| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "global_step": 25780, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00024390399999999997, | |
| "loss": 10.7678, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 3.0923821926116943, | |
| "eval_runtime": 369.7958, | |
| "eval_samples_per_second": 14.681, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0004113554106910039, | |
| "loss": 1.4821, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.4990185499191284, | |
| "eval_runtime": 398.0715, | |
| "eval_samples_per_second": 13.638, | |
| "eval_wer": 0.6425968693867077, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0004093410691003911, | |
| "loss": 0.4925, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 0.3599244952201843, | |
| "eval_runtime": 395.2724, | |
| "eval_samples_per_second": 13.735, | |
| "eval_wer": 0.522966384398255, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00040732672750977834, | |
| "loss": 0.3926, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 0.27538031339645386, | |
| "eval_runtime": 370.4342, | |
| "eval_samples_per_second": 14.656, | |
| "eval_wer": 0.4267898383371824, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.00040531238591916556, | |
| "loss": 0.3404, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 0.25353798270225525, | |
| "eval_runtime": 396.0187, | |
| "eval_samples_per_second": 13.709, | |
| "eval_wer": 0.3899409802412112, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0004032980443285528, | |
| "loss": 0.3148, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 0.23486939072608948, | |
| "eval_runtime": 394.9332, | |
| "eval_samples_per_second": 13.747, | |
| "eval_wer": 0.38388503977418525, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00040128370273794, | |
| "loss": 0.2853, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 0.20920990407466888, | |
| "eval_runtime": 393.4003, | |
| "eval_samples_per_second": 13.8, | |
| "eval_wer": 0.32840646651270206, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00039926936114732727, | |
| "loss": 0.2404, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 0.19739878177642822, | |
| "eval_runtime": 367.4384, | |
| "eval_samples_per_second": 14.775, | |
| "eval_wer": 0.3072106748781114, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00039725501955671443, | |
| "loss": 0.2325, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_loss": 0.19151932001113892, | |
| "eval_runtime": 398.5555, | |
| "eval_samples_per_second": 13.622, | |
| "eval_wer": 0.3065434949961509, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00039524067796610165, | |
| "loss": 0.2325, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "eval_loss": 0.18169228732585907, | |
| "eval_runtime": 393.6421, | |
| "eval_samples_per_second": 13.792, | |
| "eval_wer": 0.2896587118296125, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.00039322633637548887, | |
| "loss": 0.2271, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.1825951784849167, | |
| "eval_runtime": 368.1876, | |
| "eval_samples_per_second": 14.745, | |
| "eval_wer": 0.29001796253528356, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00039121199478487614, | |
| "loss": 0.2263, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_loss": 0.1739877462387085, | |
| "eval_runtime": 397.804, | |
| "eval_samples_per_second": 13.647, | |
| "eval_wer": 0.2704131383115217, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00038919765319426336, | |
| "loss": 0.2187, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 0.16721388697624207, | |
| "eval_runtime": 396.6468, | |
| "eval_samples_per_second": 13.687, | |
| "eval_wer": 0.2596869386707724, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.0003871833116036506, | |
| "loss": 0.2015, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 0.1631636619567871, | |
| "eval_runtime": 391.6354, | |
| "eval_samples_per_second": 13.862, | |
| "eval_wer": 0.25517064408519374, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0003851689700130378, | |
| "loss": 0.1792, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.17116306722164154, | |
| "eval_runtime": 368.9065, | |
| "eval_samples_per_second": 14.716, | |
| "eval_wer": 0.26646138054914037, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.000383154628422425, | |
| "loss": 0.1773, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "eval_loss": 0.16152945160865784, | |
| "eval_runtime": 398.4133, | |
| "eval_samples_per_second": 13.627, | |
| "eval_wer": 0.24146779574031307, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 0.00038114028683181223, | |
| "loss": 0.179, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.15521417558193207, | |
| "eval_runtime": 396.2358, | |
| "eval_samples_per_second": 13.701, | |
| "eval_wer": 0.24613805491403642, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 0.00037912594524119945, | |
| "loss": 0.1717, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "eval_loss": 0.15384772419929504, | |
| "eval_runtime": 393.2934, | |
| "eval_samples_per_second": 13.804, | |
| "eval_wer": 0.2374133949191686, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.0003771116036505867, | |
| "loss": 0.1744, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 0.14799486100673676, | |
| "eval_runtime": 370.8447, | |
| "eval_samples_per_second": 14.64, | |
| "eval_wer": 0.23828586091865536, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 0.0003750972620599739, | |
| "loss": 0.166, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_loss": 0.1445951759815216, | |
| "eval_runtime": 393.1105, | |
| "eval_samples_per_second": 13.81, | |
| "eval_wer": 0.22992045162945857, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 0.0003730829204693611, | |
| "loss": 0.1532, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "eval_loss": 0.1482468694448471, | |
| "eval_runtime": 395.8029, | |
| "eval_samples_per_second": 13.716, | |
| "eval_wer": 0.22037464716448552, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.00037106857887874833, | |
| "loss": 0.142, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_loss": 0.15104271471500397, | |
| "eval_runtime": 369.3329, | |
| "eval_samples_per_second": 14.699, | |
| "eval_wer": 0.228175519630485, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "learning_rate": 0.0003690542372881356, | |
| "loss": 0.1406, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "eval_loss": 0.1345045417547226, | |
| "eval_runtime": 401.0837, | |
| "eval_samples_per_second": 13.536, | |
| "eval_wer": 0.20795483705414422, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.0003670398956975228, | |
| "loss": 0.1458, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "eval_loss": 0.13937360048294067, | |
| "eval_runtime": 390.779, | |
| "eval_samples_per_second": 13.893, | |
| "eval_wer": 0.2111367718758019, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.00036502555410691004, | |
| "loss": 0.1449, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "eval_loss": 0.14066852629184723, | |
| "eval_runtime": 395.9112, | |
| "eval_samples_per_second": 13.713, | |
| "eval_wer": 0.20846805234795998, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 0.00036301121251629726, | |
| "loss": 0.137, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "eval_loss": 0.1394476294517517, | |
| "eval_runtime": 367.4977, | |
| "eval_samples_per_second": 14.773, | |
| "eval_wer": 0.20744162176032846, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 0.0003609968709256845, | |
| "loss": 0.1388, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "eval_loss": 0.1351165473461151, | |
| "eval_runtime": 397.2254, | |
| "eval_samples_per_second": 13.667, | |
| "eval_wer": 0.19964074929432898, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 0.0003589825293350717, | |
| "loss": 0.1307, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "eval_loss": 0.1401081681251526, | |
| "eval_runtime": 396.6542, | |
| "eval_samples_per_second": 13.687, | |
| "eval_wer": 0.20446497305619707, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 0.0003569681877444589, | |
| "loss": 0.122, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "eval_loss": 0.14438064396381378, | |
| "eval_runtime": 389.1513, | |
| "eval_samples_per_second": 13.951, | |
| "eval_wer": 0.20369515011547343, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0003549538461538462, | |
| "loss": 0.1193, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_loss": 0.14555728435516357, | |
| "eval_runtime": 367.57, | |
| "eval_samples_per_second": 14.77, | |
| "eval_wer": 0.21524249422632794, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 0.00035293950456323335, | |
| "loss": 0.1344, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "eval_loss": 0.1305903196334839, | |
| "eval_runtime": 395.5909, | |
| "eval_samples_per_second": 13.724, | |
| "eval_wer": 0.19522709776751349, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.00035092516297262057, | |
| "loss": 0.1267, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "eval_loss": 0.1303185522556305, | |
| "eval_runtime": 392.9371, | |
| "eval_samples_per_second": 13.816, | |
| "eval_wer": 0.18963305106492173, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 0.0003489108213820078, | |
| "loss": 0.1214, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 0.1294514238834381, | |
| "eval_runtime": 369.1477, | |
| "eval_samples_per_second": 14.707, | |
| "eval_wer": 0.19240441365152683, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 0.00034689647979139506, | |
| "loss": 0.1179, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_loss": 0.12323788553476334, | |
| "eval_runtime": 391.972, | |
| "eval_samples_per_second": 13.85, | |
| "eval_wer": 0.19183987682832948, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 0.0003448821382007823, | |
| "loss": 0.0998, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "eval_loss": 0.12908631563186646, | |
| "eval_runtime": 393.594, | |
| "eval_samples_per_second": 13.793, | |
| "eval_wer": 0.18516807800872467, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 0.0003428677966101695, | |
| "loss": 0.1017, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "eval_loss": 0.1282510757446289, | |
| "eval_runtime": 393.5495, | |
| "eval_samples_per_second": 13.795, | |
| "eval_wer": 0.1841929689504747, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 0.00034085345501955666, | |
| "loss": 0.1024, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "eval_loss": 0.12217988073825836, | |
| "eval_runtime": 368.9647, | |
| "eval_samples_per_second": 14.714, | |
| "eval_wer": 0.1739286630741596, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "learning_rate": 0.00033883911342894393, | |
| "loss": 0.1042, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 5.53, | |
| "eval_loss": 0.1310078352689743, | |
| "eval_runtime": 394.8352, | |
| "eval_samples_per_second": 13.75, | |
| "eval_wer": 0.18137028483448805, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "learning_rate": 0.00033682477183833115, | |
| "loss": 0.1096, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 5.67, | |
| "eval_loss": 0.12405824661254883, | |
| "eval_runtime": 392.553, | |
| "eval_samples_per_second": 13.83, | |
| "eval_wer": 0.17890685142417245, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "learning_rate": 0.0003348171447196871, | |
| "loss": 0.1091, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.82, | |
| "eval_loss": 0.12054669111967087, | |
| "eval_runtime": 396.4385, | |
| "eval_samples_per_second": 13.694, | |
| "eval_wer": 0.17942006671798819, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 0.0003328028031290743, | |
| "loss": 0.1039, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_loss": 0.11941063404083252, | |
| "eval_runtime": 368.032, | |
| "eval_samples_per_second": 14.751, | |
| "eval_wer": 0.17829099307159355, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "learning_rate": 0.0003307884615384615, | |
| "loss": 0.0963, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 6.11, | |
| "eval_loss": 0.13149844110012054, | |
| "eval_runtime": 395.5282, | |
| "eval_samples_per_second": 13.726, | |
| "eval_wer": 0.17823967154221196, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 0.00032877411994784874, | |
| "loss": 0.0947, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "eval_loss": 0.11894866079092026, | |
| "eval_runtime": 394.9332, | |
| "eval_samples_per_second": 13.747, | |
| "eval_wer": 0.16925840390043623, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00032675977835723596, | |
| "loss": 0.0946, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.12667880952358246, | |
| "eval_runtime": 365.722, | |
| "eval_samples_per_second": 14.845, | |
| "eval_wer": 0.1762894534257121, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 0.0003247454367666232, | |
| "loss": 0.0935, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "eval_loss": 0.12555569410324097, | |
| "eval_runtime": 401.4704, | |
| "eval_samples_per_second": 13.523, | |
| "eval_wer": 0.1729022324865281, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 0.00032273780964797916, | |
| "loss": 0.088, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "eval_loss": 0.12047957628965378, | |
| "eval_runtime": 389.5023, | |
| "eval_samples_per_second": 13.938, | |
| "eval_wer": 0.16546061072619964, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 0.00032072346805736633, | |
| "loss": 0.0947, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "eval_loss": 0.1205919086933136, | |
| "eval_runtime": 397.9241, | |
| "eval_samples_per_second": 13.643, | |
| "eval_wer": 0.17202976648704132, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 0.00031870912646675355, | |
| "loss": 0.0897, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_loss": 0.1126471757888794, | |
| "eval_runtime": 364.5953, | |
| "eval_samples_per_second": 14.89, | |
| "eval_wer": 0.16556325378496278, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "learning_rate": 0.00031669478487614077, | |
| "loss": 0.0853, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "eval_loss": 0.1229565218091011, | |
| "eval_runtime": 386.8026, | |
| "eval_samples_per_second": 14.036, | |
| "eval_wer": 0.1657685399024891, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 0.00031468044328552804, | |
| "loss": 0.0844, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "eval_loss": 0.1236405223608017, | |
| "eval_runtime": 399.2408, | |
| "eval_samples_per_second": 13.598, | |
| "eval_wer": 0.15991788555298947, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "learning_rate": 0.00031266610169491526, | |
| "loss": 0.0836, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 7.42, | |
| "eval_loss": 0.12277714908123016, | |
| "eval_runtime": 395.0189, | |
| "eval_samples_per_second": 13.744, | |
| "eval_wer": 0.17233769566333076, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.00016603860759493668, | |
| "loss": 0.0827, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_loss": 0.11456963419914246, | |
| "eval_runtime": 391.5783, | |
| "eval_samples_per_second": 13.864, | |
| "eval_wer": 0.14801129073646394, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 0.00016114936708860758, | |
| "loss": 0.0774, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "eval_loss": 0.10517676174640656, | |
| "eval_runtime": 387.996, | |
| "eval_samples_per_second": 13.992, | |
| "eval_wer": 0.14251988709263536, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 0.00015626012658227848, | |
| "loss": 0.0662, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "eval_loss": 0.10210742056369781, | |
| "eval_runtime": 393.9886, | |
| "eval_samples_per_second": 13.78, | |
| "eval_wer": 0.13456505003849115, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 0.00015138718354430379, | |
| "loss": 0.0709, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "eval_loss": 0.09889358282089233, | |
| "eval_runtime": 373.1692, | |
| "eval_samples_per_second": 14.548, | |
| "eval_wer": 0.13451372850910956, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "learning_rate": 0.00014649794303797466, | |
| "loss": 0.068, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_loss": 0.09926428645849228, | |
| "eval_runtime": 387.0555, | |
| "eval_samples_per_second": 14.026, | |
| "eval_wer": 0.1367718758018989, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 0.00014160870253164559, | |
| "loss": 0.0641, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "eval_loss": 0.09896103292703629, | |
| "eval_runtime": 394.68, | |
| "eval_samples_per_second": 13.755, | |
| "eval_wer": 0.12948421862971515, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "learning_rate": 0.00013671946202531646, | |
| "loss": 0.0633, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 6.75, | |
| "eval_loss": 0.09418534487485886, | |
| "eval_runtime": 389.6282, | |
| "eval_samples_per_second": 13.934, | |
| "eval_wer": 0.12799589427764949, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "learning_rate": 0.00013183022151898733, | |
| "loss": 0.0662, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "eval_loss": 0.09439662098884583, | |
| "eval_runtime": 385.8863, | |
| "eval_samples_per_second": 14.069, | |
| "eval_wer": 0.12553246086733386, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 0.00012694098101265823, | |
| "loss": 0.0626, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_loss": 0.09515263140201569, | |
| "eval_runtime": 386.6219, | |
| "eval_samples_per_second": 14.042, | |
| "eval_wer": 0.12199127534000513, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 0.00012205174050632911, | |
| "loss": 0.0554, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_loss": 0.09619747847318649, | |
| "eval_runtime": 388.8787, | |
| "eval_samples_per_second": 13.961, | |
| "eval_wer": 0.12158070310495253, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "learning_rate": 0.0001171625, | |
| "loss": 0.0532, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 7.21, | |
| "eval_loss": 0.10246041417121887, | |
| "eval_runtime": 392.3678, | |
| "eval_samples_per_second": 13.837, | |
| "eval_wer": 0.12178598922247882, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "learning_rate": 0.00011228955696202532, | |
| "loss": 0.0522, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 7.33, | |
| "eval_loss": 0.09738663583993912, | |
| "eval_runtime": 387.5929, | |
| "eval_samples_per_second": 14.007, | |
| "eval_wer": 0.1191172696946369, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 0.0001074003164556962, | |
| "loss": 0.0514, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "eval_loss": 0.096110999584198, | |
| "eval_runtime": 385.6979, | |
| "eval_samples_per_second": 14.076, | |
| "eval_wer": 0.12045162945855786, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 0.00010251107594936708, | |
| "loss": 0.0507, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "eval_loss": 0.09606499969959259, | |
| "eval_runtime": 388.8775, | |
| "eval_samples_per_second": 13.961, | |
| "eval_wer": 0.11778290993071594, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 9.76381329113924e-05, | |
| "loss": 0.0512, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "eval_loss": 0.0907408595085144, | |
| "eval_runtime": 369.8847, | |
| "eval_samples_per_second": 14.678, | |
| "eval_wer": 0.11311265075699256, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 9.274889240506329e-05, | |
| "loss": 0.0488, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "eval_loss": 0.08931271731853485, | |
| "eval_runtime": 392.0795, | |
| "eval_samples_per_second": 13.847, | |
| "eval_wer": 0.1107005388760585, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 8.785965189873417e-05, | |
| "loss": 0.0489, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_loss": 0.08885154128074646, | |
| "eval_runtime": 387.663, | |
| "eval_samples_per_second": 14.004, | |
| "eval_wer": 0.1098793944059533, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 8.297041139240506e-05, | |
| "loss": 0.0458, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "eval_loss": 0.0913764014840126, | |
| "eval_runtime": 388.0349, | |
| "eval_samples_per_second": 13.991, | |
| "eval_wer": 0.10644085193738774, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 7.808117088607595e-05, | |
| "loss": 0.0432, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "eval_loss": 0.0885995477437973, | |
| "eval_runtime": 390.3347, | |
| "eval_samples_per_second": 13.909, | |
| "eval_wer": 0.10746728252501925, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "learning_rate": 7.319193037974683e-05, | |
| "loss": 0.0402, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_loss": 0.09285327792167664, | |
| "eval_runtime": 390.0106, | |
| "eval_samples_per_second": 13.92, | |
| "eval_wer": 0.1060816012317167, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 6.830268987341772e-05, | |
| "loss": 0.043, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "eval_loss": 0.09266022592782974, | |
| "eval_runtime": 390.9718, | |
| "eval_samples_per_second": 13.886, | |
| "eval_wer": 0.10515781370284835, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 6.34134493670886e-05, | |
| "loss": 0.041, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "eval_loss": 0.0903313010931015, | |
| "eval_runtime": 392.2575, | |
| "eval_samples_per_second": 13.84, | |
| "eval_wer": 0.10325891711573004, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 5.8524208860759495e-05, | |
| "loss": 0.0387, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "eval_loss": 0.09009859710931778, | |
| "eval_runtime": 381.4418, | |
| "eval_samples_per_second": 14.233, | |
| "eval_wer": 0.10212984346933539, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "learning_rate": 5.365126582278481e-05, | |
| "loss": 0.0381, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 8.73, | |
| "eval_loss": 0.0909804031252861, | |
| "eval_runtime": 387.8645, | |
| "eval_samples_per_second": 13.997, | |
| "eval_wer": 0.10028226841159867, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 4.87620253164557e-05, | |
| "loss": 0.0368, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "eval_loss": 0.08640129864215851, | |
| "eval_runtime": 389.071, | |
| "eval_samples_per_second": 13.954, | |
| "eval_wer": 0.09899923017705928, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "learning_rate": 4.387278481012658e-05, | |
| "loss": 0.0393, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_loss": 0.08331574499607086, | |
| "eval_runtime": 370.5127, | |
| "eval_samples_per_second": 14.653, | |
| "eval_wer": 0.0987426225301514, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 3.898354430379747e-05, | |
| "loss": 0.0384, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_loss": 0.08789246529340744, | |
| "eval_runtime": 384.7677, | |
| "eval_samples_per_second": 14.11, | |
| "eval_wer": 0.09905055170644085, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 3.409430379746835e-05, | |
| "loss": 0.0335, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "eval_loss": 0.0857023224234581, | |
| "eval_runtime": 380.952, | |
| "eval_samples_per_second": 14.251, | |
| "eval_wer": 0.09612522453169105, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 2.9205063291139242e-05, | |
| "loss": 0.0356, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "eval_loss": 0.08656106889247894, | |
| "eval_runtime": 388.804, | |
| "eval_samples_per_second": 13.963, | |
| "eval_wer": 0.0953554015909674, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "learning_rate": 2.4315822784810126e-05, | |
| "loss": 0.0354, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "eval_loss": 0.08513284474611282, | |
| "eval_runtime": 388.1691, | |
| "eval_samples_per_second": 13.986, | |
| "eval_wer": 0.09597125994354631, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 1.9426582278481013e-05, | |
| "loss": 0.0311, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "eval_loss": 0.08604401350021362, | |
| "eval_runtime": 387.2711, | |
| "eval_samples_per_second": 14.019, | |
| "eval_wer": 0.09402104182704644, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 1.4537341772151898e-05, | |
| "loss": 0.0328, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 0.08348560333251953, | |
| "eval_runtime": 387.9108, | |
| "eval_samples_per_second": 13.995, | |
| "eval_wer": 0.09412368488580959, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "learning_rate": 9.648101265822784e-06, | |
| "loss": 0.0284, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 9.78, | |
| "eval_loss": 0.0844394788146019, | |
| "eval_runtime": 389.4719, | |
| "eval_samples_per_second": 13.939, | |
| "eval_wer": 0.09319989735694123, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 4.758860759493671e-06, | |
| "loss": 0.0305, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "eval_loss": 0.08396408706903458, | |
| "eval_runtime": 389.8127, | |
| "eval_samples_per_second": 13.927, | |
| "eval_wer": 0.09196818065178343, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 25780, | |
| "total_flos": 1.6778924993531563e+20, | |
| "train_runtime": 66940.1986, | |
| "train_samples_per_second": 0.385 | |
| } | |
| ], | |
| "max_steps": 25780, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.6778924993531563e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |