| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 27.272727272727273, | |
| "global_step": 3600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.3999999999999997e-05, | |
| "loss": 17.8657, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 4.7999999999999994e-05, | |
| "loss": 12.104, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "learning_rate": 7.199999999999999e-05, | |
| "loss": 4.7542, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 9.599999999999999e-05, | |
| "loss": 3.5251, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00011999999999999999, | |
| "loss": 3.2968, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00014399999999999998, | |
| "loss": 3.1885, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 0.000168, | |
| "loss": 3.173, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00019199999999999998, | |
| "loss": 3.1139, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 0.00021599999999999996, | |
| "loss": 3.103, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 0.00023999999999999998, | |
| "loss": 3.0862, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "eval_loss": 3.177677869796753, | |
| "eval_runtime": 53.6086, | |
| "eval_samples_per_second": 22.403, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.00026399999999999997, | |
| "loss": 3.0227, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 0.00028799999999999995, | |
| "loss": 2.6934, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.94, | |
| "learning_rate": 0.0002982658959537572, | |
| "loss": 2.1587, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 0.00029479768786127165, | |
| "loss": 1.7134, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 0.0002913294797687861, | |
| "loss": 1.4758, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.85, | |
| "learning_rate": 0.00028786127167630053, | |
| "loss": 1.311, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.15, | |
| "learning_rate": 0.000284393063583815, | |
| "loss": 1.1944, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 0.00028092485549132947, | |
| "loss": 1.082, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.0002774566473988439, | |
| "loss": 1.024, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 0.00027398843930635835, | |
| "loss": 0.9718, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "eval_loss": 1.0074659585952759, | |
| "eval_runtime": 53.837, | |
| "eval_samples_per_second": 22.308, | |
| "eval_wer": 0.8955532574974147, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 0.0002705202312138728, | |
| "loss": 0.8674, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 0.0002670520231213873, | |
| "loss": 0.8369, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 0.0002635838150289017, | |
| "loss": 0.8437, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 0.00026011560693641616, | |
| "loss": 0.7578, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 0.00025664739884393063, | |
| "loss": 0.7275, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 7.88, | |
| "learning_rate": 0.00025317919075144504, | |
| "loss": 0.7207, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.18, | |
| "learning_rate": 0.0002497109826589595, | |
| "loss": 0.6415, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 0.000246242774566474, | |
| "loss": 0.6446, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 0.00024277456647398842, | |
| "loss": 0.6243, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 0.0002393063583815029, | |
| "loss": 0.6085, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "eval_loss": 0.8931847214698792, | |
| "eval_runtime": 53.9711, | |
| "eval_samples_per_second": 22.253, | |
| "eval_wer": 0.797015807357069, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 0.00023583815028901733, | |
| "loss": 0.5279, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 9.7, | |
| "learning_rate": 0.00023236994219653174, | |
| "loss": 0.5279, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0002289017341040462, | |
| "loss": 0.5421, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 10.3, | |
| "learning_rate": 0.00022543352601156065, | |
| "loss": 0.4856, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 0.00022196531791907512, | |
| "loss": 0.5182, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 0.00021849710982658956, | |
| "loss": 0.4756, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 11.21, | |
| "learning_rate": 0.00021502890173410403, | |
| "loss": 0.479, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 0.00021156069364161847, | |
| "loss": 0.4419, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 11.82, | |
| "learning_rate": 0.00020809248554913294, | |
| "loss": 0.448, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 0.00020462427745664738, | |
| "loss": 0.4295, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "eval_loss": 0.9030922055244446, | |
| "eval_runtime": 54.1136, | |
| "eval_samples_per_second": 22.194, | |
| "eval_wer": 0.7748559609986704, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 12.42, | |
| "learning_rate": 0.00020115606936416184, | |
| "loss": 0.3976, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 0.00019768786127167629, | |
| "loss": 0.3699, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 0.00019421965317919073, | |
| "loss": 0.4159, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 0.0001907514450867052, | |
| "loss": 0.3221, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 0.00018728323699421963, | |
| "loss": 0.3642, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 0.0001838150289017341, | |
| "loss": 0.3832, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 14.24, | |
| "learning_rate": 0.00018034682080924854, | |
| "loss": 0.3367, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 0.000176878612716763, | |
| "loss": 0.3152, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 0.00017341040462427745, | |
| "loss": 0.3246, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "learning_rate": 0.00016994219653179192, | |
| "loss": 0.3012, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.15, | |
| "eval_loss": 0.9842168688774109, | |
| "eval_runtime": 54.3687, | |
| "eval_samples_per_second": 22.09, | |
| "eval_wer": 0.7636283055104152, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 0.00016647398843930633, | |
| "loss": 0.3051, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 15.76, | |
| "learning_rate": 0.00016300578034682077, | |
| "loss": 0.2966, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 0.00015953757225433524, | |
| "loss": 0.309, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 0.00015606936416184968, | |
| "loss": 0.2553, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 16.67, | |
| "learning_rate": 0.00015260115606936415, | |
| "loss": 0.279, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 0.0001491329479768786, | |
| "loss": 0.2962, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 17.27, | |
| "learning_rate": 0.00014566473988439306, | |
| "loss": 0.2829, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 17.58, | |
| "learning_rate": 0.0001421965317919075, | |
| "loss": 0.2428, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 17.88, | |
| "learning_rate": 0.00013872832369942194, | |
| "loss": 0.2434, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 0.0001352601156069364, | |
| "loss": 0.2513, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "eval_loss": 1.0579547882080078, | |
| "eval_runtime": 84.4423, | |
| "eval_samples_per_second": 14.223, | |
| "eval_wer": 0.7682080070911508, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 0.00013179190751445085, | |
| "loss": 0.2331, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 18.79, | |
| "learning_rate": 0.00012832369942196532, | |
| "loss": 0.2275, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 0.00012485549132947976, | |
| "loss": 0.2275, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 19.39, | |
| "learning_rate": 0.00012138728323699421, | |
| "loss": 0.2174, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 19.7, | |
| "learning_rate": 0.00011791907514450866, | |
| "loss": 0.2202, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0001144508670520231, | |
| "loss": 0.2291, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 20.3, | |
| "learning_rate": 0.00011098265895953756, | |
| "loss": 0.1909, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 20.61, | |
| "learning_rate": 0.00010751445086705201, | |
| "loss": 0.1897, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 20.91, | |
| "learning_rate": 0.00010404624277456647, | |
| "loss": 0.22, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "learning_rate": 0.00010057803468208092, | |
| "loss": 0.194, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.21, | |
| "eval_loss": 1.0953465700149536, | |
| "eval_runtime": 79.9574, | |
| "eval_samples_per_second": 15.02, | |
| "eval_wer": 0.7524006500221598, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 21.52, | |
| "learning_rate": 9.710982658959536e-05, | |
| "loss": 0.1862, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 21.82, | |
| "learning_rate": 9.364161849710982e-05, | |
| "loss": 0.2012, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 22.12, | |
| "learning_rate": 9.017341040462427e-05, | |
| "loss": 0.1823, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 22.42, | |
| "learning_rate": 8.670520231213873e-05, | |
| "loss": 0.1885, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 22.73, | |
| "learning_rate": 8.323699421965317e-05, | |
| "loss": 0.1616, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 23.03, | |
| "learning_rate": 7.976878612716762e-05, | |
| "loss": 0.1893, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 23.33, | |
| "learning_rate": 7.630057803468207e-05, | |
| "loss": 0.1879, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 23.64, | |
| "learning_rate": 7.283236994219653e-05, | |
| "loss": 0.1759, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 23.94, | |
| "learning_rate": 6.936416184971097e-05, | |
| "loss": 0.1703, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "learning_rate": 6.589595375722542e-05, | |
| "loss": 0.1679, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 24.24, | |
| "eval_loss": 1.1073198318481445, | |
| "eval_runtime": 79.6701, | |
| "eval_samples_per_second": 15.075, | |
| "eval_wer": 0.7289112128822574, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 24.55, | |
| "learning_rate": 6.242774566473988e-05, | |
| "loss": 0.1663, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 24.85, | |
| "learning_rate": 5.895953757225433e-05, | |
| "loss": 0.1726, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 25.15, | |
| "learning_rate": 5.549132947976878e-05, | |
| "loss": 0.1733, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 25.45, | |
| "learning_rate": 5.2023121387283234e-05, | |
| "loss": 0.1764, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 25.76, | |
| "learning_rate": 4.855491329479768e-05, | |
| "loss": 0.14, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 26.06, | |
| "learning_rate": 4.5086705202312136e-05, | |
| "loss": 0.164, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 26.36, | |
| "learning_rate": 4.161849710982658e-05, | |
| "loss": 0.1569, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 26.67, | |
| "learning_rate": 3.815028901734104e-05, | |
| "loss": 0.1674, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "learning_rate": 3.4682080924855485e-05, | |
| "loss": 0.1635, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 3.121387283236994e-05, | |
| "loss": 0.1372, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "eval_loss": 1.149294376373291, | |
| "eval_runtime": 54.3637, | |
| "eval_samples_per_second": 22.092, | |
| "eval_wer": 0.7225587235928498, | |
| "step": 3600 | |
| } | |
| ], | |
| "max_steps": 3960, | |
| "num_train_epochs": 30, | |
| "total_flos": 1.107485248885501e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |