate": 0.00022893371757925075, |
|
"loss": 0.4548, |
|
"step": 1520 |
|
}, |
|
{ |
|
"epoch": 43.71, |
|
"learning_rate": 0.00022778097982708936, |
|
"loss": 0.4375, |
|
"step": 1530 |
|
}, |
|
{ |
|
"epoch": 43.99, |
|
"learning_rate": 0.00022662824207492793, |
|
"loss": 0.4867, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 43.99, |
|
"eval_loss": 0.889893651008606, |
|
"eval_runtime": 8.7946, |
|
"eval_samples_per_second": 7.163, |
|
"eval_steps_per_second": 1.819, |
|
"eval_wer": 0.40742302825811894, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 44.28, |
|
"learning_rate": 0.0002254755043227666, |
|
"loss": 0.4903, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 44.57, |
|
"learning_rate": 0.00022432276657060517, |
|
"loss": 0.5027, |
|
"step": 1560 |
|
}, |
|
{ |
|
"epoch": 44.85, |
|
"learning_rate": 0.00022317002881844383, |
|
"loss": 0.4892, |
|
"step": 1570 |
|
}, |
|
{ |
|
"epoch": 44.99, |
|
"eval_loss": 0.8419014811515808, |
|
"eval_runtime": 8.7769, |
|
"eval_samples_per_second": 7.178, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.4053142134120624, |
|
"step": 1575 |
|
}, |
|
{ |
|
"epoch": 45.14, |
|
"learning_rate": 0.00022201729106628243, |
|
"loss": 0.4887, |
|
"step": 1580 |
|
}, |
|
{ |
|
"epoch": 45.43, |
|
"learning_rate": 0.00022086455331412106, |
|
"loss": 0.4164, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 45.71, |
|
"learning_rate": 0.00021971181556195967, |
|
"loss": 0.5044, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 45.99, |
|
"learning_rate": 0.0002185590778097983, |
|
"loss": 0.4338, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 45.99, |
|
"eval_loss": 0.8296654224395752, |
|
"eval_runtime": 8.777, |
|
"eval_samples_per_second": 7.178, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.4023618726275833, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 46.28, |
|
"learning_rate": 0.0002174063400576369, |
|
"loss": 0.4882, |
|
"step": 1620 |
|
}, |
|
{ |
|
"epoch": 46.57, |
|
"learning_rate": 0.00021625360230547553, |
|
"loss": 0.4222, |
|
"step": 1630 |
|
}, |
|
{ |
|
"epoch": 46.85, |
|
"learning_rate": 0.00021510086455331414, |
|
"loss": 0.4038, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 46.99, |
|
"eval_loss": 0.9688724875450134, |
|
"eval_runtime": 8.956, |
|
"eval_samples_per_second": 7.034, |
|
"eval_steps_per_second": 1.787, |
|
"eval_wer": 0.38253901307465205, |
|
"step": 1645 |
|
}, |
|
{ |
|
"epoch": 47.14, |
|
"learning_rate": 0.0002139481268011527, |
|
"loss": 0.5038, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 47.43, |
|
"learning_rate": 0.00021279538904899137, |
|
"loss": 0.4728, |
|
"step": 1660 |
|
}, |
|
{ |
|
"epoch": 47.71, |
|
"learning_rate": 0.00021164265129682998, |
|
"loss": 0.6479, |
|
"step": 1670 |
|
}, |
|
{ |
|
"epoch": 47.99, |
|
"learning_rate": 0.0002104899135446686, |
|
"loss": 0.4519, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 47.99, |
|
"eval_loss": 0.8536186814308167, |
|
"eval_runtime": 8.8435, |
|
"eval_samples_per_second": 7.124, |
|
"eval_steps_per_second": 1.809, |
|
"eval_wer": 0.4053142134120624, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 48.28, |
|
"learning_rate": 0.0002093371757925072, |
|
"loss": 0.4692, |
|
"step": 1690 |
|
}, |
|
{ |
|
"epoch": 48.57, |
|
"learning_rate": 0.00020818443804034584, |
|
"loss": 0.4635, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 48.85, |
|
"learning_rate": 0.00020703170028818445, |
|
"loss": 0.4298, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 48.99, |
|
"eval_loss": 0.9737337827682495, |
|
"eval_runtime": 8.8175, |
|
"eval_samples_per_second": 7.145, |
|
"eval_steps_per_second": 1.815, |
|
"eval_wer": 0.3795866722901729, |
|
"step": 1715 |
|
}, |
|
{ |
|
"epoch": 49.14, |
|
"learning_rate": 0.00020587896253602308, |
|
"loss": 0.4744, |
|
"step": 1720 |
|
}, |
|
{ |
|
"epoch": 49.43, |
|
"learning_rate": 0.00020472622478386168, |
|
"loss": 0.4103, |
|
"step": 1730 |
|
}, |
|
{ |
|
"epoch": 49.71, |
|
"learning_rate": 0.0002035734870317003, |
|
"loss": 0.4445, |
|
"step": 1740 |
|
}, |
|
{ |
|
"epoch": 49.99, |
|
"learning_rate": 0.00020242074927953892, |
|
"loss": 0.4622, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 49.99, |
|
"eval_loss": 0.9053709506988525, |
|
"eval_runtime": 8.8571, |
|
"eval_samples_per_second": 7.113, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.40742302825811894, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 50.28, |
|
"learning_rate": 0.00020126801152737752, |
|
"loss": 0.499, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 50.57, |
|
"learning_rate": 0.00020011527377521615, |
|
"loss": 0.4354, |
|
"step": 1770 |
|
}, |
|
{ |
|
"epoch": 50.85, |
|
"learning_rate": 0.00019896253602305476, |
|
"loss": 0.4358, |
|
"step": 1780 |
|
}, |
|
{ |
|
"epoch": 50.99, |
|
"eval_loss": 0.7809118628501892, |
|
"eval_runtime": 8.7922, |
|
"eval_samples_per_second": 7.165, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.38127372416701816, |
|
"step": 1785 |
|
}, |
|
{ |
|
"epoch": 51.14, |
|
"learning_rate": 0.0001978097982708934, |
|
"loss": 0.4844, |
|
"step": 1790 |
|
}, |
|
{ |
|
"epoch": 51.43, |
|
"learning_rate": 0.00019665706051873202, |
|
"loss": 0.4778, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 51.71, |
|
"learning_rate": 0.00019550432276657062, |
|
"loss": 0.4856, |
|
"step": 1810 |
|
}, |
|
{ |
|
"epoch": 51.99, |
|
"learning_rate": 0.00019435158501440923, |
|
"loss": 0.4277, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 51.99, |
|
"eval_loss": 0.8463531732559204, |
|
"eval_runtime": 8.7437, |
|
"eval_samples_per_second": 7.205, |
|
"eval_steps_per_second": 1.83, |
|
"eval_wer": 0.392239561366512, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 52.28, |
|
"learning_rate": 0.00019319884726224783, |
|
"loss": 0.5665, |
|
"step": 1830 |
|
}, |
|
{ |
|
"epoch": 52.57, |
|
"learning_rate": 0.00019204610951008646, |
|
"loss": 0.3976, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 52.85, |
|
"learning_rate": 0.00019089337175792507, |
|
"loss": 0.4186, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 52.99, |
|
"eval_loss": 0.8106096386909485, |
|
"eval_runtime": 8.7521, |
|
"eval_samples_per_second": 7.198, |
|
"eval_steps_per_second": 1.828, |
|
"eval_wer": 0.39561366512020246, |
|
"step": 1855 |
|
}, |
|
{ |
|
"epoch": 53.14, |
|
"learning_rate": 0.0001897406340057637, |
|
"loss": 0.3918, |
|
"step": 1860 |
|
}, |
|
{ |
|
"epoch": 53.43, |
|
"learning_rate": 0.0001885878962536023, |
|
"loss": 0.4269, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 53.71, |
|
"learning_rate": 0.00018743515850144093, |
|
"loss": 0.457, |
|
"step": 1880 |
|
}, |
|
{ |
|
"epoch": 53.99, |
|
"learning_rate": 0.00018628242074927956, |
|
"loss": 0.413, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 53.99, |
|
"eval_loss": 0.9219353199005127, |
|
"eval_runtime": 8.7499, |
|
"eval_samples_per_second": 7.2, |
|
"eval_steps_per_second": 1.829, |
|
"eval_wer": 0.38127372416701816, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 54.28, |
|
"learning_rate": 0.00018512968299711817, |
|
"loss": 0.4764, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 54.57, |
|
"learning_rate": 0.0001839769452449568, |
|
"loss": 0.3499, |
|
"step": 1910 |
|
}, |
|
{ |
|
"epoch": 54.85, |
|
"learning_rate": 0.0001828242074927954, |
|
"loss": 0.4262, |
|
"step": 1920 |
|
}, |
|
{ |
|
"epoch": 54.99, |
|
"eval_loss": 0.9600013494491577, |
|
"eval_runtime": 8.8474, |
|
"eval_samples_per_second": 7.121, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.39898776887389287, |
|
"step": 1925 |
|
}, |
|
{ |
|
"epoch": 55.14, |
|
"learning_rate": 0.000181671469740634, |
|
"loss": 0.4998, |
|
"step": 1930 |
|
}, |
|
{ |
|
"epoch": 55.43, |
|
"learning_rate": 0.0001805187319884726, |
|
"loss": 0.4746, |
|
"step": 1940 |
|
}, |
|
{ |
|
"epoch": 55.71, |
|
"learning_rate": 0.00017936599423631124, |
|
"loss": 0.5197, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 55.99, |
|
"learning_rate": 0.00017821325648414985, |
|
"loss": 0.4542, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 55.99, |
|
"eval_loss": 0.8444246649742126, |
|
"eval_runtime": 8.8584, |
|
"eval_samples_per_second": 7.112, |
|
"eval_steps_per_second": 1.806, |
|
"eval_wer": 0.40573597638127373, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 56.28, |
|
"learning_rate": 0.00017706051873198848, |
|
"loss": 0.5553, |
|
"step": 1970 |
|
}, |
|
{ |
|
"epoch": 56.57, |
|
"learning_rate": 0.0001759077809798271, |
|
"loss": 0.431, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 56.85, |
|
"learning_rate": 0.0001747550432276657, |
|
"loss": 0.3966, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 56.99, |
|
"eval_loss": 0.7813870310783386, |
|
"eval_runtime": 8.9227, |
|
"eval_samples_per_second": 7.061, |
|
"eval_steps_per_second": 1.793, |
|
"eval_wer": 0.39139603542808943, |
|
"step": 1995 |
|
}, |
|
{ |
|
"epoch": 57.14, |
|
"learning_rate": 0.00017360230547550434, |
|
"loss": 0.4681, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 57.43, |
|
"learning_rate": 0.00017244956772334295, |
|
"loss": 0.4188, |
|
"step": 2010 |
|
}, |
|
{ |
|
"epoch": 57.71, |
|
"learning_rate": 0.00017129682997118158, |
|
"loss": 0.4498, |
|
"step": 2020 |
|
}, |
|
{ |
|
"epoch": 57.99, |
|
"learning_rate": 0.00017014409221902018, |
|
"loss": 0.444, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 57.99, |
|
"eval_loss": 0.8330946564674377, |
|
"eval_runtime": 9.0864, |
|
"eval_samples_per_second": 6.933, |
|
"eval_steps_per_second": 1.761, |
|
"eval_wer": 0.3770560944749051, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 58.28, |
|
"learning_rate": 0.0001689913544668588, |
|
"loss": 0.4516, |
|
"step": 2040 |
|
}, |
|
{ |
|
"epoch": 58.57, |
|
"learning_rate": 0.00016783861671469742, |
|
"loss": 0.6793, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 58.85, |
|
"learning_rate": 0.00016668587896253602, |
|
"loss": 0.4673, |
|
"step": 2060 |
|
}, |
|
{ |
|
"epoch": 58.99, |
|
"eval_loss": 0.7872037291526794, |
|
"eval_runtime": 8.8855, |
|
"eval_samples_per_second": 7.09, |
|
"eval_steps_per_second": 1.801, |
|
"eval_wer": 0.3960354280894138, |
|
"step": 2065 |
|
}, |
|
{ |
|
"epoch": 59.14, |
|
"learning_rate": 0.00016553314121037465, |
|
"loss": 0.4654, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 59.43, |
|
"learning_rate": 0.00016438040345821326, |
|
"loss": 0.8131, |
|
"step": 2080 |
|
}, |
|
{ |
|
"epoch": 59.71, |
|
"learning_rate": 0.0001632276657060519, |
|
"loss": 0.4366, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 59.99, |
|
"learning_rate": 0.0001620749279538905, |
|
"loss": 0.483, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 59.99, |
|
"eval_loss": 1.0760446786880493, |
|
"eval_runtime": 8.7291, |
|
"eval_samples_per_second": 7.217, |
|
"eval_steps_per_second": 1.833, |
|
"eval_wer": 0.4036271615352172, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 60.28, |
|
"learning_rate": 0.00016092219020172912, |
|
"loss": 0.3937, |
|
"step": 2110 |
|
}, |
|
{ |
|
"epoch": 60.57, |
|
"learning_rate": 0.00015976945244956773, |
|
"loss": 0.8094, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 60.85, |
|
"learning_rate": 0.00015861671469740636, |
|
"loss": 0.5059, |
|
"step": 2130 |
|
}, |
|
{ |
|
"epoch": 60.99, |
|
"eval_loss": 0.8132546544075012, |
|
"eval_runtime": 8.8063, |
|
"eval_samples_per_second": 7.154, |
|
"eval_steps_per_second": 1.817, |
|
"eval_wer": 0.39814424293547024, |
|
"step": 2135 |
|
}, |
|
{ |
|
"epoch": 61.14, |
|
"learning_rate": 0.00015746397694524496, |
|
"loss": 0.426, |
|
"step": 2140 |
|
}, |
|
{ |
|
"epoch": 61.43, |
|
"learning_rate": 0.0001563112391930836, |
|
"loss": 0.3629, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 61.71, |
|
"learning_rate": 0.0001551585014409222, |
|
"loss": 0.4456, |
|
"step": 2160 |
|
}, |
|
{ |
|
"epoch": 61.99, |
|
"learning_rate": 0.0001540057636887608, |
|
"loss": 0.3927, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 61.99, |
|
"eval_loss": 0.8600913286209106, |
|
"eval_runtime": 8.7725, |
|
"eval_samples_per_second": 7.182, |
|
"eval_steps_per_second": 1.824, |
|
"eval_wer": 0.4032053985660059, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 62.28, |
|
"learning_rate": 0.00015285302593659943, |
|
"loss": 0.4491, |
|
"step": 2180 |
|
}, |
|
{ |
|
"epoch": 62.57, |
|
"learning_rate": 0.00015170028818443804, |
|
"loss": 0.3922, |
|
"step": 2190 |
|
}, |
|
{ |
|
"epoch": 62.85, |
|
"learning_rate": 0.00015054755043227667, |
|
"loss": 0.4297, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 62.99, |
|
"eval_loss": 0.736254096031189, |
|
"eval_runtime": 8.7949, |
|
"eval_samples_per_second": 7.163, |
|
"eval_steps_per_second": 1.819, |
|
"eval_wer": 0.38802193167439897, |
|
"step": 2205 |
|
}, |
|
{ |
|
"epoch": 63.14, |
|
"learning_rate": 0.00014939481268011527, |
|
"loss": 0.4252, |
|
"step": 2210 |
|
}, |
|
{ |
|
"epoch": 63.43, |
|
"learning_rate": 0.0001482420749279539, |
|
"loss": 0.4017, |
|
"step": 2220 |
|
}, |
|
{ |
|
"epoch": 63.71, |
|
"learning_rate": 0.0001470893371757925, |
|
"loss": 0.4588, |
|
"step": 2230 |
|
}, |
|
{ |
|
"epoch": 63.99, |
|
"learning_rate": 0.00014593659942363114, |
|
"loss": 0.4034, |
|
"step": 2240 |
|
}, |
|
{ |
|
"epoch": 63.99, |
|
"eval_loss": 0.7638913989067078, |
|
"eval_runtime": 8.8236, |
|
"eval_samples_per_second": 7.14, |
|
"eval_steps_per_second": 1.813, |
|
"eval_wer": 0.4027836355967946, |
|
"step": 2240 |
|
}, |
|
{ |
|
"epoch": 64.28, |
|
"learning_rate": 0.00014478386167146974, |
|
"loss": 0.3817, |
|
"step": 2250 |
|
}, |
|
{ |
|
"epoch": 64.57, |
|
"learning_rate": 0.00014363112391930838, |
|
"loss": 0.3861, |
|
"step": 2260 |
|
}, |
|
{ |
|
"epoch": 64.85, |
|
"learning_rate": 0.00014247838616714698, |
|
"loss": 0.3731, |
|
"step": 2270 |
|
}, |
|
{ |
|
"epoch": 64.99, |
|
"eval_loss": 0.813651978969574, |
|
"eval_runtime": 8.8337, |
|
"eval_samples_per_second": 7.132, |
|
"eval_steps_per_second": 1.811, |
|
"eval_wer": 0.36862083509067906, |
|
"step": 2275 |
|
}, |
|
{ |
|
"epoch": 65.14, |
|
"learning_rate": 0.00014132564841498558, |
|
"loss": 0.4263, |
|
"step": 2280 |
|
}, |
|
{ |
|
"epoch": 65.43, |
|
"learning_rate": 0.00014017291066282421, |
|
"loss": 0.3928, |
|
"step": 2290 |
|
}, |
|
{ |
|
"epoch": 65.71, |
|
"learning_rate": 0.00013902017291066282, |
|
"loss": 0.444, |
|
"step": 2300 |
|
}, |
|
{ |
|
"epoch": 65.99, |
|
"learning_rate": 0.00013786743515850145, |
|
"loss": 0.3793, |
|
"step": 2310 |
|
}, |
|
{ |
|
"epoch": 65.99, |
|
"eval_loss": 0.764647364616394, |
|
"eval_runtime": 8.7926, |
|
"eval_samples_per_second": 7.165, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.3787431463517503, |
|
"step": 2310 |
|
}, |
|
{ |
|
"epoch": 66.28, |
|
"learning_rate": 0.00013671469740634005, |
|
"loss": 0.4453, |
|
"step": 2320 |
|
}, |
|
{ |
|
"epoch": 66.57, |
|
"learning_rate": 0.00013556195965417869, |
|
"loss": 0.3687, |
|
"step": 2330 |
|
}, |
|
{ |
|
"epoch": 66.85, |
|
"learning_rate": 0.0001344092219020173, |
|
"loss": 0.3593, |
|
"step": 2340 |
|
}, |
|
{ |
|
"epoch": 66.99, |
|
"eval_loss": 0.7878236174583435, |
|
"eval_runtime": 8.8924, |
|
"eval_samples_per_second": 7.085, |
|
"eval_steps_per_second": 1.799, |
|
"eval_wer": 0.39519190215099115, |
|
"step": 2345 |
|
}, |
|
{ |
|
"epoch": 67.14, |
|
"learning_rate": 0.00013325648414985592, |
|
"loss": 0.4116, |
|
"step": 2350 |
|
}, |
|
{ |
|
"epoch": 67.43, |
|
"learning_rate": 0.00013210374639769455, |
|
"loss": 0.4169, |
|
"step": 2360 |
|
}, |
|
{ |
|
"epoch": 67.71, |
|
"learning_rate": 0.00013095100864553316, |
|
"loss": 0.3684, |
|
"step": 2370 |
|
}, |
|
{ |
|
"epoch": 67.99, |
|
"learning_rate": 0.00012979827089337176, |
|
"loss": 0.3616, |
|
"step": 2380 |
|
}, |
|
{ |
|
"epoch": 67.99, |
|
"eval_loss": 0.7935976982116699, |
|
"eval_runtime": 8.8812, |
|
"eval_samples_per_second": 7.094, |
|
"eval_steps_per_second": 1.802, |
|
"eval_wer": 0.4044706874736398, |
|
"step": 2380 |
|
}, |
|
{ |
|
"epoch": 68.28, |
|
"learning_rate": 0.00012864553314121036, |
|
"loss": 0.3535, |
|
"step": 2390 |
|
}, |
|
{ |
|
"epoch": 68.57, |
|
"learning_rate": 0.000127492795389049, |
|
"loss": 0.3636, |
|
"step": 2400 |
|
}, |
|
{ |
|
"epoch": 68.85, |
|
"learning_rate": 0.0001263400576368876, |
|
"loss": 0.3991, |
|
"step": 2410 |
|
}, |
|
{ |
|
"epoch": 68.99, |
|
"eval_loss": 0.7425219416618347, |
|
"eval_runtime": 8.9048, |
|
"eval_samples_per_second": 7.075, |
|
"eval_steps_per_second": 1.797, |
|
"eval_wer": 0.3774778574441164, |
|
"step": 2415 |
|
}, |
|
{ |
|
"epoch": 69.14, |
|
"learning_rate": 0.00012518731988472623, |
|
"loss": 0.434, |
|
"step": 2420 |
|
}, |
|
{ |
|
"epoch": 69.43, |
|
"learning_rate": 0.00012403458213256483, |
|
"loss": 0.3775, |
|
"step": 2430 |
|
}, |
|
{ |
|
"epoch": 69.71, |
|
"learning_rate": 0.00012288184438040347, |
|
"loss": 0.4287, |
|
"step": 2440 |
|
}, |
|
{ |
|
"epoch": 69.99, |
|
"learning_rate": 0.00012172910662824208, |
|
"loss": 0.3709, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 69.99, |
|
"eval_loss": 0.6933401226997375, |
|
"eval_runtime": 8.9112, |
|
"eval_samples_per_second": 7.07, |
|
"eval_steps_per_second": 1.795, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2450 |
|
}, |
|
{ |
|
"epoch": 70.28, |
|
"learning_rate": 0.0001205763688760807, |
|
"loss": 0.4291, |
|
"step": 2460 |
|
}, |
|
{ |
|
"epoch": 70.57, |
|
"learning_rate": 0.00011942363112391932, |
|
"loss": 0.3197, |
|
"step": 2470 |
|
}, |
|
{ |
|
"epoch": 70.85, |
|
"learning_rate": 0.00011827089337175794, |
|
"loss": 0.3886, |
|
"step": 2480 |
|
}, |
|
{ |
|
"epoch": 70.99, |
|
"eval_loss": 0.7044313549995422, |
|
"eval_runtime": 8.8012, |
|
"eval_samples_per_second": 7.158, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.3728384647827921, |
|
"step": 2485 |
|
}, |
|
{ |
|
"epoch": 71.14, |
|
"learning_rate": 0.00011711815561959654, |
|
"loss": 0.3949, |
|
"step": 2490 |
|
}, |
|
{ |
|
"epoch": 71.43, |
|
"learning_rate": 0.00011596541786743516, |
|
"loss": 0.4272, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 71.71, |
|
"learning_rate": 0.00011481268011527378, |
|
"loss": 0.3614, |
|
"step": 2510 |
|
}, |
|
{ |
|
"epoch": 71.99, |
|
"learning_rate": 0.0001136599423631124, |
|
"loss": 0.3624, |
|
"step": 2520 |
|
}, |
|
{ |
|
"epoch": 71.99, |
|
"eval_loss": 0.6915799975395203, |
|
"eval_runtime": 8.804, |
|
"eval_samples_per_second": 7.156, |
|
"eval_steps_per_second": 1.817, |
|
"eval_wer": 0.392239561366512, |
|
"step": 2520 |
|
}, |
|
{ |
|
"epoch": 72.28, |
|
"learning_rate": 0.00011250720461095101, |
|
"loss": 0.422, |
|
"step": 2530 |
|
}, |
|
{ |
|
"epoch": 72.57, |
|
"learning_rate": 0.00011135446685878963, |
|
"loss": 0.3656, |
|
"step": 2540 |
|
}, |
|
{ |
|
"epoch": 72.85, |
|
"learning_rate": 0.00011020172910662825, |
|
"loss": 0.3477, |
|
"step": 2550 |
|
}, |
|
{ |
|
"epoch": 72.99, |
|
"eval_loss": 0.7245261669158936, |
|
"eval_runtime": 8.8472, |
|
"eval_samples_per_second": 7.121, |
|
"eval_steps_per_second": 1.808, |
|
"eval_wer": 0.3871784057359764, |
|
"step": 2555 |
|
}, |
|
{ |
|
"epoch": 73.14, |
|
"learning_rate": 0.00010904899135446686, |
|
"loss": 0.3975, |
|
"step": 2560 |
|
}, |
|
{ |
|
"epoch": 73.43, |
|
"learning_rate": 0.00010789625360230548, |
|
"loss": 0.3748, |
|
"step": 2570 |
|
}, |
|
{ |
|
"epoch": 73.71, |
|
"learning_rate": 0.0001067435158501441, |
|
"loss": 0.4629, |
|
"step": 2580 |
|
}, |
|
{ |
|
"epoch": 73.99, |
|
"learning_rate": 0.00010559077809798272, |
|
"loss": 0.4116, |
|
"step": 2590 |
|
}, |
|
{ |
|
"epoch": 73.99, |
|
"eval_loss": 0.6822821497917175, |
|
"eval_runtime": 8.7923, |
|
"eval_samples_per_second": 7.165, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.3850695908899199, |
|
"step": 2590 |
|
}, |
|
{ |
|
"epoch": 74.28, |
|
"learning_rate": 0.00010443804034582132, |
|
"loss": 0.4009, |
|
"step": 2600 |
|
}, |
|
{ |
|
"epoch": 74.57, |
|
"learning_rate": 0.00010328530259365994, |
|
"loss": 0.3213, |
|
"step": 2610 |
|
}, |
|
{ |
|
"epoch": 74.85, |
|
"learning_rate": 0.00010213256484149856, |
|
"loss": 0.3956, |
|
"step": 2620 |
|
}, |
|
{ |
|
"epoch": 74.99, |
|
"eval_loss": 0.7742548584938049, |
|
"eval_runtime": 8.7905, |
|
"eval_samples_per_second": 7.167, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.38464782792070856, |
|
"step": 2625 |
|
}, |
|
{ |
|
"epoch": 75.14, |
|
"learning_rate": 0.00010097982708933717, |
|
"loss": 0.3585, |
|
"step": 2630 |
|
}, |
|
{ |
|
"epoch": 75.43, |
|
"learning_rate": 9.982708933717579e-05, |
|
"loss": 0.4401, |
|
"step": 2640 |
|
}, |
|
{ |
|
"epoch": 75.71, |
|
"learning_rate": 9.867435158501441e-05, |
|
"loss": 0.3411, |
|
"step": 2650 |
|
}, |
|
{ |
|
"epoch": 75.99, |
|
"learning_rate": 9.752161383285303e-05, |
|
"loss": 0.386, |
|
"step": 2660 |
|
}, |
|
{ |
|
"epoch": 75.99, |
|
"eval_loss": 0.7772214412689209, |
|
"eval_runtime": 8.8141, |
|
"eval_samples_per_second": 7.148, |
|
"eval_steps_per_second": 1.815, |
|
"eval_wer": 0.3943483762125685, |
|
"step": 2660 |
|
}, |
|
{ |
|
"epoch": 76.28, |
|
"learning_rate": 9.636887608069164e-05, |
|
"loss": 0.4381, |
|
"step": 2670 |
|
}, |
|
{ |
|
"epoch": 76.57, |
|
"learning_rate": 9.521613832853028e-05, |
|
"loss": 0.3696, |
|
"step": 2680 |
|
}, |
|
{ |
|
"epoch": 76.85, |
|
"learning_rate": 9.406340057636888e-05, |
|
"loss": 0.3755, |
|
"step": 2690 |
|
}, |
|
{ |
|
"epoch": 76.99, |
|
"eval_loss": 0.7822830080986023, |
|
"eval_runtime": 8.8764, |
|
"eval_samples_per_second": 7.097, |
|
"eval_steps_per_second": 1.803, |
|
"eval_wer": 0.374103753690426, |
|
"step": 2695 |
|
}, |
|
{ |
|
"epoch": 77.14, |
|
"learning_rate": 9.29106628242075e-05, |
|
"loss": 0.3742, |
|
"step": 2700 |
|
}, |
|
{ |
|
"epoch": 77.43, |
|
"learning_rate": 9.175792507204611e-05, |
|
"loss": 0.3448, |
|
"step": 2710 |
|
}, |
|
{ |
|
"epoch": 77.71, |
|
"learning_rate": 9.060518731988473e-05, |
|
"loss": 0.3741, |
|
"step": 2720 |
|
}, |
|
{ |
|
"epoch": 77.99, |
|
"learning_rate": 8.945244956772335e-05, |
|
"loss": 0.3569, |
|
"step": 2730 |
|
}, |
|
{ |
|
"epoch": 77.99, |
|
"eval_loss": 0.7801348567008972, |
|
"eval_runtime": 8.843, |
|
"eval_samples_per_second": 7.124, |
|
"eval_steps_per_second": 1.809, |
|
"eval_wer": 0.38802193167439897, |
|
"step": 2730 |
|
}, |
|
{ |
|
"epoch": 78.28, |
|
"learning_rate": 8.829971181556195e-05, |
|
"loss": 0.4016, |
|
"step": 2740 |
|
}, |
|
{ |
|
"epoch": 78.57, |
|
"learning_rate": 8.714697406340057e-05, |
|
"loss": 0.3491, |
|
"step": 2750 |
|
}, |
|
{ |
|
"epoch": 78.85, |
|
"learning_rate": 8.599423631123919e-05, |
|
"loss": 0.3403, |
|
"step": 2760 |
|
}, |
|
{ |
|
"epoch": 78.99, |
|
"eval_loss": 0.7619218826293945, |
|
"eval_runtime": 8.7919, |
|
"eval_samples_per_second": 7.166, |
|
"eval_steps_per_second": 1.82, |
|
"eval_wer": 0.378321383382539, |
|
"step": 2765 |
|
}, |
|
{ |
|
"epoch": 79.14, |
|
"learning_rate": 8.484149855907782e-05, |
|
"loss": 0.3985, |
|
"step": 2770 |
|
}, |
|
{ |
|
"epoch": 79.43, |
|
"learning_rate": 8.368876080691644e-05, |
|
"loss": 0.3658, |
|
"step": 2780 |
|
}, |
|
{ |
|
"epoch": 79.71, |
|
"learning_rate": 8.253602305475506e-05, |
|
"loss": 0.4117, |
|
"step": 2790 |
|
}, |
|
{ |
|
"epoch": 79.99, |
|
"learning_rate": 8.138328530259366e-05, |
|
"loss": 0.3623, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 79.99, |
|
"eval_loss": 0.7293934226036072, |
|
"eval_runtime": 8.7589, |
|
"eval_samples_per_second": 7.193, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2800 |
|
}, |
|
{ |
|
"epoch": 80.28, |
|
"learning_rate": 8.023054755043228e-05, |
|
"loss": 0.3391, |
|
"step": 2810 |
|
}, |
|
{ |
|
"epoch": 80.57, |
|
"learning_rate": 7.90778097982709e-05, |
|
"loss": 0.3421, |
|
"step": 2820 |
|
}, |
|
{ |
|
"epoch": 80.85, |
|
"learning_rate": 7.792507204610951e-05, |
|
"loss": 0.4157, |
|
"step": 2830 |
|
}, |
|
{ |
|
"epoch": 80.99, |
|
"eval_loss": 0.734462559223175, |
|
"eval_runtime": 8.7583, |
|
"eval_samples_per_second": 7.193, |
|
"eval_steps_per_second": 1.827, |
|
"eval_wer": 0.3854913538591312, |
|
"step": 2835 |
|
}, |
|
{ |
|
"epoch": 81.14, |
|
"learning_rate": 7.677233429394813e-05, |
|
"loss": 0.4109, |
|
"step": 2840 |
|
}, |
|
{ |
|
"epoch": 81.43, |
|
"learning_rate": 7.561959654178675e-05, |
|
"loss": 0.3738, |
|
"step": 2850 |
|
}, |
|
{ |
|
"epoch": 81.71, |
|
"learning_rate": 7.446685878962537e-05, |
|
"loss": 0.313, |
|
"step": 2860 |
|
}, |
|
{ |
|
"epoch": 81.99, |
|
"learning_rate": 7.331412103746398e-05, |
|
"loss": 0.3569, |
|
"step": 2870 |
|
}, |
|
{ |
|
"epoch": 81.99, |
|
"eval_loss": 0.7348675727844238, |
|
"eval_runtime": 8.8356, |
|
"eval_samples_per_second": 7.13, |
|
"eval_steps_per_second": 1.811, |
|
"eval_wer": 0.38043019822859553, |
|
"step": 2870 |
|
}, |
|
{ |
|
"epoch": 82.28, |
|
"learning_rate": 7.21613832853026e-05, |
|
"loss": 0.3578, |
|
"step": 2880 |
|
}, |
|
{ |
|
"epoch": 82.57, |
|
"learning_rate": 7.100864553314122e-05, |
|
"loss": 0.3075, |
|
"step": 2890 |
|
}, |
|
{ |
|
"epoch": 82.85, |
|
"learning_rate": 6.985590778097984e-05, |
|
"loss": 0.3988, |
|
"step": 2900 |
|
}, |
|
{ |
|
"epoch": 82.99, |
|
"eval_loss": 0.723199188709259, |
|
"eval_runtime": 8.755, |
|
"eval_samples_per_second": 7.196, |
|
"eval_steps_per_second": 1.828, |
|
"eval_wer": 0.3833825390130747, |
|
"step": 2905 |
|
}, |
|
{ |
|
"epoch": 83.14, |
|
"learning_rate": 6.870317002881844e-05, |
|
"loss": 0.3896, |
|
"step": 2910 |
|
}, |
|
{ |
|
"epoch": 83.43, |
|
"learning_rate": 6.755043227665706e-05, |
|
"loss": 0.3423, |
|
"step": 2920 |
|
}, |
|
{ |
|
"epoch": 83.71, |
|
"learning_rate": 6.639769452449568e-05, |
|
"loss": 0.4063, |
|
"step": 2930 |
|
}, |
|
{ |
|
"epoch": 83.99, |
|
"learning_rate": 6.52449567723343e-05, |
|
"loss": 0.3425, |
|
"step": 2940 |
|
}, |
|
{ |
|
"epoch": 83.99, |
|
"eval_loss": 0.7238648533821106, |
|
"eval_runtime": 8.814, |
|
"eval_samples_per_second": 7.148, |
|
"eval_steps_per_second": 1.815, |
|
"eval_wer": 0.37916490932096164, |
|
"step": 2940 |
|
}, |
|
{ |
|
"epoch": 84.28, |
|
"learning_rate": 6.409221902017291e-05, |
|
"loss": 0.3631, |
|
"step": 2950 |
|
}, |
|
{ |
|
"epoch": 84.57, |
|
"learning_rate": 6.293948126801153e-05, |
|
"loss": 0.3077, |
|
"step": 2960 |
|
}, |
|
{ |
|
"epoch": 84.85, |
|
"learning_rate": 6.178674351585015e-05, |
|
"loss": 0.353, |
|
"step": 2970 |
|
}, |
|
{ |
|
"epoch": 84.99, |
|
"eval_loss": 0.73667973279953, |
|
"eval_runtime": 8.9739, |
|
"eval_samples_per_second": 7.02, |
|
"eval_steps_per_second": 1.783, |
|
"eval_wer": 0.3757908055672712, |
|
"step": 2975 |
|
}, |
|
{ |
|
"epoch": 85.14, |
|
"learning_rate": 6.0634005763688764e-05, |
|
"loss": 0.3622, |
|
"step": 2980 |
|
}, |
|
{ |
|
"epoch": 85.43, |
|
"learning_rate": 5.948126801152738e-05, |
|
"loss": 0.3514, |
|
"step": 2990 |
|
}, |
|
{ |
|
"epoch": 85.71, |
|
"learning_rate": 5.8328530259366e-05, |
|
"loss": 0.335, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 85.99, |
|
"learning_rate": 5.717579250720462e-05, |
|
"loss": 0.3756, |
|
"step": 3010 |
|
}, |
|
{ |
|
"epoch": 85.99, |
|
"eval_loss": 0.7282797694206238, |
|
"eval_runtime": 8.8955, |
|
"eval_samples_per_second": 7.082, |
|
"eval_steps_per_second": 1.799, |
|
"eval_wer": 0.3728384647827921, |
|
"step": 3010 |
|
}, |
|
{ |
|
"epoch": 86.28, |
|
"learning_rate": 5.602305475504323e-05, |
|
"loss": 0.3863, |
|
"step": 3020 |
|
}, |
|
{ |
|
"epoch": 86.57, |
|
"learning_rate": 5.4870317002881845e-05, |
|
"loss": 0.3053, |
|
"step": 3030 |
|
}, |
|
{ |
|
"epoch": 86.85, |
|
"learning_rate": 5.371757925072046e-05, |
|
"loss": 0.3702, |
|
"step": 3040 |
|
}, |
|
{ |
|
"epoch": 86.99, |
|
"eval_loss": 0.7044227719306946, |
|
"eval_runtime": 8.9143, |
|
"eval_samples_per_second": 7.067, |
|
"eval_steps_per_second": 1.795, |
|
"eval_wer": 0.37916490932096164, |
|
"step": 3045 |
|
}, |
|
{ |
|
"epoch": 87.14, |
|
"learning_rate": 5.256484149855908e-05, |
|
"loss": 0.3741, |
|
"step": 3050 |
|
}, |
|
{ |
|
"epoch": 87.43, |
|
"learning_rate": 5.14121037463977e-05, |
|
"loss": 0.3763, |
|
"step": 3060 |
|
}, |
|
{ |
|
"epoch": 87.71, |
|
"learning_rate": 5.0259365994236316e-05, |
|
"loss": 0.3985, |
|
"step": 3070 |
|
}, |
|
{ |
|
"epoch": 87.99, |
|
"learning_rate": 4.9106628242074933e-05, |
|
"loss": 0.3339, |
|
"step": 3080 |
|
}, |
|
{ |
|
"epoch": 87.99, |
|
"eval_loss": 0.7279428839683533, |
|
"eval_runtime": 8.7882, |
|
"eval_samples_per_second": 7.169, |
|
"eval_steps_per_second": 1.821, |
|
"eval_wer": 0.3766343315056938, |
|
"step": 3080 |
|
}, |
|
{ |
|
"epoch": 88.28, |
|
"learning_rate": 4.7953890489913544e-05, |
|
"loss": 0.4032, |
|
"step": 3090 |
|
}, |
|
{ |
|
"epoch": 88.57, |
|
"learning_rate": 4.680115273775216e-05, |
|
"loss": 0.2959, |
|
"step": 3100 |
|
}, |
|
{ |
|
"epoch": 88.85, |
|
"learning_rate": 4.564841498559078e-05, |
|
"loss": 0.3161, |
|
"step": 3110 |
|
}, |
|
{ |
|
"epoch": 88.99, |
|
"eval_loss": 0.7680357694625854, |
|
"eval_runtime": 8.788, |
|
"eval_samples_per_second": 7.169, |
|
"eval_steps_per_second": 1.821, |
|
"eval_wer": 0.3795866722901729, |
|
"step": 3115 |
|
}, |
|
{ |
|
"epoch": 89.14, |
|
"learning_rate": 4.44956772334294e-05, |
|
"loss": 0.3987, |
|
"step": 3120 |
|
}, |
|
{ |
|
"epoch": 89.43, |
|
"learning_rate": 4.3342939481268015e-05, |
|
"loss": 0.3587, |
|
"step": 3130 |
|
}, |
|
{ |
|
"epoch": 89.71, |
|
"learning_rate": 4.219020172910663e-05, |
|
"loss": 0.3373, |
|
"step": 3140 |
|
}, |
|
{ |
|
"epoch": 89.99, |
|
"learning_rate": 4.103746397694524e-05, |
|
"loss": 0.3573, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 89.99, |
|
"eval_loss": 0.7497588992118835, |
|
"eval_runtime": 8.7271, |
|
"eval_samples_per_second": 7.219, |
|
"eval_steps_per_second": 1.833, |
|
"eval_wer": 0.37326022775200335, |
|
"step": 3150 |
|
}, |
|
{ |
|
"epoch": 90.28, |
|
"learning_rate": 3.988472622478386e-05, |
|
"loss": 0.3477, |
|
"step": 3160 |
|
}, |
|
{ |
|
"epoch": 90.57, |
|
"learning_rate": 3.8731988472622485e-05, |
|
"loss": 0.3027, |
|
"step": 3170 |
|
}, |
|
{ |
|
"epoch": 90.85, |
|
"learning_rate": 3.7579250720461096e-05, |
|
"loss": 0.3557, |
|
"step": 3180 |
|
}, |
|
{ |
|
"epoch": 90.99, |
|
"eval_loss": 0.7433480620384216, |
|
"eval_runtime": 8.6925, |
|
"eval_samples_per_second": 7.248, |
|
"eval_steps_per_second": 1.841, |
|
"eval_wer": 0.3778996204133277, |
|
"step": 3185 |
|
}, |
|
{ |
|
"epoch": 91.14, |
|
"learning_rate": 3.6426512968299714e-05, |
|
"loss": 0.3406, |
|
"step": 3190 |
|
}, |
|
{ |
|
"epoch": 91.43, |
|
"learning_rate": 3.527377521613833e-05, |
|
"loss": 0.3995, |
|
"step": 3200 |
|
}, |
|
{ |
|
"epoch": 91.71, |
|
"learning_rate": 3.412103746397695e-05, |
|
"loss": 0.3406, |
|
"step": 3210 |
|
}, |
|
{ |
|
"epoch": 91.99, |
|
"learning_rate": 3.296829971181557e-05, |
|
"loss": 0.3563, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 91.99, |
|
"eval_loss": 0.7249133586883545, |
|
"eval_runtime": 8.7779, |
|
"eval_samples_per_second": 7.177, |
|
"eval_steps_per_second": 1.823, |
|
"eval_wer": 0.3787431463517503, |
|
"step": 3220 |
|
}, |
|
{ |
|
"epoch": 92.28, |
|
"learning_rate": 3.181556195965418e-05, |
|
"loss": 0.341, |
|
"step": 3230 |
|
}, |
|
{ |
|
"epoch": 92.57, |
|
"learning_rate": 3.0662824207492795e-05, |
|
"loss": 0.3215, |
|
"step": 3240 |
|
}, |
|
{ |
|
"epoch": 92.85, |
|
"learning_rate": 2.9510086455331416e-05, |
|
"loss": 0.3304, |
|
"step": 3250 |
|
}, |
|
{ |
|
"epoch": 92.99, |
|
"eval_loss": 0.7542759776115417, |
|
"eval_runtime": 8.8038, |
|
"eval_samples_per_second": 7.156, |
|
"eval_steps_per_second": 1.817, |
|
"eval_wer": 0.378321383382539, |
|
"step": 3255 |
|
}, |
|
{ |
|
"epoch": 93.14, |
|
"learning_rate": 2.8357348703170027e-05, |
|
"loss": 0.3499, |
|
"step": 3260 |
|
}, |
|
{ |
|
"epoch": 93.43, |
|
"learning_rate": 2.7204610951008648e-05, |
|
"loss": 0.3892, |
|
"step": 3270 |
|
}, |
|
{ |
|
"epoch": 93.71, |
|
"learning_rate": 2.6051873198847266e-05, |
|
"loss": 0.3168, |
|
"step": 3280 |
|
}, |
|
{ |
|
"epoch": 93.99, |
|
"learning_rate": 2.489913544668588e-05, |
|
"loss": 0.3596, |
|
"step": 3290 |
|
}, |
|
{ |
|
"epoch": 93.99, |
|
"eval_loss": 0.7328776121139526, |
|
"eval_runtime": 8.7016, |
|
"eval_samples_per_second": 7.24, |
|
"eval_steps_per_second": 1.839, |
|
"eval_wer": 0.37326022775200335, |
|
"step": 3290 |
|
}, |
|
{ |
|
"epoch": 94.28, |
|
"learning_rate": 2.3746397694524498e-05, |
|
"loss": 0.3471, |
|
"step": 3300 |
|
}, |
|
{ |
|
"epoch": 94.57, |
|
"learning_rate": 2.2593659942363112e-05, |
|
"loss": 0.3556, |
|
"step": 3310 |
|
}, |
|
{ |
|
"epoch": 94.85, |
|
"learning_rate": 2.144092219020173e-05, |
|
"loss": 0.3548, |
|
"step": 3320 |
|
}, |
|
{ |
|
"epoch": 94.99, |
|
"eval_loss": 0.7530630230903625, |
|
"eval_runtime": 8.7431, |
|
"eval_samples_per_second": 7.206, |
|
"eval_steps_per_second": 1.83, |
|
"eval_wer": 0.37199493884436946, |
|
"step": 3325 |
|
}, |
|
{ |
|
"epoch": 95.14, |
|
"learning_rate": 2.0288184438040347e-05, |
|
"loss": 0.3257, |
|
"step": 3330 |
|
}, |
|
{ |
|
"epoch": 95.43, |
|
"learning_rate": 1.9135446685878965e-05, |
|
"loss": 0.3159, |
|
"step": 3340 |
|
}, |
|
{ |
|
"epoch": 95.71, |
|
"learning_rate": 1.798270893371758e-05, |
|
"loss": 0.3744, |
|
"step": 3350 |
|
}, |
|
{ |
|
"epoch": 95.99, |
|
"learning_rate": 1.6829971181556197e-05, |
|
"loss": 0.3269, |
|
"step": 3360 |
|
}, |
|
{ |
|
"epoch": 95.99, |
|
"eval_loss": 0.7376612424850464, |
|
"eval_runtime": 8.7042, |
|
"eval_samples_per_second": 7.238, |
|
"eval_steps_per_second": 1.838, |
|
"eval_wer": 0.37115141290594683, |
|
"step": 3360 |
|
}, |
|
{ |
|
"epoch": 96.28, |
|
"learning_rate": 1.5677233429394814e-05, |
|
"loss": 0.4905, |
|
"step": 3370 |
|
}, |
|
{ |
|
"epoch": 96.57, |
|
"learning_rate": 1.452449567723343e-05, |
|
"loss": 0.2965, |
|
"step": 3380 |
|
}, |
|
{ |
|
"epoch": 96.85, |
|
"learning_rate": 1.3371757925072045e-05, |
|
"loss": 0.3289, |
|
"step": 3390 |
|
}, |
|
{ |
|
"epoch": 96.99, |
|
"eval_loss": 0.7378009557723999, |
|
"eval_runtime": 8.7739, |
|
"eval_samples_per_second": 7.18, |
|
"eval_steps_per_second": 1.824, |
|
"eval_wer": 0.3749472796288486, |
|
"step": 3395 |
|
}, |
|
{ |
|
"epoch": 97.14, |
|
"learning_rate": 1.2219020172910664e-05, |
|
"loss": 0.3329, |
|
"step": 3400 |
|
}, |
|
{ |
|
"epoch": 97.43, |
|
"learning_rate": 1.106628242074928e-05, |
|
"loss": 0.3637, |
|
"step": 3410 |
|
}, |
|
{ |
|
"epoch": 97.71, |
|
"learning_rate": 9.913544668587897e-06, |
|
"loss": 0.3848, |
|
"step": 3420 |
|
}, |
|
{ |
|
"epoch": 97.99, |
|
"learning_rate": 8.760806916426513e-06, |
|
"loss": 0.2978, |
|
"step": 3430 |
|
}, |
|
{ |
|
"epoch": 97.99, |
|
"eval_loss": 0.7200188636779785, |
|
"eval_runtime": 8.7723, |
|
"eval_samples_per_second": 7.182, |
|
"eval_steps_per_second": 1.824, |
|
"eval_wer": 0.3728384647827921, |
|
"step": 3430 |
|
}, |
|
{ |
|
"epoch": 98.28, |
|
"learning_rate": 7.60806916426513e-06, |
|
"loss": 0.3115, |
|
"step": 3440 |
|
}, |
|
{ |
|
"epoch": 98.57, |
|
"learning_rate": 6.455331412103746e-06, |
|
"loss": 0.3524, |
|
"step": 3450 |
|
}, |
|
{ |
|
"epoch": 98.85, |
|
"learning_rate": 5.302593659942364e-06, |
|
"loss": 0.3075, |
|
"step": 3460 |
|
}, |
|
{ |
|
"epoch": 98.99, |
|
"eval_loss": 0.7210068702697754, |
|
"eval_runtime": 8.8032, |
|
"eval_samples_per_second": 7.157, |
|
"eval_steps_per_second": 1.818, |
|
"eval_wer": 0.3724167018135808, |
|
"step": 3465 |
|
}, |
|
{ |
|
"epoch": 99.14, |
|
"learning_rate": 4.14985590778098e-06, |
|
"loss": 0.3749, |
|
"step": 3470 |
|
}, |
|
{ |
|
"epoch": 99.43, |
|
"learning_rate": 2.9971181556195965e-06, |
|
"loss": 0.3387, |
|
"step": 3480 |
|
}, |
|
{ |
|
"epoch": 99.71, |
|
"learning_rate": 1.8443804034582135e-06, |
|
"loss": 0.3714, |
|
"step": 3490 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"learning_rate": 6.9164265129683e-07, |
|
"loss": 0.3402, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"eval_loss": 0.7173439264297485, |
|
"eval_runtime": 8.8444, |
|
"eval_samples_per_second": 7.123, |
|
"eval_steps_per_second": 1.809, |
|
"eval_wer": 0.37368199072121466, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 99.99, |
|
"step": 3500, |
|
"total_flos": 5.739635222597828e+18, |
|
"train_loss": 0.9361405113765172, |
|
"train_runtime": 11874.1269, |
|
"train_samples_per_second": 4.741, |
|
"train_steps_per_second": 0.295 |
|
} |
|
], |
|
"max_steps": 3500, |
|
"num_train_epochs": 100, |
|
"total_flos": 5.739635222597828e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|