| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 29.666666666666668, | |
| "global_step": 2700, | |
| "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 | |
| } | |
| ], | |
| "max_steps": 4550, | |
| "num_train_epochs": 50, | |
| "total_flos": 4.176183820424104e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |