| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 49.994535519125684, | |
| "global_step": 4550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.5e-06, | |
| "loss": 3.5867, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 5e-06, | |
| "loss": 3.5457, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 7.5e-06, | |
| "loss": 3.4513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 1e-05, | |
| "loss": 3.3432, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.25e-05, | |
| "loss": 3.3533, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 3.2806732654571533, | |
| "eval_runtime": 190.4728, | |
| "eval_samples_per_second": 25.426, | |
| "eval_steps_per_second": 0.798, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.2217, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 3.1765, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2e-05, | |
| "loss": 3.1408, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 2.2499999999999998e-05, | |
| "loss": 3.1165, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 2.5e-05, | |
| "loss": 3.1709, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_loss": 3.1325438022613525, | |
| "eval_runtime": 192.4978, | |
| "eval_samples_per_second": 25.159, | |
| "eval_steps_per_second": 0.79, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 2.75e-05, | |
| "loss": 3.079, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 3e-05, | |
| "loss": 3.0677, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 3.0656, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.5000000000000004e-05, | |
| "loss": 3.1463, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 3.75e-05, | |
| "loss": 3.0573, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "eval_loss": 3.0614514350891113, | |
| "eval_runtime": 194.36, | |
| "eval_samples_per_second": 24.918, | |
| "eval_steps_per_second": 0.782, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 4e-05, | |
| "loss": 3.0511, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.25e-05, | |
| "loss": 3.0358, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 4.4999999999999996e-05, | |
| "loss": 3.0416, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 4.75e-05, | |
| "loss": 3.1053, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 5e-05, | |
| "loss": 3.0314, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "eval_loss": 3.0990231037139893, | |
| "eval_runtime": 198.3688, | |
| "eval_samples_per_second": 24.414, | |
| "eval_steps_per_second": 0.766, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "learning_rate": 5.25e-05, | |
| "loss": 3.0309, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 5.5e-05, | |
| "loss": 3.0259, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 5.75e-05, | |
| "loss": 3.0998, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 5.27, | |
| "learning_rate": 6e-05, | |
| "loss": 3.0152, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 6.25e-05, | |
| "loss": 3.0129, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "eval_loss": 3.039973497390747, | |
| "eval_runtime": 190.8567, | |
| "eval_samples_per_second": 25.375, | |
| "eval_steps_per_second": 0.796, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 3.0088, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 6.75e-05, | |
| "loss": 3.0051, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 3.073, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 7.25e-05, | |
| "loss": 3.0031, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "learning_rate": 7.5e-05, | |
| "loss": 2.9964, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.59, | |
| "eval_loss": 2.998962640762329, | |
| "eval_runtime": 193.5213, | |
| "eval_samples_per_second": 25.026, | |
| "eval_steps_per_second": 0.785, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 7.75e-05, | |
| "loss": 2.9921, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 8e-05, | |
| "loss": 3.0665, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 7.25, | |
| "learning_rate": 8.25e-05, | |
| "loss": 2.9826, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 8.5e-05, | |
| "loss": 2.9689, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 8.75e-05, | |
| "loss": 2.9602, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "eval_loss": 2.9620397090911865, | |
| "eval_runtime": 193.5851, | |
| "eval_samples_per_second": 25.017, | |
| "eval_steps_per_second": 0.785, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 8.999999999999999e-05, | |
| "loss": 2.9639, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "learning_rate": 9.25e-05, | |
| "loss": 3.0215, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 9.5e-05, | |
| "loss": 2.9454, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "learning_rate": 9.750000000000001e-05, | |
| "loss": 2.9239, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8756, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "eval_loss": 2.7302000522613525, | |
| "eval_runtime": 191.8065, | |
| "eval_samples_per_second": 25.249, | |
| "eval_steps_per_second": 0.792, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 0.0001025, | |
| "loss": 2.8933, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 0.000105, | |
| "loss": 2.7318, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 0.0001075, | |
| "loss": 2.5941, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 0.00011, | |
| "loss": 2.4441, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 2.2931, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "eval_loss": 1.5057899951934814, | |
| "eval_runtime": 196.4368, | |
| "eval_samples_per_second": 24.654, | |
| "eval_steps_per_second": 0.774, | |
| "eval_wer": 0.9775759296054499, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.11, | |
| "learning_rate": 0.000115, | |
| "loss": 2.1999, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 10.33, | |
| "learning_rate": 0.0001175, | |
| "loss": 2.0574, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9672, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 0.0001225, | |
| "loss": 1.9015, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "learning_rate": 0.000125, | |
| "loss": 1.8427, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_loss": 0.9154536724090576, | |
| "eval_runtime": 203.1168, | |
| "eval_samples_per_second": 23.843, | |
| "eval_steps_per_second": 0.748, | |
| "eval_wer": 0.7832245245529378, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 0.0001275, | |
| "loss": 1.8155, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 1.6767, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 0.00013250000000000002, | |
| "loss": 1.5184, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 0.000135, | |
| "loss": 1.4505, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "learning_rate": 0.0001375, | |
| "loss": 1.4286, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.09, | |
| "eval_loss": 0.4074769914150238, | |
| "eval_runtime": 194.7798, | |
| "eval_samples_per_second": 24.864, | |
| "eval_steps_per_second": 0.78, | |
| "eval_wer": 0.37964802724950325, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 0.00014000000000000001, | |
| "loss": 1.3478, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 12.52, | |
| "learning_rate": 0.0001425, | |
| "loss": 1.3164, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 0.000145, | |
| "loss": 1.269, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "learning_rate": 0.0001475, | |
| "loss": 1.2292, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "learning_rate": 0.00015, | |
| "loss": 1.2229, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "eval_loss": 0.28933778405189514, | |
| "eval_runtime": 192.6398, | |
| "eval_samples_per_second": 25.14, | |
| "eval_steps_per_second": 0.789, | |
| "eval_wer": 0.26520011353959694, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 0.0001525, | |
| "loss": 1.182, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 13.62, | |
| "learning_rate": 0.000155, | |
| "loss": 1.1768, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 13.84, | |
| "learning_rate": 0.0001575, | |
| "loss": 1.1534, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 0.00016, | |
| "loss": 1.1832, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 1.1106, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "eval_loss": 0.24685777723789215, | |
| "eval_runtime": 198.5067, | |
| "eval_samples_per_second": 24.397, | |
| "eval_steps_per_second": 0.766, | |
| "eval_wer": 0.2253760999148453, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 0.000165, | |
| "loss": 1.1217, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 14.72, | |
| "learning_rate": 0.0001675, | |
| "loss": 1.1236, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 14.94, | |
| "learning_rate": 0.00017, | |
| "loss": 1.0969, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 15.16, | |
| "learning_rate": 0.0001725, | |
| "loss": 1.0971, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 0.000175, | |
| "loss": 1.0663, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "eval_loss": 0.22192780673503876, | |
| "eval_runtime": 189.4566, | |
| "eval_samples_per_second": 25.563, | |
| "eval_steps_per_second": 0.802, | |
| "eval_wer": 0.19727504967357365, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 0.0001775, | |
| "loss": 1.0733, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 15.82, | |
| "learning_rate": 0.00017999999999999998, | |
| "loss": 1.0635, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 16.04, | |
| "learning_rate": 0.0001825, | |
| "loss": 1.1079, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 16.26, | |
| "learning_rate": 0.000185, | |
| "loss": 1.0467, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 0.0001875, | |
| "loss": 1.0667, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "eval_loss": 0.21288961172103882, | |
| "eval_runtime": 189.0078, | |
| "eval_samples_per_second": 25.623, | |
| "eval_steps_per_second": 0.804, | |
| "eval_wer": 0.18944081748509792, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.7, | |
| "learning_rate": 0.00019, | |
| "loss": 1.0204, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 0.00019250000000000002, | |
| "loss": 1.0356, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 0.00019500000000000002, | |
| "loss": 1.0538, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 17.36, | |
| "learning_rate": 0.0001975, | |
| "loss": 1.0228, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0193, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "eval_loss": 0.1991206556558609, | |
| "eval_runtime": 191.1717, | |
| "eval_samples_per_second": 25.333, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 0.17885325007096226, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "learning_rate": 0.00020250000000000002, | |
| "loss": 1.0322, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 0.000205, | |
| "loss": 1.0176, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 18.24, | |
| "learning_rate": 0.0002075, | |
| "loss": 1.0272, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 0.00021, | |
| "loss": 0.9675, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 18.68, | |
| "learning_rate": 0.0002125, | |
| "loss": 0.9816, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 18.68, | |
| "eval_loss": 0.19402356445789337, | |
| "eval_runtime": 194.75, | |
| "eval_samples_per_second": 24.868, | |
| "eval_steps_per_second": 0.78, | |
| "eval_wer": 0.18010218563724098, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 0.000215, | |
| "loss": 1.0023, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 19.12, | |
| "learning_rate": 0.0002175, | |
| "loss": 1.013, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 19.34, | |
| "learning_rate": 0.00022, | |
| "loss": 0.9664, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 19.56, | |
| "learning_rate": 0.00022250000000000001, | |
| "loss": 0.9736, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 0.9814, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 19.78, | |
| "eval_loss": 0.18596723675727844, | |
| "eval_runtime": 188.4639, | |
| "eval_samples_per_second": 25.697, | |
| "eval_steps_per_second": 0.807, | |
| "eval_wer": 0.16667612829974454, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0002275, | |
| "loss": 1.0064, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 20.22, | |
| "learning_rate": 0.00023, | |
| "loss": 0.9583, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 20.44, | |
| "learning_rate": 0.0002325, | |
| "loss": 0.9646, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 20.66, | |
| "learning_rate": 0.000235, | |
| "loss": 0.9762, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "learning_rate": 0.0002375, | |
| "loss": 0.9787, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 20.87, | |
| "eval_loss": 0.18878202140331268, | |
| "eval_runtime": 188.244, | |
| "eval_samples_per_second": 25.727, | |
| "eval_steps_per_second": 0.807, | |
| "eval_wer": 0.16420664206642066, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 21.1, | |
| "learning_rate": 0.00024, | |
| "loss": 1.0218, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 21.32, | |
| "learning_rate": 0.00024249999999999999, | |
| "loss": 0.9505, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 0.000245, | |
| "loss": 0.9554, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 21.75, | |
| "learning_rate": 0.0002475, | |
| "loss": 0.9728, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 21.97, | |
| "learning_rate": 0.00025, | |
| "loss": 0.9699, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 21.97, | |
| "eval_loss": 0.18748582899570465, | |
| "eval_runtime": 190.7875, | |
| "eval_samples_per_second": 25.384, | |
| "eval_steps_per_second": 0.797, | |
| "eval_wer": 0.17042293499858074, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 22.2, | |
| "learning_rate": 0.00024803921568627453, | |
| "loss": 0.9624, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 22.42, | |
| "learning_rate": 0.000246078431372549, | |
| "loss": 0.9419, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 22.63, | |
| "learning_rate": 0.00024411764705882354, | |
| "loss": 0.9563, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 22.85, | |
| "learning_rate": 0.00024215686274509804, | |
| "loss": 0.9643, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 0.00024019607843137256, | |
| "loss": 0.9616, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "eval_loss": 0.18017500638961792, | |
| "eval_runtime": 191.3932, | |
| "eval_samples_per_second": 25.304, | |
| "eval_steps_per_second": 0.794, | |
| "eval_wer": 0.16173715583309678, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 23.3, | |
| "learning_rate": 0.00023823529411764704, | |
| "loss": 0.917, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 23.51, | |
| "learning_rate": 0.00023627450980392157, | |
| "loss": 0.945, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 23.73, | |
| "learning_rate": 0.0002343137254901961, | |
| "loss": 0.9243, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 23.95, | |
| "learning_rate": 0.0002323529411764706, | |
| "loss": 0.9288, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 24.17, | |
| "learning_rate": 0.0002303921568627451, | |
| "loss": 0.9378, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 24.17, | |
| "eval_loss": 0.17928896844387054, | |
| "eval_runtime": 189.7619, | |
| "eval_samples_per_second": 25.521, | |
| "eval_steps_per_second": 0.801, | |
| "eval_wer": 0.1577348850411581, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 24.39, | |
| "learning_rate": 0.0002284313725490196, | |
| "loss": 0.9071, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 24.61, | |
| "learning_rate": 0.00022647058823529412, | |
| "loss": 0.9054, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 24.83, | |
| "learning_rate": 0.0002246078431372549, | |
| "loss": 0.9303, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 25.05, | |
| "learning_rate": 0.00022264705882352943, | |
| "loss": 0.9376, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 25.27, | |
| "learning_rate": 0.0002206862745098039, | |
| "loss": 0.888, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 25.27, | |
| "eval_loss": 0.17642559111118317, | |
| "eval_runtime": 187.0437, | |
| "eval_samples_per_second": 25.892, | |
| "eval_steps_per_second": 0.813, | |
| "eval_wer": 0.15452739142776042, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 25.49, | |
| "learning_rate": 0.00021872549019607843, | |
| "loss": 0.9135, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 25.71, | |
| "learning_rate": 0.00021676470588235294, | |
| "loss": 0.9094, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 25.93, | |
| "learning_rate": 0.00021480392156862746, | |
| "loss": 0.8879, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 26.15, | |
| "learning_rate": 0.00021284313725490196, | |
| "loss": 0.929, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 26.37, | |
| "learning_rate": 0.00021088235294117647, | |
| "loss": 0.8942, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 26.37, | |
| "eval_loss": 0.16744859516620636, | |
| "eval_runtime": 190.6796, | |
| "eval_samples_per_second": 25.399, | |
| "eval_steps_per_second": 0.797, | |
| "eval_wer": 0.14916264547260857, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 26.59, | |
| "learning_rate": 0.000208921568627451, | |
| "loss": 0.8717, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 26.81, | |
| "learning_rate": 0.0002069607843137255, | |
| "loss": 0.8952, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 27.03, | |
| "learning_rate": 0.000205, | |
| "loss": 0.8849, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 27.25, | |
| "learning_rate": 0.00020313725490196078, | |
| "loss": 0.8781, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "learning_rate": 0.0002011764705882353, | |
| "loss": 0.8701, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 27.47, | |
| "eval_loss": 0.1738910973072052, | |
| "eval_runtime": 187.7205, | |
| "eval_samples_per_second": 25.799, | |
| "eval_steps_per_second": 0.81, | |
| "eval_wer": 0.1511779733181947, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "learning_rate": 0.0001992156862745098, | |
| "loss": 0.8674, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 0.00019725490196078433, | |
| "loss": 0.8752, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 28.13, | |
| "learning_rate": 0.0001952941176470588, | |
| "loss": 0.91, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 28.35, | |
| "learning_rate": 0.00019333333333333333, | |
| "loss": 0.8693, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "learning_rate": 0.00019137254901960786, | |
| "loss": 0.8555, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 28.57, | |
| "eval_loss": 0.1689654141664505, | |
| "eval_runtime": 196.2665, | |
| "eval_samples_per_second": 24.676, | |
| "eval_steps_per_second": 0.774, | |
| "eval_wer": 0.14459267669599773, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 28.79, | |
| "learning_rate": 0.00018941176470588236, | |
| "loss": 0.8796, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 29.01, | |
| "learning_rate": 0.00018745098039215686, | |
| "loss": 0.9045, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 29.23, | |
| "learning_rate": 0.00018549019607843137, | |
| "loss": 0.8515, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 29.45, | |
| "learning_rate": 0.0001835294117647059, | |
| "loss": 0.861, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 29.67, | |
| "learning_rate": 0.0001815686274509804, | |
| "loss": 0.8513, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 29.67, | |
| "eval_loss": 0.16488835215568542, | |
| "eval_runtime": 189.1938, | |
| "eval_samples_per_second": 25.598, | |
| "eval_steps_per_second": 0.803, | |
| "eval_wer": 0.14774340051092819, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 29.89, | |
| "learning_rate": 0.0001796078431372549, | |
| "loss": 0.8573, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 30.11, | |
| "learning_rate": 0.00017764705882352942, | |
| "loss": 0.8686, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 30.33, | |
| "learning_rate": 0.00017568627450980392, | |
| "loss": 0.8462, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 30.55, | |
| "learning_rate": 0.00017372549019607845, | |
| "loss": 0.8419, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 0.00017176470588235293, | |
| "loss": 0.8659, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "eval_loss": 0.1636602133512497, | |
| "eval_runtime": 190.245, | |
| "eval_samples_per_second": 25.457, | |
| "eval_steps_per_second": 0.799, | |
| "eval_wer": 0.1422367300596083, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 30.98, | |
| "learning_rate": 0.00016980392156862745, | |
| "loss": 0.8607, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 31.21, | |
| "learning_rate": 0.00016784313725490196, | |
| "loss": 0.8484, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 31.43, | |
| "learning_rate": 0.00016588235294117648, | |
| "loss": 0.8335, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 31.64, | |
| "learning_rate": 0.00016392156862745098, | |
| "loss": 0.8321, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 31.86, | |
| "learning_rate": 0.00016196078431372549, | |
| "loss": 0.8419, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 31.86, | |
| "eval_loss": 0.16137854754924774, | |
| "eval_runtime": 190.2053, | |
| "eval_samples_per_second": 25.462, | |
| "eval_steps_per_second": 0.799, | |
| "eval_wer": 0.1397104740278172, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 32.09, | |
| "learning_rate": 0.00016, | |
| "loss": 0.8634, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 32.31, | |
| "learning_rate": 0.00015803921568627451, | |
| "loss": 0.8351, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 0.00015607843137254901, | |
| "loss": 0.8427, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 32.74, | |
| "learning_rate": 0.00015411764705882352, | |
| "loss": 0.8324, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "learning_rate": 0.00015215686274509804, | |
| "loss": 0.8491, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_loss": 0.1594749242067337, | |
| "eval_runtime": 191.4631, | |
| "eval_samples_per_second": 25.295, | |
| "eval_steps_per_second": 0.794, | |
| "eval_wer": 0.1401362475163213, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 33.19, | |
| "learning_rate": 0.00015019607843137257, | |
| "loss": 0.8263, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 33.4, | |
| "learning_rate": 0.00014823529411764705, | |
| "loss": 0.8203, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 33.62, | |
| "learning_rate": 0.00014627450980392157, | |
| "loss": 0.8277, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 33.84, | |
| "learning_rate": 0.00014431372549019607, | |
| "loss": 0.8183, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 34.07, | |
| "learning_rate": 0.0001423529411764706, | |
| "loss": 0.8395, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 34.07, | |
| "eval_loss": 0.1606692373752594, | |
| "eval_runtime": 190.8222, | |
| "eval_samples_per_second": 25.38, | |
| "eval_steps_per_second": 0.797, | |
| "eval_wer": 0.13755322168606302, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 34.28, | |
| "learning_rate": 0.0001403921568627451, | |
| "loss": 0.8207, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 34.5, | |
| "learning_rate": 0.0001384313725490196, | |
| "loss": 0.8106, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 34.72, | |
| "learning_rate": 0.00013647058823529413, | |
| "loss": 0.8329, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 34.94, | |
| "learning_rate": 0.00013450980392156863, | |
| "loss": 0.8048, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 35.16, | |
| "learning_rate": 0.00013254901960784313, | |
| "loss": 0.83, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 35.16, | |
| "eval_loss": 0.15384173393249512, | |
| "eval_runtime": 191.1912, | |
| "eval_samples_per_second": 25.331, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 0.1379222253760999, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 35.38, | |
| "learning_rate": 0.00013058823529411764, | |
| "loss": 0.7963, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 35.6, | |
| "learning_rate": 0.00012862745098039216, | |
| "loss": 0.7895, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 35.82, | |
| "learning_rate": 0.0001266666666666667, | |
| "loss": 0.7964, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 36.04, | |
| "learning_rate": 0.0001247058823529412, | |
| "loss": 0.7931, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 36.26, | |
| "learning_rate": 0.0001227450980392157, | |
| "loss": 0.7835, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 36.26, | |
| "eval_loss": 0.16020993888378143, | |
| "eval_runtime": 192.1405, | |
| "eval_samples_per_second": 25.206, | |
| "eval_steps_per_second": 0.791, | |
| "eval_wer": 0.1408458699971615, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 36.48, | |
| "learning_rate": 0.00012078431372549021, | |
| "loss": 0.772, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 36.7, | |
| "learning_rate": 0.00011882352941176471, | |
| "loss": 0.7818, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 36.92, | |
| "learning_rate": 0.00011696078431372549, | |
| "loss": 0.8016, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 37.14, | |
| "learning_rate": 0.000115, | |
| "loss": 0.8061, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 37.36, | |
| "learning_rate": 0.0001130392156862745, | |
| "loss": 0.7703, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 37.36, | |
| "eval_loss": 0.16011376678943634, | |
| "eval_runtime": 187.5367, | |
| "eval_samples_per_second": 25.824, | |
| "eval_steps_per_second": 0.811, | |
| "eval_wer": 0.13692875390292364, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 37.58, | |
| "learning_rate": 0.00011107843137254903, | |
| "loss": 0.7713, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 37.8, | |
| "learning_rate": 0.00010911764705882353, | |
| "loss": 0.7712, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "learning_rate": 0.00010715686274509805, | |
| "loss": 0.8149, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 38.24, | |
| "learning_rate": 0.00010519607843137255, | |
| "loss": 0.7885, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "learning_rate": 0.00010323529411764706, | |
| "loss": 0.7474, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 38.46, | |
| "eval_loss": 0.1514146625995636, | |
| "eval_runtime": 190.8228, | |
| "eval_samples_per_second": 25.38, | |
| "eval_steps_per_second": 0.797, | |
| "eval_wer": 0.1342321884757309, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 38.68, | |
| "learning_rate": 0.00010127450980392156, | |
| "loss": 0.785, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 38.9, | |
| "learning_rate": 9.931372549019609e-05, | |
| "loss": 0.7677, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 39.12, | |
| "learning_rate": 9.73529411764706e-05, | |
| "loss": 0.7849, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 39.34, | |
| "learning_rate": 9.539215686274511e-05, | |
| "loss": 0.7637, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 39.56, | |
| "learning_rate": 9.343137254901961e-05, | |
| "loss": 0.7719, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 39.56, | |
| "eval_loss": 0.15932896733283997, | |
| "eval_runtime": 189.6806, | |
| "eval_samples_per_second": 25.532, | |
| "eval_steps_per_second": 0.801, | |
| "eval_wer": 0.1352540448481408, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 39.78, | |
| "learning_rate": 9.147058823529412e-05, | |
| "loss": 0.7591, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 39.99, | |
| "learning_rate": 8.950980392156862e-05, | |
| "loss": 0.7706, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 40.22, | |
| "learning_rate": 8.754901960784314e-05, | |
| "loss": 0.7805, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 40.44, | |
| "learning_rate": 8.558823529411765e-05, | |
| "loss": 0.7753, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 40.66, | |
| "learning_rate": 8.362745098039217e-05, | |
| "loss": 0.7638, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 40.66, | |
| "eval_loss": 0.15362653136253357, | |
| "eval_runtime": 192.1118, | |
| "eval_samples_per_second": 25.209, | |
| "eval_steps_per_second": 0.791, | |
| "eval_wer": 0.13380641498722678, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 40.87, | |
| "learning_rate": 8.166666666666667e-05, | |
| "loss": 0.7791, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 41.1, | |
| "learning_rate": 7.970588235294118e-05, | |
| "loss": 0.7931, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 41.32, | |
| "learning_rate": 7.774509803921568e-05, | |
| "loss": 0.741, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 41.54, | |
| "learning_rate": 7.57843137254902e-05, | |
| "loss": 0.7434, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 41.75, | |
| "learning_rate": 7.38235294117647e-05, | |
| "loss": 0.771, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 41.75, | |
| "eval_loss": 0.1530592143535614, | |
| "eval_runtime": 186.414, | |
| "eval_samples_per_second": 25.98, | |
| "eval_steps_per_second": 0.815, | |
| "eval_wer": 0.13170593244393983, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 41.97, | |
| "learning_rate": 7.186274509803923e-05, | |
| "loss": 0.7765, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 42.2, | |
| "learning_rate": 6.990196078431373e-05, | |
| "loss": 0.7599, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 42.42, | |
| "learning_rate": 6.794117647058824e-05, | |
| "loss": 0.7782, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 42.63, | |
| "learning_rate": 6.598039215686274e-05, | |
| "loss": 0.7395, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 42.85, | |
| "learning_rate": 6.401960784313726e-05, | |
| "loss": 0.7594, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 42.85, | |
| "eval_loss": 0.14983513951301575, | |
| "eval_runtime": 190.4439, | |
| "eval_samples_per_second": 25.43, | |
| "eval_steps_per_second": 0.798, | |
| "eval_wer": 0.12883905762134545, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 43.08, | |
| "learning_rate": 6.205882352941176e-05, | |
| "loss": 0.7691, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 43.3, | |
| "learning_rate": 6.0098039215686274e-05, | |
| "loss": 0.728, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 43.51, | |
| "learning_rate": 5.813725490196078e-05, | |
| "loss": 0.7585, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 43.73, | |
| "learning_rate": 5.6176470588235296e-05, | |
| "loss": 0.7564, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 43.95, | |
| "learning_rate": 5.4215686274509804e-05, | |
| "loss": 0.7383, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 43.95, | |
| "eval_loss": 0.15269021689891815, | |
| "eval_runtime": 188.5816, | |
| "eval_samples_per_second": 25.681, | |
| "eval_steps_per_second": 0.806, | |
| "eval_wer": 0.13003122338915696, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 44.17, | |
| "learning_rate": 5.225490196078431e-05, | |
| "loss": 0.7693, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 44.39, | |
| "learning_rate": 5.0294117647058826e-05, | |
| "loss": 0.7347, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 44.61, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 0.7185, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 44.83, | |
| "learning_rate": 4.637254901960784e-05, | |
| "loss": 0.7394, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 45.05, | |
| "learning_rate": 4.4411764705882356e-05, | |
| "loss": 0.7565, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 45.05, | |
| "eval_loss": 0.14823457598686218, | |
| "eval_runtime": 191.1254, | |
| "eval_samples_per_second": 25.339, | |
| "eval_steps_per_second": 0.795, | |
| "eval_wer": 0.12886744252057905, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 45.27, | |
| "learning_rate": 4.2450980392156864e-05, | |
| "loss": 0.7535, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 45.49, | |
| "learning_rate": 4.049019607843137e-05, | |
| "loss": 0.737, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 45.71, | |
| "learning_rate": 3.852941176470588e-05, | |
| "loss": 0.7501, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 45.93, | |
| "learning_rate": 3.6568627450980393e-05, | |
| "loss": 0.7285, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "learning_rate": 3.46078431372549e-05, | |
| "loss": 0.7697, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 46.15, | |
| "eval_loss": 0.1494804471731186, | |
| "eval_runtime": 190.3868, | |
| "eval_samples_per_second": 25.438, | |
| "eval_steps_per_second": 0.798, | |
| "eval_wer": 0.1271927334657962, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 46.37, | |
| "learning_rate": 3.264705882352941e-05, | |
| "loss": 0.7283, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 46.59, | |
| "learning_rate": 3.0686274509803923e-05, | |
| "loss": 0.7218, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 46.81, | |
| "learning_rate": 2.872549019607843e-05, | |
| "loss": 0.7341, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 47.03, | |
| "learning_rate": 2.6764705882352942e-05, | |
| "loss": 0.7293, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 47.25, | |
| "learning_rate": 2.480392156862745e-05, | |
| "loss": 0.7194, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 47.25, | |
| "eval_loss": 0.14928147196769714, | |
| "eval_runtime": 189.625, | |
| "eval_samples_per_second": 25.54, | |
| "eval_steps_per_second": 0.802, | |
| "eval_wer": 0.12693726937269373, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 47.47, | |
| "learning_rate": 2.284313725490196e-05, | |
| "loss": 0.7326, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 47.69, | |
| "learning_rate": 2.088235294117647e-05, | |
| "loss": 0.724, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 47.91, | |
| "learning_rate": 1.892156862745098e-05, | |
| "loss": 0.7316, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 48.13, | |
| "learning_rate": 1.696078431372549e-05, | |
| "loss": 0.7566, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 48.35, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.7479, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 48.35, | |
| "eval_loss": 0.1490115076303482, | |
| "eval_runtime": 188.3681, | |
| "eval_samples_per_second": 25.71, | |
| "eval_steps_per_second": 0.807, | |
| "eval_wer": 0.12761850695430033, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 48.57, | |
| "learning_rate": 1.303921568627451e-05, | |
| "loss": 0.7139, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 48.79, | |
| "learning_rate": 1.107843137254902e-05, | |
| "loss": 0.7345, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 9.117647058823529e-06, | |
| "loss": 0.756, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 49.23, | |
| "learning_rate": 7.15686274509804e-06, | |
| "loss": 0.7196, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 49.45, | |
| "learning_rate": 5.196078431372549e-06, | |
| "loss": 0.7132, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 49.45, | |
| "eval_loss": 0.1501067876815796, | |
| "eval_runtime": 190.1989, | |
| "eval_samples_per_second": 25.463, | |
| "eval_steps_per_second": 0.799, | |
| "eval_wer": 0.1265398807834232, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 49.67, | |
| "learning_rate": 3.235294117647059e-06, | |
| "loss": 0.7278, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 49.89, | |
| "learning_rate": 1.2745098039215688e-06, | |
| "loss": 0.7431, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 49.99, | |
| "step": 4550, | |
| "total_flos": 7.03270196503015e+19, | |
| "train_loss": 1.3462627654023223, | |
| "train_runtime": 35973.7, | |
| "train_samples_per_second": 16.248, | |
| "train_steps_per_second": 0.126 | |
| } | |
| ], | |
| "max_steps": 4550, | |
| "num_train_epochs": 50, | |
| "total_flos": 7.03270196503015e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |