| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 50.0, | |
| "global_step": 81000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 4.983333333333334e-06, | |
| "loss": 23.3878, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_loss": 21.917871475219727, | |
| "eval_runtime": 24.5965, | |
| "eval_samples_per_second": 30.939, | |
| "eval_steps_per_second": 3.903, | |
| "eval_wer": 1.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 4.905974842767296e-06, | |
| "loss": 8.8862, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_loss": 6.059910297393799, | |
| "eval_runtime": 24.7094, | |
| "eval_samples_per_second": 30.798, | |
| "eval_steps_per_second": 3.885, | |
| "eval_wer": 1.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 4.811635220125786e-06, | |
| "loss": 4.3701, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "eval_loss": 4.383728981018066, | |
| "eval_runtime": 24.9717, | |
| "eval_samples_per_second": 30.474, | |
| "eval_steps_per_second": 3.844, | |
| "eval_wer": 1.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 4.717295597484277e-06, | |
| "loss": 4.113, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "eval_loss": 4.269829750061035, | |
| "eval_runtime": 25.0095, | |
| "eval_samples_per_second": 30.428, | |
| "eval_steps_per_second": 3.839, | |
| "eval_wer": 0.9982187388671179, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "learning_rate": 4.6229559748427675e-06, | |
| "loss": 3.9666, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.63, | |
| "eval_loss": 3.9726266860961914, | |
| "eval_runtime": 24.8315, | |
| "eval_samples_per_second": 30.647, | |
| "eval_steps_per_second": 3.866, | |
| "eval_wer": 0.9989312433202707, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "learning_rate": 4.5286792452830195e-06, | |
| "loss": 3.5965, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.56, | |
| "eval_loss": 3.712357759475708, | |
| "eval_runtime": 25.2229, | |
| "eval_samples_per_second": 30.171, | |
| "eval_steps_per_second": 3.806, | |
| "eval_wer": 0.9975062344139651, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 4.43433962264151e-06, | |
| "loss": 3.3944, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "eval_loss": 3.5005130767822266, | |
| "eval_runtime": 25.0342, | |
| "eval_samples_per_second": 30.398, | |
| "eval_steps_per_second": 3.835, | |
| "eval_wer": 1.0057000356252226, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 4.340062893081761e-06, | |
| "loss": 3.304, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "eval_loss": 3.3709797859191895, | |
| "eval_runtime": 24.7151, | |
| "eval_samples_per_second": 30.791, | |
| "eval_steps_per_second": 3.884, | |
| "eval_wer": 1.0042750267189169, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 4.245723270440252e-06, | |
| "loss": 3.2482, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "eval_loss": 3.4201204776763916, | |
| "eval_runtime": 25.1168, | |
| "eval_samples_per_second": 30.298, | |
| "eval_steps_per_second": 3.822, | |
| "eval_wer": 1.015496971856074, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "learning_rate": 4.151446540880503e-06, | |
| "loss": 3.212, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "eval_loss": 3.373223304748535, | |
| "eval_runtime": 24.7052, | |
| "eval_samples_per_second": 30.803, | |
| "eval_steps_per_second": 3.886, | |
| "eval_wer": 1.0151407196294977, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "learning_rate": 4.057106918238994e-06, | |
| "loss": 3.1778, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "eval_loss": 3.2763421535491943, | |
| "eval_runtime": 25.3297, | |
| "eval_samples_per_second": 30.044, | |
| "eval_steps_per_second": 3.79, | |
| "eval_wer": 1.0008906305664411, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "learning_rate": 3.962767295597484e-06, | |
| "loss": 3.1027, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 11.11, | |
| "eval_loss": 3.1943371295928955, | |
| "eval_runtime": 24.6308, | |
| "eval_samples_per_second": 30.896, | |
| "eval_steps_per_second": 3.898, | |
| "eval_wer": 1.0024937655860349, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 3.868490566037736e-06, | |
| "loss": 2.9905, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "eval_loss": 2.808248281478882, | |
| "eval_runtime": 24.4561, | |
| "eval_samples_per_second": 31.117, | |
| "eval_steps_per_second": 3.925, | |
| "eval_wer": 0.9702529390808693, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 3.774213836477988e-06, | |
| "loss": 2.7095, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_loss": 2.4993014335632324, | |
| "eval_runtime": 24.4488, | |
| "eval_samples_per_second": 31.126, | |
| "eval_steps_per_second": 3.927, | |
| "eval_wer": 0.9301745635910225, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "learning_rate": 3.6799371069182395e-06, | |
| "loss": 2.4862, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 13.89, | |
| "eval_loss": 2.3071722984313965, | |
| "eval_runtime": 26.0996, | |
| "eval_samples_per_second": 29.158, | |
| "eval_steps_per_second": 3.678, | |
| "eval_wer": 0.9139650872817955, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 3.58559748427673e-06, | |
| "loss": 2.3271, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "eval_loss": 2.1398162841796875, | |
| "eval_runtime": 25.2522, | |
| "eval_samples_per_second": 30.136, | |
| "eval_steps_per_second": 3.802, | |
| "eval_wer": 0.8949055931599572, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 15.74, | |
| "learning_rate": 3.4912578616352204e-06, | |
| "loss": 2.1968, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 15.74, | |
| "eval_loss": 2.059424877166748, | |
| "eval_runtime": 24.8853, | |
| "eval_samples_per_second": 30.58, | |
| "eval_steps_per_second": 3.858, | |
| "eval_wer": 0.8817242607766299, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 3.396981132075472e-06, | |
| "loss": 2.111, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "eval_loss": 1.9404386281967163, | |
| "eval_runtime": 24.703, | |
| "eval_samples_per_second": 30.806, | |
| "eval_steps_per_second": 3.886, | |
| "eval_wer": 0.8630210188813681, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 3.3027044025157235e-06, | |
| "loss": 2.0387, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "eval_loss": 1.8895297050476074, | |
| "eval_runtime": 24.8094, | |
| "eval_samples_per_second": 30.674, | |
| "eval_steps_per_second": 3.87, | |
| "eval_wer": 0.8496615603847524, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "learning_rate": 3.208364779874214e-06, | |
| "loss": 1.9504, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 18.52, | |
| "eval_loss": 1.7960795164108276, | |
| "eval_runtime": 25.0414, | |
| "eval_samples_per_second": 30.39, | |
| "eval_steps_per_second": 3.834, | |
| "eval_wer": 0.8314926968293552, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "learning_rate": 3.1140880503144657e-06, | |
| "loss": 1.9039, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 19.44, | |
| "eval_loss": 1.7433359622955322, | |
| "eval_runtime": 25.2242, | |
| "eval_samples_per_second": 30.169, | |
| "eval_steps_per_second": 3.806, | |
| "eval_wer": 0.8213395083719274, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 20.37, | |
| "learning_rate": 3.0198113207547176e-06, | |
| "loss": 1.8342, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 20.37, | |
| "eval_loss": 1.679031491279602, | |
| "eval_runtime": 24.764, | |
| "eval_samples_per_second": 30.73, | |
| "eval_steps_per_second": 3.877, | |
| "eval_wer": 0.7994299964374777, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 21.3, | |
| "learning_rate": 2.9254716981132075e-06, | |
| "loss": 1.7824, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 21.3, | |
| "eval_loss": 1.6291165351867676, | |
| "eval_runtime": 25.2561, | |
| "eval_samples_per_second": 30.131, | |
| "eval_steps_per_second": 3.801, | |
| "eval_wer": 0.782508015675098, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "learning_rate": 2.831132075471698e-06, | |
| "loss": 1.7359, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 22.22, | |
| "eval_loss": 1.578299880027771, | |
| "eval_runtime": 24.8174, | |
| "eval_samples_per_second": 30.664, | |
| "eval_steps_per_second": 3.868, | |
| "eval_wer": 0.770573566084788, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "learning_rate": 2.73685534591195e-06, | |
| "loss": 1.7053, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 23.15, | |
| "eval_loss": 1.5247900485992432, | |
| "eval_runtime": 24.9123, | |
| "eval_samples_per_second": 30.547, | |
| "eval_steps_per_second": 3.854, | |
| "eval_wer": 0.749198432490203, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 24.07, | |
| "learning_rate": 2.6425786163522016e-06, | |
| "loss": 1.6504, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 24.07, | |
| "eval_loss": 1.4929991960525513, | |
| "eval_runtime": 26.1656, | |
| "eval_samples_per_second": 29.084, | |
| "eval_steps_per_second": 3.669, | |
| "eval_wer": 0.7406483790523691, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 2.5482389937106923e-06, | |
| "loss": 1.6263, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 1.4572067260742188, | |
| "eval_runtime": 25.0273, | |
| "eval_samples_per_second": 30.407, | |
| "eval_steps_per_second": 3.836, | |
| "eval_wer": 0.7347702173138582, | |
| "step": 40500 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "learning_rate": 2.4539622641509434e-06, | |
| "loss": 1.5893, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "eval_loss": 1.4201958179473877, | |
| "eval_runtime": 25.1981, | |
| "eval_samples_per_second": 30.201, | |
| "eval_steps_per_second": 3.81, | |
| "eval_wer": 0.7160669754185963, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 26.85, | |
| "learning_rate": 2.359622641509434e-06, | |
| "loss": 1.5669, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 26.85, | |
| "eval_loss": 1.3987159729003906, | |
| "eval_runtime": 24.7489, | |
| "eval_samples_per_second": 30.749, | |
| "eval_steps_per_second": 3.879, | |
| "eval_wer": 0.7142857142857143, | |
| "step": 43500 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "learning_rate": 2.265408805031447e-06, | |
| "loss": 1.5277, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 27.78, | |
| "eval_loss": 1.3511500358581543, | |
| "eval_runtime": 25.1221, | |
| "eval_samples_per_second": 30.292, | |
| "eval_steps_per_second": 3.821, | |
| "eval_wer": 0.6991449946562166, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "learning_rate": 2.1710691823899376e-06, | |
| "loss": 1.501, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 28.7, | |
| "eval_loss": 1.331958532333374, | |
| "eval_runtime": 24.626, | |
| "eval_samples_per_second": 30.902, | |
| "eval_steps_per_second": 3.898, | |
| "eval_wer": 0.6879230495190595, | |
| "step": 46500 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "learning_rate": 2.076729559748428e-06, | |
| "loss": 1.4781, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 29.63, | |
| "eval_loss": 1.3111568689346313, | |
| "eval_runtime": 24.7067, | |
| "eval_samples_per_second": 30.801, | |
| "eval_steps_per_second": 3.886, | |
| "eval_wer": 0.6788386177413609, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "learning_rate": 1.9824528301886794e-06, | |
| "loss": 1.4477, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 30.56, | |
| "eval_loss": 1.2850154638290405, | |
| "eval_runtime": 24.9276, | |
| "eval_samples_per_second": 30.528, | |
| "eval_steps_per_second": 3.851, | |
| "eval_wer": 0.6656572853580335, | |
| "step": 49500 | |
| }, | |
| { | |
| "epoch": 31.48, | |
| "learning_rate": 1.888176100628931e-06, | |
| "loss": 1.4483, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 31.48, | |
| "eval_loss": 1.2812527418136597, | |
| "eval_runtime": 25.2119, | |
| "eval_samples_per_second": 30.184, | |
| "eval_steps_per_second": 3.808, | |
| "eval_wer": 0.6633416458852868, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 32.41, | |
| "learning_rate": 1.7938364779874214e-06, | |
| "loss": 1.4065, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 32.41, | |
| "eval_loss": 1.2474950551986694, | |
| "eval_runtime": 25.9366, | |
| "eval_samples_per_second": 29.341, | |
| "eval_steps_per_second": 3.701, | |
| "eval_wer": 0.6540790879942999, | |
| "step": 52500 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "learning_rate": 1.699496855345912e-06, | |
| "loss": 1.3779, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 33.33, | |
| "eval_loss": 1.224350929260254, | |
| "eval_runtime": 25.3174, | |
| "eval_samples_per_second": 30.058, | |
| "eval_steps_per_second": 3.792, | |
| "eval_wer": 0.6503384396152476, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 34.26, | |
| "learning_rate": 1.6052830188679247e-06, | |
| "loss": 1.3788, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 34.26, | |
| "eval_loss": 1.2116361856460571, | |
| "eval_runtime": 24.9096, | |
| "eval_samples_per_second": 30.551, | |
| "eval_steps_per_second": 3.854, | |
| "eval_wer": 0.6407196294976844, | |
| "step": 55500 | |
| }, | |
| { | |
| "epoch": 35.19, | |
| "learning_rate": 1.5109433962264154e-06, | |
| "loss": 1.3428, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 35.19, | |
| "eval_loss": 1.1938250064849854, | |
| "eval_runtime": 24.9601, | |
| "eval_samples_per_second": 30.489, | |
| "eval_steps_per_second": 3.846, | |
| "eval_wer": 0.6351977199857499, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 36.11, | |
| "learning_rate": 1.4166666666666667e-06, | |
| "loss": 1.3453, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 36.11, | |
| "eval_loss": 1.192725419998169, | |
| "eval_runtime": 25.8368, | |
| "eval_samples_per_second": 29.454, | |
| "eval_steps_per_second": 3.716, | |
| "eval_wer": 0.6339508371927325, | |
| "step": 58500 | |
| }, | |
| { | |
| "epoch": 37.04, | |
| "learning_rate": 1.3223270440251574e-06, | |
| "loss": 1.3137, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 37.04, | |
| "eval_loss": 1.16991126537323, | |
| "eval_runtime": 24.7736, | |
| "eval_samples_per_second": 30.718, | |
| "eval_steps_per_second": 3.875, | |
| "eval_wer": 0.6252226576416102, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 37.96, | |
| "learning_rate": 1.2279874213836478e-06, | |
| "loss": 1.2984, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 37.96, | |
| "eval_loss": 1.1666393280029297, | |
| "eval_runtime": 24.7631, | |
| "eval_samples_per_second": 30.731, | |
| "eval_steps_per_second": 3.877, | |
| "eval_wer": 0.6229070181688635, | |
| "step": 61500 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "learning_rate": 1.1337106918238996e-06, | |
| "loss": 1.2927, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 38.89, | |
| "eval_loss": 1.1585460901260376, | |
| "eval_runtime": 24.9605, | |
| "eval_samples_per_second": 30.488, | |
| "eval_steps_per_second": 3.846, | |
| "eval_wer": 0.6188101175632348, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 39.81, | |
| "learning_rate": 1.039433962264151e-06, | |
| "loss": 1.2919, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 39.81, | |
| "eval_loss": 1.1618481874465942, | |
| "eval_runtime": 24.6184, | |
| "eval_samples_per_second": 30.912, | |
| "eval_steps_per_second": 3.9, | |
| "eval_wer": 0.6189882436765229, | |
| "step": 64500 | |
| }, | |
| { | |
| "epoch": 40.74, | |
| "learning_rate": 9.450943396226415e-07, | |
| "loss": 1.293, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 40.74, | |
| "eval_loss": 1.147918462753296, | |
| "eval_runtime": 24.8441, | |
| "eval_samples_per_second": 30.631, | |
| "eval_steps_per_second": 3.864, | |
| "eval_wer": 0.6180976131100819, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "learning_rate": 8.507547169811322e-07, | |
| "loss": 1.2853, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 41.67, | |
| "eval_loss": 1.1422728300094604, | |
| "eval_runtime": 24.5674, | |
| "eval_samples_per_second": 30.976, | |
| "eval_steps_per_second": 3.908, | |
| "eval_wer": 0.6202351264695405, | |
| "step": 67500 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "learning_rate": 7.565408805031447e-07, | |
| "loss": 1.2687, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 42.59, | |
| "eval_loss": 1.131521224975586, | |
| "eval_runtime": 24.9081, | |
| "eval_samples_per_second": 30.552, | |
| "eval_steps_per_second": 3.854, | |
| "eval_wer": 0.6131100819380121, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 43.52, | |
| "learning_rate": 6.622012578616354e-07, | |
| "loss": 1.2603, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 43.52, | |
| "eval_loss": 1.133251428604126, | |
| "eval_runtime": 25.026, | |
| "eval_samples_per_second": 30.408, | |
| "eval_steps_per_second": 3.836, | |
| "eval_wer": 0.6127538297114357, | |
| "step": 70500 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "learning_rate": 5.678616352201258e-07, | |
| "loss": 1.2577, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 44.44, | |
| "eval_loss": 1.1190848350524902, | |
| "eval_runtime": 24.839, | |
| "eval_samples_per_second": 30.637, | |
| "eval_steps_per_second": 3.865, | |
| "eval_wer": 0.6079444246526541, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 45.37, | |
| "learning_rate": 4.7358490566037737e-07, | |
| "loss": 1.2435, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 45.37, | |
| "eval_loss": 1.1177417039871216, | |
| "eval_runtime": 24.5078, | |
| "eval_samples_per_second": 31.051, | |
| "eval_steps_per_second": 3.917, | |
| "eval_wer": 0.6079444246526541, | |
| "step": 73500 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "learning_rate": 3.79245283018868e-07, | |
| "loss": 1.251, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 46.3, | |
| "eval_loss": 1.1210719347000122, | |
| "eval_runtime": 24.6592, | |
| "eval_samples_per_second": 30.861, | |
| "eval_steps_per_second": 3.893, | |
| "eval_wer": 0.6091913074456715, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 47.22, | |
| "learning_rate": 2.8496855345911953e-07, | |
| "loss": 1.2482, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 47.22, | |
| "eval_loss": 1.1177067756652832, | |
| "eval_runtime": 25.1033, | |
| "eval_samples_per_second": 30.315, | |
| "eval_steps_per_second": 3.824, | |
| "eval_wer": 0.6059850374064838, | |
| "step": 76500 | |
| }, | |
| { | |
| "epoch": 48.15, | |
| "learning_rate": 1.9062893081761006e-07, | |
| "loss": 1.2422, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 48.15, | |
| "eval_loss": 1.1227455139160156, | |
| "eval_runtime": 25.3506, | |
| "eval_samples_per_second": 30.019, | |
| "eval_steps_per_second": 3.787, | |
| "eval_wer": 0.6097256857855362, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 49.07, | |
| "learning_rate": 9.641509433962265e-08, | |
| "loss": 1.2485, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 49.07, | |
| "eval_loss": 1.1186823844909668, | |
| "eval_runtime": 24.6079, | |
| "eval_samples_per_second": 30.925, | |
| "eval_steps_per_second": 3.901, | |
| "eval_wer": 0.6070537940862131, | |
| "step": 79500 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "learning_rate": 2.138364779874214e-09, | |
| "loss": 1.2425, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 1.1176735162734985, | |
| "eval_runtime": 24.8722, | |
| "eval_samples_per_second": 30.596, | |
| "eval_steps_per_second": 3.86, | |
| "eval_wer": 0.6058069112931956, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 81000, | |
| "total_flos": 1.7056142305765743e+20, | |
| "train_loss": 2.504730513961227, | |
| "train_runtime": 70530.0839, | |
| "train_samples_per_second": 18.372, | |
| "train_steps_per_second": 1.148 | |
| } | |
| ], | |
| "max_steps": 81000, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.7056142305765743e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |