|
{ |
|
"best_metric": 0.5264096260070801, |
|
"best_model_checkpoint": "ai-light-dance_drums_ft_pretrain_wav2vec2-base-new-v2/checkpoint-1188", |
|
"epoch": 99.97777777777777, |
|
"global_step": 2200, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.44, |
|
"learning_rate": 0.00024, |
|
"loss": 12.475, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"learning_rate": 0.00039890410958904113, |
|
"loss": 4.6468, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"eval_loss": 3.2315070629119873, |
|
"eval_runtime": 2.3722, |
|
"eval_samples_per_second": 7.166, |
|
"eval_steps_per_second": 2.108, |
|
"eval_wer": 1.0, |
|
"step": 22 |
|
}, |
|
{ |
|
"epoch": 1.36, |
|
"learning_rate": 0.00039707762557077624, |
|
"loss": 2.2321, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 1.8, |
|
"learning_rate": 0.0003952511415525114, |
|
"loss": 1.5745, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 1.98, |
|
"eval_loss": 3.1603267192840576, |
|
"eval_runtime": 2.5264, |
|
"eval_samples_per_second": 6.729, |
|
"eval_steps_per_second": 1.979, |
|
"eval_wer": 1.0, |
|
"step": 44 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 0.0003934246575342466, |
|
"loss": 1.6456, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.71, |
|
"learning_rate": 0.00039159817351598175, |
|
"loss": 1.465, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.98, |
|
"eval_loss": 2.2551496028900146, |
|
"eval_runtime": 2.4489, |
|
"eval_samples_per_second": 6.942, |
|
"eval_steps_per_second": 2.042, |
|
"eval_wer": 1.0, |
|
"step": 66 |
|
}, |
|
{ |
|
"epoch": 3.18, |
|
"learning_rate": 0.0003897716894977169, |
|
"loss": 1.3799, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 3.62, |
|
"learning_rate": 0.0003879452054794521, |
|
"loss": 1.3168, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 3.98, |
|
"eval_loss": 1.8460932970046997, |
|
"eval_runtime": 2.5217, |
|
"eval_samples_per_second": 6.741, |
|
"eval_steps_per_second": 1.983, |
|
"eval_wer": 1.0, |
|
"step": 88 |
|
}, |
|
{ |
|
"epoch": 4.09, |
|
"learning_rate": 0.0003861187214611872, |
|
"loss": 1.4043, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 4.53, |
|
"learning_rate": 0.0003842922374429224, |
|
"loss": 1.2482, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"learning_rate": 0.00038246575342465753, |
|
"loss": 1.1359, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 4.98, |
|
"eval_loss": 1.4874340295791626, |
|
"eval_runtime": 2.4675, |
|
"eval_samples_per_second": 6.89, |
|
"eval_steps_per_second": 2.026, |
|
"eval_wer": 0.9797377830750894, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 5.44, |
|
"learning_rate": 0.0003806392694063927, |
|
"loss": 1.085, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 5.89, |
|
"learning_rate": 0.00037881278538812787, |
|
"loss": 0.9769, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 5.98, |
|
"eval_loss": 1.7358566522598267, |
|
"eval_runtime": 2.4383, |
|
"eval_samples_per_second": 6.972, |
|
"eval_steps_per_second": 2.051, |
|
"eval_wer": 0.5494636471990465, |
|
"step": 132 |
|
}, |
|
{ |
|
"epoch": 6.36, |
|
"learning_rate": 0.00037698630136986304, |
|
"loss": 1.0217, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 6.8, |
|
"learning_rate": 0.00037515981735159815, |
|
"loss": 0.9019, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 6.98, |
|
"eval_loss": 1.5833041667938232, |
|
"eval_runtime": 2.5274, |
|
"eval_samples_per_second": 6.726, |
|
"eval_steps_per_second": 1.978, |
|
"eval_wer": 0.5268176400476758, |
|
"step": 154 |
|
}, |
|
{ |
|
"epoch": 7.27, |
|
"learning_rate": 0.0003733333333333334, |
|
"loss": 0.9973, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 7.71, |
|
"learning_rate": 0.0003715068493150685, |
|
"loss": 0.8057, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 7.98, |
|
"eval_loss": 1.4891672134399414, |
|
"eval_runtime": 2.5017, |
|
"eval_samples_per_second": 6.795, |
|
"eval_steps_per_second": 1.999, |
|
"eval_wer": 0.5303933253873659, |
|
"step": 176 |
|
}, |
|
{ |
|
"epoch": 8.18, |
|
"learning_rate": 0.00036968036529680365, |
|
"loss": 0.7626, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 8.62, |
|
"learning_rate": 0.0003678538812785388, |
|
"loss": 1.0845, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 8.98, |
|
"eval_loss": 1.393895149230957, |
|
"eval_runtime": 2.5045, |
|
"eval_samples_per_second": 6.788, |
|
"eval_steps_per_second": 1.996, |
|
"eval_wer": 0.5196662693682956, |
|
"step": 198 |
|
}, |
|
{ |
|
"epoch": 9.09, |
|
"learning_rate": 0.000366027397260274, |
|
"loss": 0.788, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 9.53, |
|
"learning_rate": 0.00036420091324200916, |
|
"loss": 0.6903, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 9.98, |
|
"learning_rate": 0.0003623744292237443, |
|
"loss": 0.7562, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 9.98, |
|
"eval_loss": 1.1237640380859375, |
|
"eval_runtime": 2.4848, |
|
"eval_samples_per_second": 6.842, |
|
"eval_steps_per_second": 2.012, |
|
"eval_wer": 0.5446960667461264, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 10.44, |
|
"learning_rate": 0.00036054794520547944, |
|
"loss": 0.702, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 10.89, |
|
"learning_rate": 0.00035872146118721466, |
|
"loss": 0.7259, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 10.98, |
|
"eval_loss": 1.2935717105865479, |
|
"eval_runtime": 2.35, |
|
"eval_samples_per_second": 7.234, |
|
"eval_steps_per_second": 2.128, |
|
"eval_wer": 0.5005959475566151, |
|
"step": 242 |
|
}, |
|
{ |
|
"epoch": 11.36, |
|
"learning_rate": 0.0003568949771689498, |
|
"loss": 0.6347, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 11.8, |
|
"learning_rate": 0.00035506849315068494, |
|
"loss": 0.7318, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 11.98, |
|
"eval_loss": 1.2763104438781738, |
|
"eval_runtime": 2.3731, |
|
"eval_samples_per_second": 7.164, |
|
"eval_steps_per_second": 2.107, |
|
"eval_wer": 0.466030989272944, |
|
"step": 264 |
|
}, |
|
{ |
|
"epoch": 12.27, |
|
"learning_rate": 0.0003532420091324201, |
|
"loss": 0.8421, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 12.71, |
|
"learning_rate": 0.0003514155251141553, |
|
"loss": 0.6452, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 12.98, |
|
"eval_loss": 1.2947018146514893, |
|
"eval_runtime": 2.3625, |
|
"eval_samples_per_second": 7.196, |
|
"eval_steps_per_second": 2.116, |
|
"eval_wer": 0.47794994040524436, |
|
"step": 286 |
|
}, |
|
{ |
|
"epoch": 13.18, |
|
"learning_rate": 0.0003495890410958904, |
|
"loss": 0.663, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 13.62, |
|
"learning_rate": 0.0003477625570776256, |
|
"loss": 0.6353, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 13.98, |
|
"eval_loss": 1.1925369501113892, |
|
"eval_runtime": 2.3586, |
|
"eval_samples_per_second": 7.208, |
|
"eval_steps_per_second": 2.12, |
|
"eval_wer": 0.45172824791418353, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 14.09, |
|
"learning_rate": 0.00034593607305936073, |
|
"loss": 0.6161, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 14.53, |
|
"learning_rate": 0.0003441095890410959, |
|
"loss": 0.657, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 14.98, |
|
"learning_rate": 0.00034228310502283106, |
|
"loss": 0.6463, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 14.98, |
|
"eval_loss": 0.8667492866516113, |
|
"eval_runtime": 2.3648, |
|
"eval_samples_per_second": 7.189, |
|
"eval_steps_per_second": 2.114, |
|
"eval_wer": 0.4100119189511323, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 15.44, |
|
"learning_rate": 0.00034045662100456623, |
|
"loss": 0.6894, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 15.89, |
|
"learning_rate": 0.00033863013698630135, |
|
"loss": 0.5381, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 15.98, |
|
"eval_loss": 1.1242961883544922, |
|
"eval_runtime": 2.3712, |
|
"eval_samples_per_second": 7.169, |
|
"eval_steps_per_second": 2.109, |
|
"eval_wer": 0.3909415971394517, |
|
"step": 352 |
|
}, |
|
{ |
|
"epoch": 16.36, |
|
"learning_rate": 0.00033680365296803657, |
|
"loss": 0.6061, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 16.8, |
|
"learning_rate": 0.0003349771689497717, |
|
"loss": 0.5637, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 16.98, |
|
"eval_loss": 0.8682847023010254, |
|
"eval_runtime": 2.3793, |
|
"eval_samples_per_second": 7.145, |
|
"eval_steps_per_second": 2.101, |
|
"eval_wer": 0.37544696066746125, |
|
"step": 374 |
|
}, |
|
{ |
|
"epoch": 17.27, |
|
"learning_rate": 0.00033315068493150685, |
|
"loss": 0.5715, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 17.71, |
|
"learning_rate": 0.000331324200913242, |
|
"loss": 0.6149, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 17.98, |
|
"eval_loss": 1.1040120124816895, |
|
"eval_runtime": 2.3707, |
|
"eval_samples_per_second": 7.171, |
|
"eval_steps_per_second": 2.109, |
|
"eval_wer": 0.3730631704410012, |
|
"step": 396 |
|
}, |
|
{ |
|
"epoch": 18.18, |
|
"learning_rate": 0.0003294977168949772, |
|
"loss": 0.5399, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 18.62, |
|
"learning_rate": 0.00032767123287671235, |
|
"loss": 0.6138, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 18.98, |
|
"eval_loss": 1.1067866086959839, |
|
"eval_runtime": 2.3709, |
|
"eval_samples_per_second": 7.17, |
|
"eval_steps_per_second": 2.109, |
|
"eval_wer": 0.38498212157330153, |
|
"step": 418 |
|
}, |
|
{ |
|
"epoch": 19.09, |
|
"learning_rate": 0.0003258447488584475, |
|
"loss": 0.5364, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 19.53, |
|
"learning_rate": 0.00032401826484018264, |
|
"loss": 0.4998, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 19.98, |
|
"learning_rate": 0.00032219178082191786, |
|
"loss": 0.7381, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 19.98, |
|
"eval_loss": 0.9203162789344788, |
|
"eval_runtime": 2.3697, |
|
"eval_samples_per_second": 7.174, |
|
"eval_steps_per_second": 2.11, |
|
"eval_wer": 0.36233611442193087, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 20.44, |
|
"learning_rate": 0.00032036529680365297, |
|
"loss": 0.5755, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 20.89, |
|
"learning_rate": 0.00031853881278538814, |
|
"loss": 0.5064, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 20.98, |
|
"eval_loss": 0.8806262016296387, |
|
"eval_runtime": 2.3863, |
|
"eval_samples_per_second": 7.124, |
|
"eval_steps_per_second": 2.095, |
|
"eval_wer": 0.3539928486293206, |
|
"step": 462 |
|
}, |
|
{ |
|
"epoch": 21.36, |
|
"learning_rate": 0.0003167123287671233, |
|
"loss": 0.5148, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 21.8, |
|
"learning_rate": 0.0003148858447488585, |
|
"loss": 0.4731, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 21.98, |
|
"eval_loss": 0.7259242534637451, |
|
"eval_runtime": 2.3798, |
|
"eval_samples_per_second": 7.144, |
|
"eval_steps_per_second": 2.101, |
|
"eval_wer": 0.36233611442193087, |
|
"step": 484 |
|
}, |
|
{ |
|
"epoch": 22.27, |
|
"learning_rate": 0.0003130593607305936, |
|
"loss": 0.5034, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 22.71, |
|
"learning_rate": 0.0003112328767123288, |
|
"loss": 0.5232, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 22.98, |
|
"eval_loss": 0.7934766411781311, |
|
"eval_runtime": 2.3715, |
|
"eval_samples_per_second": 7.168, |
|
"eval_steps_per_second": 2.108, |
|
"eval_wer": 0.35160905840286055, |
|
"step": 506 |
|
}, |
|
{ |
|
"epoch": 23.18, |
|
"learning_rate": 0.0003094063926940639, |
|
"loss": 0.4923, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 23.62, |
|
"learning_rate": 0.0003075799086757991, |
|
"loss": 0.4689, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 23.98, |
|
"eval_loss": 0.777065634727478, |
|
"eval_runtime": 2.3797, |
|
"eval_samples_per_second": 7.144, |
|
"eval_steps_per_second": 2.101, |
|
"eval_wer": 0.3539928486293206, |
|
"step": 528 |
|
}, |
|
{ |
|
"epoch": 24.09, |
|
"learning_rate": 0.00030575342465753426, |
|
"loss": 0.4745, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 24.53, |
|
"learning_rate": 0.00030392694063926943, |
|
"loss": 0.4817, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 24.98, |
|
"learning_rate": 0.00030210045662100454, |
|
"loss": 0.4902, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 24.98, |
|
"eval_loss": 0.6896762847900391, |
|
"eval_runtime": 2.3739, |
|
"eval_samples_per_second": 7.161, |
|
"eval_steps_per_second": 2.106, |
|
"eval_wer": 0.3909415971394517, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 25.44, |
|
"learning_rate": 0.00030027397260273976, |
|
"loss": 0.5525, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 25.89, |
|
"learning_rate": 0.0002984474885844749, |
|
"loss": 0.4079, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 25.98, |
|
"eval_loss": 0.8030110597610474, |
|
"eval_runtime": 2.3676, |
|
"eval_samples_per_second": 7.18, |
|
"eval_steps_per_second": 2.112, |
|
"eval_wer": 0.35518474374255066, |
|
"step": 572 |
|
}, |
|
{ |
|
"epoch": 26.36, |
|
"learning_rate": 0.00029662100456621005, |
|
"loss": 0.4054, |
|
"step": 580 |
|
}, |
|
{ |
|
"epoch": 26.8, |
|
"learning_rate": 0.0002947945205479452, |
|
"loss": 0.5045, |
|
"step": 590 |
|
}, |
|
{ |
|
"epoch": 26.98, |
|
"eval_loss": 0.6778237223625183, |
|
"eval_runtime": 2.3715, |
|
"eval_samples_per_second": 7.168, |
|
"eval_steps_per_second": 2.108, |
|
"eval_wer": 0.37902264600715135, |
|
"step": 594 |
|
}, |
|
{ |
|
"epoch": 27.27, |
|
"learning_rate": 0.0002929680365296804, |
|
"loss": 0.4829, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 27.71, |
|
"learning_rate": 0.00029114155251141555, |
|
"loss": 0.4373, |
|
"step": 610 |
|
}, |
|
{ |
|
"epoch": 27.98, |
|
"eval_loss": 0.7455542683601379, |
|
"eval_runtime": 2.3574, |
|
"eval_samples_per_second": 7.211, |
|
"eval_steps_per_second": 2.121, |
|
"eval_wer": 0.3694874851013111, |
|
"step": 616 |
|
}, |
|
{ |
|
"epoch": 28.18, |
|
"learning_rate": 0.0002893150684931507, |
|
"loss": 0.3957, |
|
"step": 620 |
|
}, |
|
{ |
|
"epoch": 28.62, |
|
"learning_rate": 0.00028748858447488583, |
|
"loss": 0.4366, |
|
"step": 630 |
|
}, |
|
{ |
|
"epoch": 28.98, |
|
"eval_loss": 0.7009137868881226, |
|
"eval_runtime": 2.3463, |
|
"eval_samples_per_second": 7.245, |
|
"eval_steps_per_second": 2.131, |
|
"eval_wer": 0.3432657926102503, |
|
"step": 638 |
|
}, |
|
{ |
|
"epoch": 29.09, |
|
"learning_rate": 0.00028566210045662105, |
|
"loss": 0.4075, |
|
"step": 640 |
|
}, |
|
{ |
|
"epoch": 29.53, |
|
"learning_rate": 0.00028383561643835617, |
|
"loss": 0.4399, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 29.98, |
|
"learning_rate": 0.00028200913242009133, |
|
"loss": 0.3944, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 29.98, |
|
"eval_loss": 0.6841340661048889, |
|
"eval_runtime": 2.4508, |
|
"eval_samples_per_second": 6.936, |
|
"eval_steps_per_second": 2.04, |
|
"eval_wer": 0.3468414779499404, |
|
"step": 660 |
|
}, |
|
{ |
|
"epoch": 30.44, |
|
"learning_rate": 0.0002801826484018265, |
|
"loss": 0.419, |
|
"step": 670 |
|
}, |
|
{ |
|
"epoch": 30.89, |
|
"learning_rate": 0.00027835616438356167, |
|
"loss": 0.4206, |
|
"step": 680 |
|
}, |
|
{ |
|
"epoch": 30.98, |
|
"eval_loss": 0.7093454599380493, |
|
"eval_runtime": 2.3527, |
|
"eval_samples_per_second": 7.226, |
|
"eval_steps_per_second": 2.125, |
|
"eval_wer": 0.33730631704410013, |
|
"step": 682 |
|
}, |
|
{ |
|
"epoch": 31.36, |
|
"learning_rate": 0.0002765296803652968, |
|
"loss": 0.406, |
|
"step": 690 |
|
}, |
|
{ |
|
"epoch": 31.8, |
|
"learning_rate": 0.000274703196347032, |
|
"loss": 0.3949, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 31.98, |
|
"eval_loss": 0.6900954842567444, |
|
"eval_runtime": 2.3751, |
|
"eval_samples_per_second": 7.157, |
|
"eval_steps_per_second": 2.105, |
|
"eval_wer": 0.3575685339690107, |
|
"step": 704 |
|
}, |
|
{ |
|
"epoch": 32.27, |
|
"learning_rate": 0.0002728767123287671, |
|
"loss": 0.4085, |
|
"step": 710 |
|
}, |
|
{ |
|
"epoch": 32.71, |
|
"learning_rate": 0.0002710502283105023, |
|
"loss": 0.4416, |
|
"step": 720 |
|
}, |
|
{ |
|
"epoch": 32.98, |
|
"eval_loss": 0.6762455105781555, |
|
"eval_runtime": 2.4676, |
|
"eval_samples_per_second": 6.889, |
|
"eval_steps_per_second": 2.026, |
|
"eval_wer": 0.3396901072705602, |
|
"step": 726 |
|
}, |
|
{ |
|
"epoch": 33.18, |
|
"learning_rate": 0.00026922374429223746, |
|
"loss": 0.4288, |
|
"step": 730 |
|
}, |
|
{ |
|
"epoch": 33.62, |
|
"learning_rate": 0.0002673972602739726, |
|
"loss": 0.4248, |
|
"step": 740 |
|
}, |
|
{ |
|
"epoch": 33.98, |
|
"eval_loss": 0.7196361422538757, |
|
"eval_runtime": 2.3986, |
|
"eval_samples_per_second": 7.088, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.3539928486293206, |
|
"step": 748 |
|
}, |
|
{ |
|
"epoch": 34.09, |
|
"learning_rate": 0.0002655707762557078, |
|
"loss": 0.4098, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 34.53, |
|
"learning_rate": 0.00026374429223744296, |
|
"loss": 0.4203, |
|
"step": 760 |
|
}, |
|
{ |
|
"epoch": 34.98, |
|
"learning_rate": 0.00026191780821917807, |
|
"loss": 0.4214, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 34.98, |
|
"eval_loss": 0.6669135093688965, |
|
"eval_runtime": 2.3951, |
|
"eval_samples_per_second": 7.098, |
|
"eval_steps_per_second": 2.088, |
|
"eval_wer": 0.3253873659117998, |
|
"step": 770 |
|
}, |
|
{ |
|
"epoch": 35.44, |
|
"learning_rate": 0.00026009132420091324, |
|
"loss": 0.4146, |
|
"step": 780 |
|
}, |
|
{ |
|
"epoch": 35.89, |
|
"learning_rate": 0.0002582648401826484, |
|
"loss": 0.416, |
|
"step": 790 |
|
}, |
|
{ |
|
"epoch": 35.98, |
|
"eval_loss": 0.6421573758125305, |
|
"eval_runtime": 2.439, |
|
"eval_samples_per_second": 6.97, |
|
"eval_steps_per_second": 2.05, |
|
"eval_wer": 0.34445768772348034, |
|
"step": 792 |
|
}, |
|
{ |
|
"epoch": 36.36, |
|
"learning_rate": 0.0002564383561643836, |
|
"loss": 0.443, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 36.8, |
|
"learning_rate": 0.00025461187214611874, |
|
"loss": 0.3687, |
|
"step": 810 |
|
}, |
|
{ |
|
"epoch": 36.98, |
|
"eval_loss": 0.6344943046569824, |
|
"eval_runtime": 2.3916, |
|
"eval_samples_per_second": 7.108, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.3504171632896305, |
|
"step": 814 |
|
}, |
|
{ |
|
"epoch": 37.27, |
|
"learning_rate": 0.0002527853881278539, |
|
"loss": 0.4141, |
|
"step": 820 |
|
}, |
|
{ |
|
"epoch": 37.71, |
|
"learning_rate": 0.000250958904109589, |
|
"loss": 0.4119, |
|
"step": 830 |
|
}, |
|
{ |
|
"epoch": 37.98, |
|
"eval_loss": 0.63063645362854, |
|
"eval_runtime": 2.4069, |
|
"eval_samples_per_second": 7.063, |
|
"eval_steps_per_second": 2.077, |
|
"eval_wer": 0.33849821215733017, |
|
"step": 836 |
|
}, |
|
{ |
|
"epoch": 38.18, |
|
"learning_rate": 0.00024913242009132425, |
|
"loss": 0.3937, |
|
"step": 840 |
|
}, |
|
{ |
|
"epoch": 38.62, |
|
"learning_rate": 0.00024730593607305936, |
|
"loss": 0.359, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 38.98, |
|
"eval_loss": 0.6538282036781311, |
|
"eval_runtime": 2.3872, |
|
"eval_samples_per_second": 7.121, |
|
"eval_steps_per_second": 2.095, |
|
"eval_wer": 0.3575685339690107, |
|
"step": 858 |
|
}, |
|
{ |
|
"epoch": 39.09, |
|
"learning_rate": 0.00024547945205479453, |
|
"loss": 0.4018, |
|
"step": 860 |
|
}, |
|
{ |
|
"epoch": 39.53, |
|
"learning_rate": 0.00024365296803652967, |
|
"loss": 0.3544, |
|
"step": 870 |
|
}, |
|
{ |
|
"epoch": 39.98, |
|
"learning_rate": 0.00024182648401826487, |
|
"loss": 0.359, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 39.98, |
|
"eval_loss": 0.661315381526947, |
|
"eval_runtime": 2.3978, |
|
"eval_samples_per_second": 7.09, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.33492252681764006, |
|
"step": 880 |
|
}, |
|
{ |
|
"epoch": 40.44, |
|
"learning_rate": 0.00024, |
|
"loss": 0.3963, |
|
"step": 890 |
|
}, |
|
{ |
|
"epoch": 40.89, |
|
"learning_rate": 0.00023817351598173517, |
|
"loss": 0.3488, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 40.98, |
|
"eval_loss": 0.5976209044456482, |
|
"eval_runtime": 2.3647, |
|
"eval_samples_per_second": 7.189, |
|
"eval_steps_per_second": 2.114, |
|
"eval_wer": 0.3468414779499404, |
|
"step": 902 |
|
}, |
|
{ |
|
"epoch": 41.36, |
|
"learning_rate": 0.00023634703196347031, |
|
"loss": 0.3752, |
|
"step": 910 |
|
}, |
|
{ |
|
"epoch": 41.8, |
|
"learning_rate": 0.0002345205479452055, |
|
"loss": 0.3543, |
|
"step": 920 |
|
}, |
|
{ |
|
"epoch": 41.98, |
|
"eval_loss": 0.6326549053192139, |
|
"eval_runtime": 2.3914, |
|
"eval_samples_per_second": 7.109, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.3432657926102503, |
|
"step": 924 |
|
}, |
|
{ |
|
"epoch": 42.27, |
|
"learning_rate": 0.00023269406392694065, |
|
"loss": 0.3711, |
|
"step": 930 |
|
}, |
|
{ |
|
"epoch": 42.71, |
|
"learning_rate": 0.00023086757990867582, |
|
"loss": 0.3647, |
|
"step": 940 |
|
}, |
|
{ |
|
"epoch": 42.98, |
|
"eval_loss": 0.6207560300827026, |
|
"eval_runtime": 2.3947, |
|
"eval_samples_per_second": 7.099, |
|
"eval_steps_per_second": 2.088, |
|
"eval_wer": 0.3599523241954708, |
|
"step": 946 |
|
}, |
|
{ |
|
"epoch": 43.18, |
|
"learning_rate": 0.00022904109589041096, |
|
"loss": 0.3462, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 43.62, |
|
"learning_rate": 0.00022721461187214615, |
|
"loss": 0.3529, |
|
"step": 960 |
|
}, |
|
{ |
|
"epoch": 43.98, |
|
"eval_loss": 0.6007660627365112, |
|
"eval_runtime": 2.4091, |
|
"eval_samples_per_second": 7.056, |
|
"eval_steps_per_second": 2.075, |
|
"eval_wer": 0.3492252681764005, |
|
"step": 968 |
|
}, |
|
{ |
|
"epoch": 44.09, |
|
"learning_rate": 0.00022538812785388127, |
|
"loss": 0.359, |
|
"step": 970 |
|
}, |
|
{ |
|
"epoch": 44.53, |
|
"learning_rate": 0.00022374429223744296, |
|
"loss": 0.329, |
|
"step": 980 |
|
}, |
|
{ |
|
"epoch": 44.98, |
|
"learning_rate": 0.0002219178082191781, |
|
"loss": 0.3691, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 44.98, |
|
"eval_loss": 0.606474757194519, |
|
"eval_runtime": 2.3824, |
|
"eval_samples_per_second": 7.136, |
|
"eval_steps_per_second": 2.099, |
|
"eval_wer": 0.3492252681764005, |
|
"step": 990 |
|
}, |
|
{ |
|
"epoch": 45.44, |
|
"learning_rate": 0.00022009132420091327, |
|
"loss": 0.3611, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 45.89, |
|
"learning_rate": 0.0002182648401826484, |
|
"loss": 0.329, |
|
"step": 1010 |
|
}, |
|
{ |
|
"epoch": 45.98, |
|
"eval_loss": 0.6287906169891357, |
|
"eval_runtime": 2.4134, |
|
"eval_samples_per_second": 7.044, |
|
"eval_steps_per_second": 2.072, |
|
"eval_wer": 0.33730631704410013, |
|
"step": 1012 |
|
}, |
|
{ |
|
"epoch": 46.36, |
|
"learning_rate": 0.00021643835616438358, |
|
"loss": 0.3256, |
|
"step": 1020 |
|
}, |
|
{ |
|
"epoch": 46.8, |
|
"learning_rate": 0.00021461187214611872, |
|
"loss": 0.3357, |
|
"step": 1030 |
|
}, |
|
{ |
|
"epoch": 46.98, |
|
"eval_loss": 0.5759619474411011, |
|
"eval_runtime": 2.3736, |
|
"eval_samples_per_second": 7.162, |
|
"eval_steps_per_second": 2.107, |
|
"eval_wer": 0.34803337306317045, |
|
"step": 1034 |
|
}, |
|
{ |
|
"epoch": 47.27, |
|
"learning_rate": 0.00021278538812785392, |
|
"loss": 0.3698, |
|
"step": 1040 |
|
}, |
|
{ |
|
"epoch": 47.71, |
|
"learning_rate": 0.00021095890410958906, |
|
"loss": 0.3318, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 47.98, |
|
"eval_loss": 0.5636705160140991, |
|
"eval_runtime": 2.409, |
|
"eval_samples_per_second": 7.057, |
|
"eval_steps_per_second": 2.076, |
|
"eval_wer": 0.3563766388557807, |
|
"step": 1056 |
|
}, |
|
{ |
|
"epoch": 48.18, |
|
"learning_rate": 0.00020913242009132422, |
|
"loss": 0.3573, |
|
"step": 1060 |
|
}, |
|
{ |
|
"epoch": 48.62, |
|
"learning_rate": 0.00020730593607305937, |
|
"loss": 0.3181, |
|
"step": 1070 |
|
}, |
|
{ |
|
"epoch": 48.98, |
|
"eval_loss": 0.5560324192047119, |
|
"eval_runtime": 2.398, |
|
"eval_samples_per_second": 7.089, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.3468414779499404, |
|
"step": 1078 |
|
}, |
|
{ |
|
"epoch": 49.09, |
|
"learning_rate": 0.00020547945205479456, |
|
"loss": 0.3274, |
|
"step": 1080 |
|
}, |
|
{ |
|
"epoch": 49.53, |
|
"learning_rate": 0.0002036529680365297, |
|
"loss": 0.3073, |
|
"step": 1090 |
|
}, |
|
{ |
|
"epoch": 49.98, |
|
"learning_rate": 0.00020182648401826487, |
|
"loss": 0.3313, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 49.98, |
|
"eval_loss": 0.5904679298400879, |
|
"eval_runtime": 2.4923, |
|
"eval_samples_per_second": 6.821, |
|
"eval_steps_per_second": 2.006, |
|
"eval_wer": 0.33373063170441003, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 50.44, |
|
"learning_rate": 0.0002, |
|
"loss": 0.3322, |
|
"step": 1110 |
|
}, |
|
{ |
|
"epoch": 50.89, |
|
"learning_rate": 0.00019817351598173518, |
|
"loss": 0.3059, |
|
"step": 1120 |
|
}, |
|
{ |
|
"epoch": 50.98, |
|
"eval_loss": 0.5443325042724609, |
|
"eval_runtime": 2.3906, |
|
"eval_samples_per_second": 7.111, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.32777115613825986, |
|
"step": 1122 |
|
}, |
|
{ |
|
"epoch": 51.36, |
|
"learning_rate": 0.00019634703196347032, |
|
"loss": 0.2972, |
|
"step": 1130 |
|
}, |
|
{ |
|
"epoch": 51.8, |
|
"learning_rate": 0.00019452054794520549, |
|
"loss": 0.3375, |
|
"step": 1140 |
|
}, |
|
{ |
|
"epoch": 51.98, |
|
"eval_loss": 0.5695326328277588, |
|
"eval_runtime": 2.3779, |
|
"eval_samples_per_second": 7.149, |
|
"eval_steps_per_second": 2.103, |
|
"eval_wer": 0.3575685339690107, |
|
"step": 1144 |
|
}, |
|
{ |
|
"epoch": 52.27, |
|
"learning_rate": 0.00019269406392694065, |
|
"loss": 0.3534, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 52.71, |
|
"learning_rate": 0.00019086757990867582, |
|
"loss": 0.3191, |
|
"step": 1160 |
|
}, |
|
{ |
|
"epoch": 52.98, |
|
"eval_loss": 0.5874321460723877, |
|
"eval_runtime": 2.5159, |
|
"eval_samples_per_second": 6.757, |
|
"eval_steps_per_second": 1.987, |
|
"eval_wer": 0.33849821215733017, |
|
"step": 1166 |
|
}, |
|
{ |
|
"epoch": 53.18, |
|
"learning_rate": 0.00018904109589041096, |
|
"loss": 0.3002, |
|
"step": 1170 |
|
}, |
|
{ |
|
"epoch": 53.62, |
|
"learning_rate": 0.00018721461187214613, |
|
"loss": 0.3115, |
|
"step": 1180 |
|
}, |
|
{ |
|
"epoch": 53.98, |
|
"eval_loss": 0.5264096260070801, |
|
"eval_runtime": 2.4342, |
|
"eval_samples_per_second": 6.984, |
|
"eval_steps_per_second": 2.054, |
|
"eval_wer": 0.3635280095351609, |
|
"step": 1188 |
|
}, |
|
{ |
|
"epoch": 54.09, |
|
"learning_rate": 0.0001853881278538813, |
|
"loss": 0.3044, |
|
"step": 1190 |
|
}, |
|
{ |
|
"epoch": 54.53, |
|
"learning_rate": 0.00018356164383561644, |
|
"loss": 0.3179, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 54.98, |
|
"learning_rate": 0.0001817351598173516, |
|
"loss": 0.3044, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 54.98, |
|
"eval_loss": 0.5479747653007507, |
|
"eval_runtime": 2.3968, |
|
"eval_samples_per_second": 7.093, |
|
"eval_steps_per_second": 2.086, |
|
"eval_wer": 0.3432657926102503, |
|
"step": 1210 |
|
}, |
|
{ |
|
"epoch": 55.44, |
|
"learning_rate": 0.00017990867579908678, |
|
"loss": 0.3619, |
|
"step": 1220 |
|
}, |
|
{ |
|
"epoch": 55.89, |
|
"learning_rate": 0.00017808219178082192, |
|
"loss": 0.3256, |
|
"step": 1230 |
|
}, |
|
{ |
|
"epoch": 55.98, |
|
"eval_loss": 0.5677239894866943, |
|
"eval_runtime": 2.4179, |
|
"eval_samples_per_second": 7.031, |
|
"eval_steps_per_second": 2.068, |
|
"eval_wer": 0.33849821215733017, |
|
"step": 1232 |
|
}, |
|
{ |
|
"epoch": 56.36, |
|
"learning_rate": 0.00017625570776255708, |
|
"loss": 0.3341, |
|
"step": 1240 |
|
}, |
|
{ |
|
"epoch": 56.8, |
|
"learning_rate": 0.00017442922374429225, |
|
"loss": 0.2938, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 56.98, |
|
"eval_loss": 0.5597422122955322, |
|
"eval_runtime": 2.3754, |
|
"eval_samples_per_second": 7.157, |
|
"eval_steps_per_second": 2.105, |
|
"eval_wer": 0.34445768772348034, |
|
"step": 1254 |
|
}, |
|
{ |
|
"epoch": 57.27, |
|
"learning_rate": 0.00017260273972602742, |
|
"loss": 0.297, |
|
"step": 1260 |
|
}, |
|
{ |
|
"epoch": 57.71, |
|
"learning_rate": 0.00017077625570776256, |
|
"loss": 0.2853, |
|
"step": 1270 |
|
}, |
|
{ |
|
"epoch": 57.98, |
|
"eval_loss": 0.5942354798316956, |
|
"eval_runtime": 2.3904, |
|
"eval_samples_per_second": 7.112, |
|
"eval_steps_per_second": 2.092, |
|
"eval_wer": 0.33730631704410013, |
|
"step": 1276 |
|
}, |
|
{ |
|
"epoch": 58.18, |
|
"learning_rate": 0.00016894977168949773, |
|
"loss": 0.3409, |
|
"step": 1280 |
|
}, |
|
{ |
|
"epoch": 58.62, |
|
"learning_rate": 0.0001671232876712329, |
|
"loss": 0.3348, |
|
"step": 1290 |
|
}, |
|
{ |
|
"epoch": 58.98, |
|
"eval_loss": 0.5733348727226257, |
|
"eval_runtime": 2.3901, |
|
"eval_samples_per_second": 7.113, |
|
"eval_steps_per_second": 2.092, |
|
"eval_wer": 0.3420738974970203, |
|
"step": 1298 |
|
}, |
|
{ |
|
"epoch": 59.09, |
|
"learning_rate": 0.00016529680365296804, |
|
"loss": 0.341, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 59.53, |
|
"learning_rate": 0.0001634703196347032, |
|
"loss": 0.3381, |
|
"step": 1310 |
|
}, |
|
{ |
|
"epoch": 59.98, |
|
"learning_rate": 0.00016164383561643837, |
|
"loss": 0.3024, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 59.98, |
|
"eval_loss": 0.5603709816932678, |
|
"eval_runtime": 2.3597, |
|
"eval_samples_per_second": 7.204, |
|
"eval_steps_per_second": 2.119, |
|
"eval_wer": 0.3432657926102503, |
|
"step": 1320 |
|
}, |
|
{ |
|
"epoch": 60.44, |
|
"learning_rate": 0.00015981735159817351, |
|
"loss": 0.2983, |
|
"step": 1330 |
|
}, |
|
{ |
|
"epoch": 60.89, |
|
"learning_rate": 0.00015799086757990868, |
|
"loss": 0.2655, |
|
"step": 1340 |
|
}, |
|
{ |
|
"epoch": 60.98, |
|
"eval_loss": 0.5348132252693176, |
|
"eval_runtime": 2.4009, |
|
"eval_samples_per_second": 7.081, |
|
"eval_steps_per_second": 2.083, |
|
"eval_wer": 0.3468414779499404, |
|
"step": 1342 |
|
}, |
|
{ |
|
"epoch": 61.36, |
|
"learning_rate": 0.00015616438356164385, |
|
"loss": 0.3171, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 61.8, |
|
"learning_rate": 0.00015433789954337902, |
|
"loss": 0.3029, |
|
"step": 1360 |
|
}, |
|
{ |
|
"epoch": 61.98, |
|
"eval_loss": 0.5751910209655762, |
|
"eval_runtime": 2.4011, |
|
"eval_samples_per_second": 7.08, |
|
"eval_steps_per_second": 2.082, |
|
"eval_wer": 0.3206197854588796, |
|
"step": 1364 |
|
}, |
|
{ |
|
"epoch": 62.27, |
|
"learning_rate": 0.00015251141552511416, |
|
"loss": 0.2712, |
|
"step": 1370 |
|
}, |
|
{ |
|
"epoch": 62.71, |
|
"learning_rate": 0.00015068493150684933, |
|
"loss": 0.3435, |
|
"step": 1380 |
|
}, |
|
{ |
|
"epoch": 62.98, |
|
"eval_loss": 0.548903226852417, |
|
"eval_runtime": 2.3839, |
|
"eval_samples_per_second": 7.131, |
|
"eval_steps_per_second": 2.097, |
|
"eval_wer": 0.3063170441001192, |
|
"step": 1386 |
|
}, |
|
{ |
|
"epoch": 63.18, |
|
"learning_rate": 0.0001488584474885845, |
|
"loss": 0.2861, |
|
"step": 1390 |
|
}, |
|
{ |
|
"epoch": 63.62, |
|
"learning_rate": 0.00014703196347031963, |
|
"loss": 0.3125, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 63.98, |
|
"eval_loss": 0.5735787749290466, |
|
"eval_runtime": 2.5834, |
|
"eval_samples_per_second": 6.581, |
|
"eval_steps_per_second": 1.935, |
|
"eval_wer": 0.3075089392133492, |
|
"step": 1408 |
|
}, |
|
{ |
|
"epoch": 64.09, |
|
"learning_rate": 0.0001452054794520548, |
|
"loss": 0.2642, |
|
"step": 1410 |
|
}, |
|
{ |
|
"epoch": 64.53, |
|
"learning_rate": 0.00014337899543378997, |
|
"loss": 0.3023, |
|
"step": 1420 |
|
}, |
|
{ |
|
"epoch": 64.98, |
|
"learning_rate": 0.0001415525114155251, |
|
"loss": 0.263, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 64.98, |
|
"eval_loss": 0.5505235195159912, |
|
"eval_runtime": 2.3983, |
|
"eval_samples_per_second": 7.088, |
|
"eval_steps_per_second": 2.085, |
|
"eval_wer": 0.3206197854588796, |
|
"step": 1430 |
|
}, |
|
{ |
|
"epoch": 65.44, |
|
"learning_rate": 0.00013972602739726028, |
|
"loss": 0.2621, |
|
"step": 1440 |
|
}, |
|
{ |
|
"epoch": 65.89, |
|
"learning_rate": 0.00013789954337899545, |
|
"loss": 0.2665, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 65.98, |
|
"eval_loss": 0.5391205549240112, |
|
"eval_runtime": 2.4038, |
|
"eval_samples_per_second": 7.072, |
|
"eval_steps_per_second": 2.08, |
|
"eval_wer": 0.3230035756853397, |
|
"step": 1452 |
|
}, |
|
{ |
|
"epoch": 66.36, |
|
"learning_rate": 0.00013607305936073061, |
|
"loss": 0.3153, |
|
"step": 1460 |
|
}, |
|
{ |
|
"epoch": 66.8, |
|
"learning_rate": 0.00013424657534246576, |
|
"loss": 0.299, |
|
"step": 1470 |
|
}, |
|
{ |
|
"epoch": 66.98, |
|
"eval_loss": 0.5388566255569458, |
|
"eval_runtime": 2.4363, |
|
"eval_samples_per_second": 6.978, |
|
"eval_steps_per_second": 2.052, |
|
"eval_wer": 0.3134684147794994, |
|
"step": 1474 |
|
}, |
|
{ |
|
"epoch": 67.27, |
|
"learning_rate": 0.00013242009132420092, |
|
"loss": 0.2966, |
|
"step": 1480 |
|
}, |
|
{ |
|
"epoch": 67.71, |
|
"learning_rate": 0.0001305936073059361, |
|
"loss": 0.2909, |
|
"step": 1490 |
|
}, |
|
{ |
|
"epoch": 67.98, |
|
"eval_loss": 0.5840542912483215, |
|
"eval_runtime": 2.4363, |
|
"eval_samples_per_second": 6.978, |
|
"eval_steps_per_second": 2.052, |
|
"eval_wer": 0.3098927294398093, |
|
"step": 1496 |
|
}, |
|
{ |
|
"epoch": 68.18, |
|
"learning_rate": 0.00012876712328767123, |
|
"loss": 0.2848, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 68.62, |
|
"learning_rate": 0.0001269406392694064, |
|
"loss": 0.2988, |
|
"step": 1510 |
|
}, |
|
{ |
|
"epoch": 68.98, |
|
"eval_loss": 0.5847446918487549, |
|
"eval_runtime": 2.3833, |
|
"eval_samples_per_second": 7.133, |
|
"eval_steps_per_second": 2.098, |
|
"eval_wer": 0.300357568533969, |
|
"step": 1518 |
|
}, |
|
{ |
|
"epoch": 69.09, |
|
"learning_rate": 0.00012511415525114157, |
|
"loss": 0.2722, |
|
"step": 1520 |
|
}, |
|
{ |
|
"epoch": 69.53, |
|
"learning_rate": 0.0001232876712328767, |
|
"loss": 0.2633, |
|
"step": 1530 |
|
}, |
|
{ |
|
"epoch": 69.98, |
|
"learning_rate": 0.00012146118721461188, |
|
"loss": 0.2879, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 69.98, |
|
"eval_loss": 0.5940636992454529, |
|
"eval_runtime": 2.6597, |
|
"eval_samples_per_second": 6.392, |
|
"eval_steps_per_second": 1.88, |
|
"eval_wer": 0.2967818831942789, |
|
"step": 1540 |
|
}, |
|
{ |
|
"epoch": 70.44, |
|
"learning_rate": 0.00011963470319634704, |
|
"loss": 0.2957, |
|
"step": 1550 |
|
}, |
|
{ |
|
"epoch": 70.89, |
|
"learning_rate": 0.0001178082191780822, |
|
"loss": 0.2802, |
|
"step": 1560 |
|
}, |
|
{ |
|
"epoch": 70.98, |
|
"eval_loss": 0.6612115502357483, |
|
"eval_runtime": 2.3737, |
|
"eval_samples_per_second": 7.162, |
|
"eval_steps_per_second": 2.106, |
|
"eval_wer": 0.29201430274135876, |
|
"step": 1562 |
|
}, |
|
{ |
|
"epoch": 71.36, |
|
"learning_rate": 0.00011598173515981737, |
|
"loss": 0.2691, |
|
"step": 1570 |
|
}, |
|
{ |
|
"epoch": 71.8, |
|
"learning_rate": 0.00011415525114155252, |
|
"loss": 0.2877, |
|
"step": 1580 |
|
}, |
|
{ |
|
"epoch": 71.98, |
|
"eval_loss": 0.5640567541122437, |
|
"eval_runtime": 2.4284, |
|
"eval_samples_per_second": 7.0, |
|
"eval_steps_per_second": 2.059, |
|
"eval_wer": 0.30512514898688914, |
|
"step": 1584 |
|
}, |
|
{ |
|
"epoch": 72.27, |
|
"learning_rate": 0.00011232876712328768, |
|
"loss": 0.295, |
|
"step": 1590 |
|
}, |
|
{ |
|
"epoch": 72.71, |
|
"learning_rate": 0.00011050228310502284, |
|
"loss": 0.2727, |
|
"step": 1600 |
|
}, |
|
{ |
|
"epoch": 72.98, |
|
"eval_loss": 0.6137677431106567, |
|
"eval_runtime": 2.404, |
|
"eval_samples_per_second": 7.072, |
|
"eval_steps_per_second": 2.08, |
|
"eval_wer": 0.3063170441001192, |
|
"step": 1606 |
|
}, |
|
{ |
|
"epoch": 73.18, |
|
"learning_rate": 0.000108675799086758, |
|
"loss": 0.2989, |
|
"step": 1610 |
|
}, |
|
{ |
|
"epoch": 73.62, |
|
"learning_rate": 0.00010684931506849317, |
|
"loss": 0.2668, |
|
"step": 1620 |
|
}, |
|
{ |
|
"epoch": 73.98, |
|
"eval_loss": 0.608710527420044, |
|
"eval_runtime": 2.3622, |
|
"eval_samples_per_second": 7.197, |
|
"eval_steps_per_second": 2.117, |
|
"eval_wer": 0.29201430274135876, |
|
"step": 1628 |
|
}, |
|
{ |
|
"epoch": 74.09, |
|
"learning_rate": 0.00010502283105022832, |
|
"loss": 0.2665, |
|
"step": 1630 |
|
}, |
|
{ |
|
"epoch": 74.53, |
|
"learning_rate": 0.00010319634703196347, |
|
"loss": 0.2829, |
|
"step": 1640 |
|
}, |
|
{ |
|
"epoch": 74.98, |
|
"learning_rate": 0.00010136986301369864, |
|
"loss": 0.2675, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 74.98, |
|
"eval_loss": 0.5876254439353943, |
|
"eval_runtime": 2.3686, |
|
"eval_samples_per_second": 7.177, |
|
"eval_steps_per_second": 2.111, |
|
"eval_wer": 0.2932061978545888, |
|
"step": 1650 |
|
}, |
|
{ |
|
"epoch": 75.44, |
|
"learning_rate": 9.954337899543378e-05, |
|
"loss": 0.2837, |
|
"step": 1660 |
|
}, |
|
{ |
|
"epoch": 75.89, |
|
"learning_rate": 9.771689497716895e-05, |
|
"loss": 0.264, |
|
"step": 1670 |
|
}, |
|
{ |
|
"epoch": 75.98, |
|
"eval_loss": 0.6042572855949402, |
|
"eval_runtime": 2.4461, |
|
"eval_samples_per_second": 6.95, |
|
"eval_steps_per_second": 2.044, |
|
"eval_wer": 0.29797377830750893, |
|
"step": 1672 |
|
}, |
|
{ |
|
"epoch": 76.36, |
|
"learning_rate": 9.58904109589041e-05, |
|
"loss": 0.2769, |
|
"step": 1680 |
|
}, |
|
{ |
|
"epoch": 76.8, |
|
"learning_rate": 9.406392694063927e-05, |
|
"loss": 0.2352, |
|
"step": 1690 |
|
}, |
|
{ |
|
"epoch": 76.98, |
|
"eval_loss": 0.5829324722290039, |
|
"eval_runtime": 2.389, |
|
"eval_samples_per_second": 7.116, |
|
"eval_steps_per_second": 2.093, |
|
"eval_wer": 0.2932061978545888, |
|
"step": 1694 |
|
}, |
|
{ |
|
"epoch": 77.27, |
|
"learning_rate": 9.223744292237443e-05, |
|
"loss": 0.364, |
|
"step": 1700 |
|
}, |
|
{ |
|
"epoch": 77.71, |
|
"learning_rate": 9.041095890410958e-05, |
|
"loss": 0.2494, |
|
"step": 1710 |
|
}, |
|
{ |
|
"epoch": 77.98, |
|
"eval_loss": 0.5775046348571777, |
|
"eval_runtime": 2.3916, |
|
"eval_samples_per_second": 7.108, |
|
"eval_steps_per_second": 2.091, |
|
"eval_wer": 0.3063170441001192, |
|
"step": 1716 |
|
}, |
|
{ |
|
"epoch": 78.18, |
|
"learning_rate": 8.858447488584475e-05, |
|
"loss": 0.2494, |
|
"step": 1720 |
|
}, |
|
{ |
|
"epoch": 78.62, |
|
"learning_rate": 8.67579908675799e-05, |
|
"loss": 0.2621, |
|
"step": 1730 |
|
}, |
|
{ |
|
"epoch": 78.98, |
|
"eval_loss": 0.5675925016403198, |
|
"eval_runtime": 2.402, |
|
"eval_samples_per_second": 7.078, |
|
"eval_steps_per_second": 2.082, |
|
"eval_wer": 0.29558998808104886, |
|
"step": 1738 |
|
}, |
|
{ |
|
"epoch": 79.09, |
|
"learning_rate": 8.493150684931507e-05, |
|
"loss": 0.286, |
|
"step": 1740 |
|
}, |
|
{ |
|
"epoch": 79.53, |
|
"learning_rate": 8.310502283105023e-05, |
|
"loss": 0.2591, |
|
"step": 1750 |
|
}, |
|
{ |
|
"epoch": 79.98, |
|
"learning_rate": 8.127853881278538e-05, |
|
"loss": 0.2788, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 79.98, |
|
"eval_loss": 0.5863559246063232, |
|
"eval_runtime": 2.3868, |
|
"eval_samples_per_second": 7.122, |
|
"eval_steps_per_second": 2.095, |
|
"eval_wer": 0.2932061978545888, |
|
"step": 1760 |
|
}, |
|
{ |
|
"epoch": 80.44, |
|
"learning_rate": 7.945205479452055e-05, |
|
"loss": 0.2591, |
|
"step": 1770 |
|
}, |
|
{ |
|
"epoch": 80.89, |
|
"learning_rate": 7.76255707762557e-05, |
|
"loss": 0.2615, |
|
"step": 1780 |
|
}, |
|
{ |
|
"epoch": 80.98, |
|
"eval_loss": 0.5754203796386719, |
|
"eval_runtime": 2.4207, |
|
"eval_samples_per_second": 7.023, |
|
"eval_steps_per_second": 2.066, |
|
"eval_wer": 0.30154946364719903, |
|
"step": 1782 |
|
}, |
|
{ |
|
"epoch": 81.36, |
|
"learning_rate": 7.579908675799087e-05, |
|
"loss": 0.307, |
|
"step": 1790 |
|
}, |
|
{ |
|
"epoch": 81.8, |
|
"learning_rate": 7.397260273972603e-05, |
|
"loss": 0.2542, |
|
"step": 1800 |
|
}, |
|
{ |
|
"epoch": 81.98, |
|
"eval_loss": 0.5650574564933777, |
|
"eval_runtime": 2.403, |
|
"eval_samples_per_second": 7.075, |
|
"eval_steps_per_second": 2.081, |
|
"eval_wer": 0.30274135876042907, |
|
"step": 1804 |
|
}, |
|
{ |
|
"epoch": 82.27, |
|
"learning_rate": 7.21461187214612e-05, |
|
"loss": 0.2653, |
|
"step": 1810 |
|
}, |
|
{ |
|
"epoch": 82.71, |
|
"learning_rate": 7.031963470319635e-05, |
|
"loss": 0.2641, |
|
"step": 1820 |
|
}, |
|
{ |
|
"epoch": 82.98, |
|
"eval_loss": 0.5731266736984253, |
|
"eval_runtime": 2.415, |
|
"eval_samples_per_second": 7.039, |
|
"eval_steps_per_second": 2.07, |
|
"eval_wer": 0.300357568533969, |
|
"step": 1826 |
|
}, |
|
{ |
|
"epoch": 83.18, |
|
"learning_rate": 6.84931506849315e-05, |
|
"loss": 0.2684, |
|
"step": 1830 |
|
}, |
|
{ |
|
"epoch": 83.62, |
|
"learning_rate": 6.666666666666667e-05, |
|
"loss": 0.2532, |
|
"step": 1840 |
|
}, |
|
{ |
|
"epoch": 83.98, |
|
"eval_loss": 0.5782402157783508, |
|
"eval_runtime": 2.4968, |
|
"eval_samples_per_second": 6.809, |
|
"eval_steps_per_second": 2.003, |
|
"eval_wer": 0.2967818831942789, |
|
"step": 1848 |
|
}, |
|
{ |
|
"epoch": 84.09, |
|
"learning_rate": 6.484018264840182e-05, |
|
"loss": 0.2521, |
|
"step": 1850 |
|
}, |
|
{ |
|
"epoch": 84.53, |
|
"learning_rate": 6.301369863013699e-05, |
|
"loss": 0.231, |
|
"step": 1860 |
|
}, |
|
{ |
|
"epoch": 84.98, |
|
"learning_rate": 6.118721461187215e-05, |
|
"loss": 0.2645, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 84.98, |
|
"eval_loss": 0.5717898011207581, |
|
"eval_runtime": 2.3947, |
|
"eval_samples_per_second": 7.099, |
|
"eval_steps_per_second": 2.088, |
|
"eval_wer": 0.3039332538736591, |
|
"step": 1870 |
|
}, |
|
{ |
|
"epoch": 85.44, |
|
"learning_rate": 5.936073059360731e-05, |
|
"loss": 0.2731, |
|
"step": 1880 |
|
}, |
|
{ |
|
"epoch": 85.89, |
|
"learning_rate": 5.753424657534247e-05, |
|
"loss": 0.2296, |
|
"step": 1890 |
|
}, |
|
{ |
|
"epoch": 85.98, |
|
"eval_loss": 0.5628200173377991, |
|
"eval_runtime": 2.3724, |
|
"eval_samples_per_second": 7.166, |
|
"eval_steps_per_second": 2.108, |
|
"eval_wer": 0.3146603098927294, |
|
"step": 1892 |
|
}, |
|
{ |
|
"epoch": 86.36, |
|
"learning_rate": 5.570776255707762e-05, |
|
"loss": 0.2658, |
|
"step": 1900 |
|
}, |
|
{ |
|
"epoch": 86.8, |
|
"learning_rate": 5.3881278538812784e-05, |
|
"loss": 0.2394, |
|
"step": 1910 |
|
}, |
|
{ |
|
"epoch": 86.98, |
|
"eval_loss": 0.5919906497001648, |
|
"eval_runtime": 2.3939, |
|
"eval_samples_per_second": 7.101, |
|
"eval_steps_per_second": 2.089, |
|
"eval_wer": 0.30274135876042907, |
|
"step": 1914 |
|
}, |
|
{ |
|
"epoch": 87.27, |
|
"learning_rate": 5.2054794520547945e-05, |
|
"loss": 0.2777, |
|
"step": 1920 |
|
}, |
|
{ |
|
"epoch": 87.71, |
|
"learning_rate": 5.0228310502283106e-05, |
|
"loss": 0.2636, |
|
"step": 1930 |
|
}, |
|
{ |
|
"epoch": 87.98, |
|
"eval_loss": 0.6085216999053955, |
|
"eval_runtime": 2.389, |
|
"eval_samples_per_second": 7.116, |
|
"eval_steps_per_second": 2.093, |
|
"eval_wer": 0.2967818831942789, |
|
"step": 1936 |
|
}, |
|
{ |
|
"epoch": 88.18, |
|
"learning_rate": 4.840182648401827e-05, |
|
"loss": 0.2717, |
|
"step": 1940 |
|
}, |
|
{ |
|
"epoch": 88.62, |
|
"learning_rate": 4.657534246575342e-05, |
|
"loss": 0.2371, |
|
"step": 1950 |
|
}, |
|
{ |
|
"epoch": 88.98, |
|
"eval_loss": 0.5808770060539246, |
|
"eval_runtime": 2.3775, |
|
"eval_samples_per_second": 7.15, |
|
"eval_steps_per_second": 2.103, |
|
"eval_wer": 0.3075089392133492, |
|
"step": 1958 |
|
}, |
|
{ |
|
"epoch": 89.09, |
|
"learning_rate": 4.474885844748858e-05, |
|
"loss": 0.2616, |
|
"step": 1960 |
|
}, |
|
{ |
|
"epoch": 89.53, |
|
"learning_rate": 4.2922374429223744e-05, |
|
"loss": 0.2526, |
|
"step": 1970 |
|
}, |
|
{ |
|
"epoch": 89.98, |
|
"learning_rate": 4.1095890410958905e-05, |
|
"loss": 0.2364, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 89.98, |
|
"eval_loss": 0.592748761177063, |
|
"eval_runtime": 2.3694, |
|
"eval_samples_per_second": 7.175, |
|
"eval_steps_per_second": 2.11, |
|
"eval_wer": 0.3039332538736591, |
|
"step": 1980 |
|
}, |
|
{ |
|
"epoch": 90.44, |
|
"learning_rate": 3.9269406392694066e-05, |
|
"loss": 0.2658, |
|
"step": 1990 |
|
}, |
|
{ |
|
"epoch": 90.89, |
|
"learning_rate": 3.744292237442922e-05, |
|
"loss": 0.2812, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 90.98, |
|
"eval_loss": 0.5713174343109131, |
|
"eval_runtime": 2.3963, |
|
"eval_samples_per_second": 7.094, |
|
"eval_steps_per_second": 2.087, |
|
"eval_wer": 0.31227651966626935, |
|
"step": 2002 |
|
}, |
|
{ |
|
"epoch": 91.36, |
|
"learning_rate": 3.561643835616438e-05, |
|
"loss": 0.2737, |
|
"step": 2010 |
|
}, |
|
{ |
|
"epoch": 91.8, |
|
"learning_rate": 3.378995433789954e-05, |
|
"loss": 0.2141, |
|
"step": 2020 |
|
}, |
|
{ |
|
"epoch": 91.98, |
|
"eval_loss": 0.5742636919021606, |
|
"eval_runtime": 2.3763, |
|
"eval_samples_per_second": 7.154, |
|
"eval_steps_per_second": 2.104, |
|
"eval_wer": 0.3039332538736591, |
|
"step": 2024 |
|
}, |
|
{ |
|
"epoch": 92.27, |
|
"learning_rate": 3.1963470319634704e-05, |
|
"loss": 0.2495, |
|
"step": 2030 |
|
}, |
|
{ |
|
"epoch": 92.71, |
|
"learning_rate": 3.0136986301369862e-05, |
|
"loss": 0.2919, |
|
"step": 2040 |
|
}, |
|
{ |
|
"epoch": 92.98, |
|
"eval_loss": 0.5836674571037292, |
|
"eval_runtime": 2.3949, |
|
"eval_samples_per_second": 7.098, |
|
"eval_steps_per_second": 2.088, |
|
"eval_wer": 0.3063170441001192, |
|
"step": 2046 |
|
}, |
|
{ |
|
"epoch": 93.18, |
|
"learning_rate": 2.8310502283105023e-05, |
|
"loss": 0.2553, |
|
"step": 2050 |
|
}, |
|
{ |
|
"epoch": 93.62, |
|
"learning_rate": 2.6484018264840184e-05, |
|
"loss": 0.2288, |
|
"step": 2060 |
|
}, |
|
{ |
|
"epoch": 93.98, |
|
"eval_loss": 0.5859872102737427, |
|
"eval_runtime": 2.3921, |
|
"eval_samples_per_second": 7.107, |
|
"eval_steps_per_second": 2.09, |
|
"eval_wer": 0.30154946364719903, |
|
"step": 2068 |
|
}, |
|
{ |
|
"epoch": 94.09, |
|
"learning_rate": 2.4657534246575342e-05, |
|
"loss": 0.2978, |
|
"step": 2070 |
|
}, |
|
{ |
|
"epoch": 94.53, |
|
"learning_rate": 2.2831050228310503e-05, |
|
"loss": 0.2202, |
|
"step": 2080 |
|
}, |
|
{ |
|
"epoch": 94.98, |
|
"learning_rate": 2.100456621004566e-05, |
|
"loss": 0.2585, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 94.98, |
|
"eval_loss": 0.5775898098945618, |
|
"eval_runtime": 2.4084, |
|
"eval_samples_per_second": 7.059, |
|
"eval_steps_per_second": 2.076, |
|
"eval_wer": 0.3146603098927294, |
|
"step": 2090 |
|
}, |
|
{ |
|
"epoch": 95.44, |
|
"learning_rate": 1.9178082191780822e-05, |
|
"loss": 0.2722, |
|
"step": 2100 |
|
}, |
|
{ |
|
"epoch": 95.89, |
|
"learning_rate": 1.7351598173515983e-05, |
|
"loss": 0.2529, |
|
"step": 2110 |
|
}, |
|
{ |
|
"epoch": 95.98, |
|
"eval_loss": 0.5624773502349854, |
|
"eval_runtime": 2.4174, |
|
"eval_samples_per_second": 7.032, |
|
"eval_steps_per_second": 2.068, |
|
"eval_wer": 0.31585220500595945, |
|
"step": 2112 |
|
}, |
|
{ |
|
"epoch": 96.36, |
|
"learning_rate": 1.552511415525114e-05, |
|
"loss": 0.2285, |
|
"step": 2120 |
|
}, |
|
{ |
|
"epoch": 96.8, |
|
"learning_rate": 1.3698630136986302e-05, |
|
"loss": 0.2343, |
|
"step": 2130 |
|
}, |
|
{ |
|
"epoch": 96.98, |
|
"eval_loss": 0.5700345039367676, |
|
"eval_runtime": 2.4887, |
|
"eval_samples_per_second": 6.831, |
|
"eval_steps_per_second": 2.009, |
|
"eval_wer": 0.30870083432657924, |
|
"step": 2134 |
|
}, |
|
{ |
|
"epoch": 97.27, |
|
"learning_rate": 1.1872146118721461e-05, |
|
"loss": 0.2933, |
|
"step": 2140 |
|
}, |
|
{ |
|
"epoch": 97.71, |
|
"learning_rate": 1.004566210045662e-05, |
|
"loss": 0.2567, |
|
"step": 2150 |
|
}, |
|
{ |
|
"epoch": 97.98, |
|
"eval_loss": 0.5728834867477417, |
|
"eval_runtime": 2.4553, |
|
"eval_samples_per_second": 6.924, |
|
"eval_steps_per_second": 2.036, |
|
"eval_wer": 0.30870083432657924, |
|
"step": 2156 |
|
}, |
|
{ |
|
"epoch": 98.18, |
|
"learning_rate": 8.21917808219178e-06, |
|
"loss": 0.2421, |
|
"step": 2160 |
|
}, |
|
{ |
|
"epoch": 98.62, |
|
"learning_rate": 6.39269406392694e-06, |
|
"loss": 0.2448, |
|
"step": 2170 |
|
}, |
|
{ |
|
"epoch": 98.98, |
|
"eval_loss": 0.5728344917297363, |
|
"eval_runtime": 2.4689, |
|
"eval_samples_per_second": 6.886, |
|
"eval_steps_per_second": 2.025, |
|
"eval_wer": 0.3110846245530393, |
|
"step": 2178 |
|
}, |
|
{ |
|
"epoch": 99.09, |
|
"learning_rate": 4.566210045662101e-06, |
|
"loss": 0.2497, |
|
"step": 2180 |
|
}, |
|
{ |
|
"epoch": 99.53, |
|
"learning_rate": 2.7397260273972604e-06, |
|
"loss": 0.2456, |
|
"step": 2190 |
|
}, |
|
{ |
|
"epoch": 99.98, |
|
"learning_rate": 9.132420091324201e-07, |
|
"loss": 0.2501, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 99.98, |
|
"eval_loss": 0.5744234323501587, |
|
"eval_runtime": 2.4546, |
|
"eval_samples_per_second": 6.926, |
|
"eval_steps_per_second": 2.037, |
|
"eval_wer": 0.3098927294398093, |
|
"step": 2200 |
|
}, |
|
{ |
|
"epoch": 99.98, |
|
"step": 2200, |
|
"total_flos": 1.783157786698179e+18, |
|
"train_loss": 0.5098017852956599, |
|
"train_runtime": 3993.9176, |
|
"train_samples_per_second": 4.482, |
|
"train_steps_per_second": 0.551 |
|
} |
|
], |
|
"max_steps": 2200, |
|
"num_train_epochs": 100, |
|
"total_flos": 1.783157786698179e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|