| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 32.30769230769231, | |
| "global_step": 2100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9e-07, | |
| "loss": 18.8043, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.8e-06, | |
| "loss": 23.3938, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 20.8738, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.8e-06, | |
| "loss": 19.9257, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 22.7793, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 5.8e-06, | |
| "loss": 18.2911, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 20.4742, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.7e-06, | |
| "loss": 19.7014, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 8.7e-06, | |
| "loss": 15.6078, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 9.7e-06, | |
| "loss": 15.1693, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 12.956756591796875, | |
| "eval_runtime": 111.4724, | |
| "eval_samples_per_second": 3.059, | |
| "eval_steps_per_second": 0.386, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 1.0700000000000001e-05, | |
| "loss": 12.7027, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.1700000000000001e-05, | |
| "loss": 10.9656, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.27e-05, | |
| "loss": 10.6914, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 1.3700000000000001e-05, | |
| "loss": 7.6051, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 1.47e-05, | |
| "loss": 8.7814, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 1.5700000000000002e-05, | |
| "loss": 7.5898, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 1.6700000000000003e-05, | |
| "loss": 7.0232, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 1.77e-05, | |
| "loss": 6.8873, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 1.87e-05, | |
| "loss": 5.7658, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 1.97e-05, | |
| "loss": 6.434, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 5.401930809020996, | |
| "eval_runtime": 111.63, | |
| "eval_samples_per_second": 3.055, | |
| "eval_steps_per_second": 0.385, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 2.07e-05, | |
| "loss": 5.5544, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 2.1700000000000002e-05, | |
| "loss": 5.2728, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.2700000000000003e-05, | |
| "loss": 5.3269, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 2.37e-05, | |
| "loss": 4.8495, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 2.47e-05, | |
| "loss": 4.7196, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 2.57e-05, | |
| "loss": 4.7143, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 2.6700000000000002e-05, | |
| "loss": 4.0896, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 2.7700000000000002e-05, | |
| "loss": 4.342, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 2.87e-05, | |
| "loss": 3.9603, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 2.97e-05, | |
| "loss": 3.9273, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 3.848849058151245, | |
| "eval_runtime": 111.5668, | |
| "eval_samples_per_second": 3.056, | |
| "eval_steps_per_second": 0.385, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 3.07e-05, | |
| "loss": 3.9211, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3.1700000000000005e-05, | |
| "loss": 3.6815, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 3.27e-05, | |
| "loss": 3.7151, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 3.3700000000000006e-05, | |
| "loss": 3.6599, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 3.4699999999999996e-05, | |
| "loss": 3.4484, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 3.57e-05, | |
| "loss": 3.4587, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 3.6700000000000004e-05, | |
| "loss": 3.3518, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.77e-05, | |
| "loss": 3.3416, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.8700000000000006e-05, | |
| "loss": 3.3674, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 3.97e-05, | |
| "loss": 3.1977, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 3.2796881198883057, | |
| "eval_runtime": 111.0722, | |
| "eval_samples_per_second": 3.07, | |
| "eval_steps_per_second": 0.387, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 4.07e-05, | |
| "loss": 3.3328, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 6.46, | |
| "learning_rate": 4.17e-05, | |
| "loss": 3.2257, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 4.27e-05, | |
| "loss": 3.2563, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 4.3700000000000005e-05, | |
| "loss": 3.264, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 6.92, | |
| "learning_rate": 4.47e-05, | |
| "loss": 3.166, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 4.5700000000000006e-05, | |
| "loss": 3.223, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 4.6700000000000003e-05, | |
| "loss": 3.1885, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 4.77e-05, | |
| "loss": 3.182, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 4.87e-05, | |
| "loss": 3.2332, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 4.97e-05, | |
| "loss": 3.1592, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 3.201068162918091, | |
| "eval_runtime": 111.8171, | |
| "eval_samples_per_second": 3.05, | |
| "eval_steps_per_second": 0.385, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.85, | |
| "learning_rate": 5.0700000000000006e-05, | |
| "loss": 3.1996, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 5.17e-05, | |
| "loss": 3.2977, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 5.270000000000001e-05, | |
| "loss": 3.1425, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 8.31, | |
| "learning_rate": 5.3700000000000004e-05, | |
| "loss": 3.2105, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 8.46, | |
| "learning_rate": 5.470000000000001e-05, | |
| "loss": 3.1593, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 5.5700000000000005e-05, | |
| "loss": 3.1665, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 8.77, | |
| "learning_rate": 5.6699999999999996e-05, | |
| "loss": 3.2244, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "learning_rate": 5.77e-05, | |
| "loss": 3.1578, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "learning_rate": 5.87e-05, | |
| "loss": 3.1924, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 5.97e-05, | |
| "loss": 3.1667, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 3.139559030532837, | |
| "eval_runtime": 109.7588, | |
| "eval_samples_per_second": 3.107, | |
| "eval_steps_per_second": 0.392, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 6.07e-05, | |
| "loss": 3.1127, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 9.54, | |
| "learning_rate": 6.170000000000001e-05, | |
| "loss": 3.1775, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 9.69, | |
| "learning_rate": 6.27e-05, | |
| "loss": 3.1132, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 6.37e-05, | |
| "loss": 3.1704, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 6.47e-05, | |
| "loss": 3.1306, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 6.570000000000001e-05, | |
| "loss": 3.0967, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 10.31, | |
| "learning_rate": 6.670000000000001e-05, | |
| "loss": 3.1375, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 10.46, | |
| "learning_rate": 6.77e-05, | |
| "loss": 3.1139, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 6.87e-05, | |
| "loss": 3.098, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "learning_rate": 6.97e-05, | |
| "loss": 3.1111, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 3.0903170108795166, | |
| "eval_runtime": 109.1662, | |
| "eval_samples_per_second": 3.124, | |
| "eval_steps_per_second": 0.394, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 10.92, | |
| "learning_rate": 7.07e-05, | |
| "loss": 3.0775, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 11.08, | |
| "learning_rate": 7.17e-05, | |
| "loss": 3.1294, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 11.23, | |
| "learning_rate": 7.27e-05, | |
| "loss": 3.0943, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 7.37e-05, | |
| "loss": 3.0812, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 11.54, | |
| "learning_rate": 7.47e-05, | |
| "loss": 3.0843, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 11.69, | |
| "learning_rate": 7.570000000000001e-05, | |
| "loss": 3.0668, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 7.670000000000001e-05, | |
| "loss": 3.0617, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 7.77e-05, | |
| "loss": 3.0469, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 12.15, | |
| "learning_rate": 7.87e-05, | |
| "loss": 2.9867, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 7.970000000000001e-05, | |
| "loss": 2.9955, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "eval_cer": 0.9953039832285115, | |
| "eval_loss": 2.895956039428711, | |
| "eval_runtime": 109.2356, | |
| "eval_samples_per_second": 3.122, | |
| "eval_steps_per_second": 0.394, | |
| "eval_wer": 1.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "learning_rate": 8.070000000000001e-05, | |
| "loss": 2.9201, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 8.17e-05, | |
| "loss": 2.8683, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 12.77, | |
| "learning_rate": 8.27e-05, | |
| "loss": 2.8432, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "learning_rate": 8.37e-05, | |
| "loss": 2.6776, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 13.08, | |
| "learning_rate": 8.47e-05, | |
| "loss": 2.5611, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 13.23, | |
| "learning_rate": 8.57e-05, | |
| "loss": 2.4031, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 8.67e-05, | |
| "loss": 2.2939, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "learning_rate": 8.77e-05, | |
| "loss": 2.0835, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 13.69, | |
| "learning_rate": 8.87e-05, | |
| "loss": 1.8785, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "learning_rate": 8.970000000000001e-05, | |
| "loss": 1.784, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "eval_cer": 0.33366876310272536, | |
| "eval_loss": 1.4985727071762085, | |
| "eval_runtime": 109.5182, | |
| "eval_samples_per_second": 3.114, | |
| "eval_steps_per_second": 0.393, | |
| "eval_wer": 0.8279052553663953, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 9.070000000000001e-05, | |
| "loss": 1.6083, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 9.17e-05, | |
| "loss": 1.5164, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 14.31, | |
| "learning_rate": 9.27e-05, | |
| "loss": 1.4019, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 14.46, | |
| "learning_rate": 9.370000000000001e-05, | |
| "loss": 1.3325, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 9.47e-05, | |
| "loss": 1.3251, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 14.77, | |
| "learning_rate": 9.57e-05, | |
| "loss": 1.2811, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 14.92, | |
| "learning_rate": 9.67e-05, | |
| "loss": 1.1927, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 15.08, | |
| "learning_rate": 9.77e-05, | |
| "loss": 1.2017, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 15.23, | |
| "learning_rate": 9.87e-05, | |
| "loss": 1.0452, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 1.1511, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "eval_cer": 0.21970649895178196, | |
| "eval_loss": 0.8053019046783447, | |
| "eval_runtime": 109.46, | |
| "eval_samples_per_second": 3.115, | |
| "eval_steps_per_second": 0.393, | |
| "eval_wer": 0.6402664692820134, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 15.54, | |
| "learning_rate": 9.994166666666667e-05, | |
| "loss": 5.2926, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 15.69, | |
| "learning_rate": 9.985833333333334e-05, | |
| "loss": 3.4895, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 15.85, | |
| "learning_rate": 9.977500000000001e-05, | |
| "loss": 3.3029, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 9.969166666666667e-05, | |
| "loss": 3.2695, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 16.15, | |
| "learning_rate": 9.960833333333333e-05, | |
| "loss": 3.1054, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 16.31, | |
| "learning_rate": 9.952500000000001e-05, | |
| "loss": 3.0923, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 9.944166666666667e-05, | |
| "loss": 2.9955, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 9.935833333333334e-05, | |
| "loss": 2.9343, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 9.9275e-05, | |
| "loss": 2.7871, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "learning_rate": 9.919166666666667e-05, | |
| "loss": 2.3674, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "eval_cer": 0.4807547169811321, | |
| "eval_loss": 1.9613749980926514, | |
| "eval_runtime": 26.9012, | |
| "eval_samples_per_second": 12.676, | |
| "eval_steps_per_second": 1.598, | |
| "eval_wer": 0.9925980754996299, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 17.08, | |
| "learning_rate": 9.910833333333333e-05, | |
| "loss": 2.1319, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 17.23, | |
| "learning_rate": 9.9025e-05, | |
| "loss": 1.6758, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 9.894166666666668e-05, | |
| "loss": 1.5306, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 17.54, | |
| "learning_rate": 9.885833333333334e-05, | |
| "loss": 1.3654, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 17.69, | |
| "learning_rate": 9.8775e-05, | |
| "loss": 1.2197, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "learning_rate": 9.869166666666668e-05, | |
| "loss": 1.1739, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 9.860833333333334e-05, | |
| "loss": 1.0703, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 18.15, | |
| "learning_rate": 9.8525e-05, | |
| "loss": 1.064, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 18.31, | |
| "learning_rate": 9.844166666666667e-05, | |
| "loss": 1.0071, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 9.835833333333334e-05, | |
| "loss": 0.8716, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "eval_cer": 0.20771488469601676, | |
| "eval_loss": 0.7716531753540039, | |
| "eval_runtime": 26.7363, | |
| "eval_samples_per_second": 12.754, | |
| "eval_steps_per_second": 1.608, | |
| "eval_wer": 0.609178386380459, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 9.8275e-05, | |
| "loss": 0.9567, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "learning_rate": 9.819166666666668e-05, | |
| "loss": 0.9223, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 18.92, | |
| "learning_rate": 9.810833333333334e-05, | |
| "loss": 0.8467, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 19.08, | |
| "learning_rate": 9.8025e-05, | |
| "loss": 0.8997, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 19.23, | |
| "learning_rate": 9.794166666666667e-05, | |
| "loss": 0.7379, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 9.785833333333334e-05, | |
| "loss": 0.7997, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 19.54, | |
| "learning_rate": 9.7775e-05, | |
| "loss": 0.8364, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 19.69, | |
| "learning_rate": 9.769166666666667e-05, | |
| "loss": 0.6978, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 19.85, | |
| "learning_rate": 9.760833333333334e-05, | |
| "loss": 0.7572, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 9.7525e-05, | |
| "loss": 0.7108, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.16545073375262054, | |
| "eval_loss": 0.5405445098876953, | |
| "eval_runtime": 26.8237, | |
| "eval_samples_per_second": 12.713, | |
| "eval_steps_per_second": 1.603, | |
| "eval_wer": 0.5159141376757957, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 20.15, | |
| "learning_rate": 9.744166666666667e-05, | |
| "loss": 0.7364, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 20.31, | |
| "learning_rate": 9.735833333333335e-05, | |
| "loss": 0.6907, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 20.46, | |
| "learning_rate": 9.7275e-05, | |
| "loss": 0.6201, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 20.62, | |
| "learning_rate": 9.719166666666667e-05, | |
| "loss": 0.6772, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 20.77, | |
| "learning_rate": 9.710833333333334e-05, | |
| "loss": 0.6928, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "learning_rate": 9.7025e-05, | |
| "loss": 0.6312, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 21.08, | |
| "learning_rate": 9.694166666666667e-05, | |
| "loss": 0.6452, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 21.23, | |
| "learning_rate": 9.685833333333333e-05, | |
| "loss": 0.551, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 21.38, | |
| "learning_rate": 9.677500000000001e-05, | |
| "loss": 0.6457, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 9.669166666666667e-05, | |
| "loss": 0.612, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "eval_cer": 0.140041928721174, | |
| "eval_loss": 0.42258089780807495, | |
| "eval_runtime": 27.0022, | |
| "eval_samples_per_second": 12.629, | |
| "eval_steps_per_second": 1.592, | |
| "eval_wer": 0.463360473723168, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 21.69, | |
| "learning_rate": 9.660833333333333e-05, | |
| "loss": 0.5377, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 21.85, | |
| "learning_rate": 9.652500000000002e-05, | |
| "loss": 0.625, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 9.644166666666668e-05, | |
| "loss": 0.5206, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 22.15, | |
| "learning_rate": 9.635833333333334e-05, | |
| "loss": 0.6004, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 22.31, | |
| "learning_rate": 9.627500000000001e-05, | |
| "loss": 0.5728, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 22.46, | |
| "learning_rate": 9.619166666666667e-05, | |
| "loss": 0.5696, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 22.62, | |
| "learning_rate": 9.610833333333333e-05, | |
| "loss": 0.5712, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "learning_rate": 9.6025e-05, | |
| "loss": 0.5303, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 22.92, | |
| "learning_rate": 9.594166666666668e-05, | |
| "loss": 0.4954, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "learning_rate": 9.585833333333334e-05, | |
| "loss": 0.4919, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 23.08, | |
| "eval_cer": 0.11941299790356394, | |
| "eval_loss": 0.3299271762371063, | |
| "eval_runtime": 27.0051, | |
| "eval_samples_per_second": 12.627, | |
| "eval_steps_per_second": 1.592, | |
| "eval_wer": 0.4052553663952628, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 9.5775e-05, | |
| "loss": 0.391, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 23.38, | |
| "learning_rate": 9.569166666666667e-05, | |
| "loss": 0.5205, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 23.54, | |
| "learning_rate": 9.560833333333333e-05, | |
| "loss": 0.5302, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 23.69, | |
| "learning_rate": 9.5525e-05, | |
| "loss": 0.4806, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 23.85, | |
| "learning_rate": 9.544166666666668e-05, | |
| "loss": 0.5882, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 9.535833333333334e-05, | |
| "loss": 0.4541, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 24.15, | |
| "learning_rate": 9.5275e-05, | |
| "loss": 0.457, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 24.31, | |
| "learning_rate": 9.519166666666667e-05, | |
| "loss": 0.4797, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 24.46, | |
| "learning_rate": 9.510833333333333e-05, | |
| "loss": 0.3973, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 9.5025e-05, | |
| "loss": 0.4999, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "eval_cer": 0.1020545073375262, | |
| "eval_loss": 0.2569698691368103, | |
| "eval_runtime": 30.0585, | |
| "eval_samples_per_second": 11.345, | |
| "eval_steps_per_second": 1.431, | |
| "eval_wer": 0.3604737231680237, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 9.494166666666668e-05, | |
| "loss": 0.452, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 24.92, | |
| "learning_rate": 9.485833333333334e-05, | |
| "loss": 0.4148, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 9.4775e-05, | |
| "loss": 0.4316, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 25.23, | |
| "learning_rate": 9.469166666666667e-05, | |
| "loss": 0.3673, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 25.38, | |
| "learning_rate": 9.460833333333335e-05, | |
| "loss": 0.4365, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 25.54, | |
| "learning_rate": 9.452500000000001e-05, | |
| "loss": 0.4191, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 9.444166666666667e-05, | |
| "loss": 0.3857, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "learning_rate": 9.435833333333334e-05, | |
| "loss": 0.4627, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 9.4275e-05, | |
| "loss": 0.3845, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 26.15, | |
| "learning_rate": 9.419166666666666e-05, | |
| "loss": 0.4349, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 26.15, | |
| "eval_cer": 0.09207547169811321, | |
| "eval_loss": 0.21595901250839233, | |
| "eval_runtime": 27.0141, | |
| "eval_samples_per_second": 12.623, | |
| "eval_steps_per_second": 1.592, | |
| "eval_wer": 0.33826794966691337, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 9.410833333333335e-05, | |
| "loss": 0.359, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 26.46, | |
| "learning_rate": 9.402500000000001e-05, | |
| "loss": 0.3897, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 26.62, | |
| "learning_rate": 9.394166666666667e-05, | |
| "loss": 0.3857, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 26.77, | |
| "learning_rate": 9.385833333333334e-05, | |
| "loss": 0.361, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 9.3775e-05, | |
| "loss": 0.4116, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 27.08, | |
| "learning_rate": 9.369166666666666e-05, | |
| "loss": 0.427, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 27.23, | |
| "learning_rate": 9.360833333333334e-05, | |
| "loss": 0.2724, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 27.38, | |
| "learning_rate": 9.352500000000001e-05, | |
| "loss": 0.416, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 27.54, | |
| "learning_rate": 9.344166666666667e-05, | |
| "loss": 0.3837, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "learning_rate": 9.335833333333333e-05, | |
| "loss": 0.3205, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "eval_cer": 0.07907756813417191, | |
| "eval_loss": 0.17836497724056244, | |
| "eval_runtime": 26.9274, | |
| "eval_samples_per_second": 12.664, | |
| "eval_steps_per_second": 1.597, | |
| "eval_wer": 0.29533678756476683, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 27.85, | |
| "learning_rate": 9.3275e-05, | |
| "loss": 0.3578, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 9.319166666666666e-05, | |
| "loss": 0.3527, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 28.15, | |
| "learning_rate": 9.310833333333334e-05, | |
| "loss": 0.3458, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 28.31, | |
| "learning_rate": 9.302500000000001e-05, | |
| "loss": 0.3527, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 28.46, | |
| "learning_rate": 9.294166666666667e-05, | |
| "loss": 0.3244, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 28.62, | |
| "learning_rate": 9.285833333333333e-05, | |
| "loss": 0.393, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 28.77, | |
| "learning_rate": 9.2775e-05, | |
| "loss": 0.3255, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "learning_rate": 9.269166666666668e-05, | |
| "loss": 0.2884, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 29.08, | |
| "learning_rate": 9.260833333333334e-05, | |
| "loss": 0.3382, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 29.23, | |
| "learning_rate": 9.252500000000001e-05, | |
| "loss": 0.2717, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 29.23, | |
| "eval_cer": 0.07127882599580712, | |
| "eval_loss": 0.14557726681232452, | |
| "eval_runtime": 26.923, | |
| "eval_samples_per_second": 12.666, | |
| "eval_steps_per_second": 1.597, | |
| "eval_wer": 0.2701702442635085, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 29.38, | |
| "learning_rate": 9.244166666666667e-05, | |
| "loss": 0.3271, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 29.54, | |
| "learning_rate": 9.235833333333333e-05, | |
| "loss": 0.3439, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 29.69, | |
| "learning_rate": 9.2275e-05, | |
| "loss": 0.3184, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "learning_rate": 9.219166666666668e-05, | |
| "loss": 0.3458, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 9.210833333333334e-05, | |
| "loss": 0.2775, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 30.15, | |
| "learning_rate": 9.2025e-05, | |
| "loss": 0.2959, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 30.31, | |
| "learning_rate": 9.194166666666667e-05, | |
| "loss": 0.2861, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 30.46, | |
| "learning_rate": 9.185833333333333e-05, | |
| "loss": 0.3047, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 30.62, | |
| "learning_rate": 9.1775e-05, | |
| "loss": 0.3244, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 9.169166666666668e-05, | |
| "loss": 0.2903, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "eval_cer": 0.06532494758909853, | |
| "eval_loss": 0.1265391856431961, | |
| "eval_runtime": 29.4587, | |
| "eval_samples_per_second": 11.576, | |
| "eval_steps_per_second": 1.46, | |
| "eval_wer": 0.2527757216876388, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 30.92, | |
| "learning_rate": 9.160833333333334e-05, | |
| "loss": 0.2849, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 31.08, | |
| "learning_rate": 9.1525e-05, | |
| "loss": 0.3266, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 31.23, | |
| "learning_rate": 9.144166666666668e-05, | |
| "loss": 0.2915, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 31.38, | |
| "learning_rate": 9.135833333333334e-05, | |
| "loss": 0.3066, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 31.54, | |
| "learning_rate": 9.1275e-05, | |
| "loss": 0.3167, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 31.69, | |
| "learning_rate": 9.119166666666667e-05, | |
| "loss": 0.2572, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 31.85, | |
| "learning_rate": 9.110833333333334e-05, | |
| "loss": 0.3269, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 9.1025e-05, | |
| "loss": 0.2422, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 32.15, | |
| "learning_rate": 9.094166666666666e-05, | |
| "loss": 0.3152, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 32.31, | |
| "learning_rate": 9.085833333333334e-05, | |
| "loss": 0.2703, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 32.31, | |
| "eval_cer": 0.060964360587002095, | |
| "eval_loss": 0.11160185188055038, | |
| "eval_runtime": 26.8072, | |
| "eval_samples_per_second": 12.72, | |
| "eval_steps_per_second": 1.604, | |
| "eval_wer": 0.24426350851221318, | |
| "step": 2100 | |
| } | |
| ], | |
| "max_steps": 13000, | |
| "num_train_epochs": 200, | |
| "total_flos": 8.493815951541043e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |