| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 2900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 8.000000000000001e-07, | |
| "loss": 9.0973, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.8e-06, | |
| "loss": 6.2616, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 3.5171, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.8e-06, | |
| "loss": 3.2442, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 2.9931, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.8e-06, | |
| "loss": 2.9325, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 2.9366, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.8e-06, | |
| "loss": 2.9523, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.8e-06, | |
| "loss": 2.9003, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 2.9043, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 2.968320369720459, | |
| "eval_runtime": 55.7229, | |
| "eval_samples_per_second": 30.149, | |
| "eval_steps_per_second": 30.149, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.08e-05, | |
| "loss": 2.9203, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.18e-05, | |
| "loss": 2.9374, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 2.9319, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3800000000000002e-05, | |
| "loss": 2.8861, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.48e-05, | |
| "loss": 2.959, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.58e-05, | |
| "loss": 2.9416, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 2.8794, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.78e-05, | |
| "loss": 2.8955, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.88e-05, | |
| "loss": 2.932, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.9800000000000004e-05, | |
| "loss": 2.8537, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.928142786026001, | |
| "eval_runtime": 55.9102, | |
| "eval_samples_per_second": 30.048, | |
| "eval_steps_per_second": 30.048, | |
| "eval_wer": 0.9997243470470677, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.08e-05, | |
| "loss": 2.8264, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.18e-05, | |
| "loss": 2.8669, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.2800000000000002e-05, | |
| "loss": 2.8596, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.38e-05, | |
| "loss": 2.8239, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.48e-05, | |
| "loss": 2.7974, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.58e-05, | |
| "loss": 2.8308, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 2.6800000000000004e-05, | |
| "loss": 2.7328, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 2.7800000000000005e-05, | |
| "loss": 2.7796, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.88e-05, | |
| "loss": 2.824, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.98e-05, | |
| "loss": 2.7803, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.7330267429351807, | |
| "eval_runtime": 55.7999, | |
| "eval_samples_per_second": 30.108, | |
| "eval_steps_per_second": 30.108, | |
| "eval_wer": 0.9999310867617669, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.08e-05, | |
| "loss": 2.6748, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.18e-05, | |
| "loss": 2.6847, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 3.2800000000000004e-05, | |
| "loss": 2.779, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.38e-05, | |
| "loss": 2.6724, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.48e-05, | |
| "loss": 2.6407, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.58e-05, | |
| "loss": 2.657, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.68e-05, | |
| "loss": 2.7049, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.7800000000000004e-05, | |
| "loss": 2.5902, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.88e-05, | |
| "loss": 2.6132, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.9800000000000005e-05, | |
| "loss": 2.6806, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 2.5792078971862793, | |
| "eval_runtime": 56.1267, | |
| "eval_samples_per_second": 29.932, | |
| "eval_steps_per_second": 29.932, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.08e-05, | |
| "loss": 2.6065, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.18e-05, | |
| "loss": 2.5238, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.2800000000000004e-05, | |
| "loss": 2.5459, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.38e-05, | |
| "loss": 2.6672, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.4800000000000005e-05, | |
| "loss": 2.5004, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.58e-05, | |
| "loss": 2.4467, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.6800000000000006e-05, | |
| "loss": 2.5114, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.78e-05, | |
| "loss": 2.5773, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.88e-05, | |
| "loss": 2.3931, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.9800000000000004e-05, | |
| "loss": 2.4136, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 2.432730197906494, | |
| "eval_runtime": 55.7997, | |
| "eval_samples_per_second": 30.108, | |
| "eval_steps_per_second": 30.108, | |
| "eval_wer": 0.9947625938942871, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.08e-05, | |
| "loss": 2.5782, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.1800000000000005e-05, | |
| "loss": 2.4132, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.28e-05, | |
| "loss": 2.3392, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.380000000000001e-05, | |
| "loss": 2.4972, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 5.4800000000000004e-05, | |
| "loss": 2.592, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 5.580000000000001e-05, | |
| "loss": 2.3403, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 5.68e-05, | |
| "loss": 2.3356, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.7799999999999995e-05, | |
| "loss": 2.5141, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 5.88e-05, | |
| "loss": 2.468, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.9800000000000003e-05, | |
| "loss": 2.1682, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 2.3507769107818604, | |
| "eval_runtime": 56.1338, | |
| "eval_samples_per_second": 29.929, | |
| "eval_steps_per_second": 29.929, | |
| "eval_wer": 0.9876645303562814, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 6.08e-05, | |
| "loss": 2.2526, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.18e-05, | |
| "loss": 2.5135, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.280000000000001e-05, | |
| "loss": 2.2708, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 6.38e-05, | |
| "loss": 2.194, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 6.48e-05, | |
| "loss": 2.3454, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 6.58e-05, | |
| "loss": 2.4744, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.680000000000001e-05, | |
| "loss": 2.1678, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.780000000000001e-05, | |
| "loss": 2.2312, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 6.879999999999999e-05, | |
| "loss": 2.4317, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 6.98e-05, | |
| "loss": 2.2577, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_loss": 2.2175588607788086, | |
| "eval_runtime": 56.2525, | |
| "eval_samples_per_second": 29.865, | |
| "eval_steps_per_second": 29.865, | |
| "eval_wer": 0.9773275446213218, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.08e-05, | |
| "loss": 2.0414, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 7.18e-05, | |
| "loss": 2.2138, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.280000000000001e-05, | |
| "loss": 2.4531, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 7.38e-05, | |
| "loss": 2.0281, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 7.48e-05, | |
| "loss": 1.9785, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 7.58e-05, | |
| "loss": 2.2353, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 7.680000000000001e-05, | |
| "loss": 2.3622, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 7.780000000000001e-05, | |
| "loss": 1.9538, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.88e-05, | |
| "loss": 2.0412, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 7.98e-05, | |
| "loss": 2.355, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_loss": 2.175335168838501, | |
| "eval_runtime": 56.1512, | |
| "eval_samples_per_second": 29.919, | |
| "eval_steps_per_second": 29.919, | |
| "eval_wer": 0.954172696575012, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 8.080000000000001e-05, | |
| "loss": 2.0461, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8.18e-05, | |
| "loss": 1.8626, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 8.28e-05, | |
| "loss": 2.1198, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 8.38e-05, | |
| "loss": 2.3944, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 8.48e-05, | |
| "loss": 1.8489, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.58e-05, | |
| "loss": 1.9309, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.680000000000001e-05, | |
| "loss": 2.2409, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 8.78e-05, | |
| "loss": 2.0836, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 8.88e-05, | |
| "loss": 1.7763, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 8.98e-05, | |
| "loss": 1.8588, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 2.065042018890381, | |
| "eval_runtime": 56.4539, | |
| "eval_samples_per_second": 29.759, | |
| "eval_steps_per_second": 29.759, | |
| "eval_wer": 0.8851216318654813, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 9.080000000000001e-05, | |
| "loss": 2.3039, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 9.180000000000001e-05, | |
| "loss": 1.9357, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 9.28e-05, | |
| "loss": 1.8426, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 9.38e-05, | |
| "loss": 2.0648, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 9.48e-05, | |
| "loss": 2.1967, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 9.58e-05, | |
| "loss": 1.7113, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 9.680000000000001e-05, | |
| "loss": 1.8058, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 9.78e-05, | |
| "loss": 2.2581, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 9.88e-05, | |
| "loss": 1.9886, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 9.98e-05, | |
| "loss": 1.6831, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 2.0109121799468994, | |
| "eval_runtime": 56.4838, | |
| "eval_samples_per_second": 29.743, | |
| "eval_steps_per_second": 29.743, | |
| "eval_wer": 0.8618289573427055, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 9.957894736842106e-05, | |
| "loss": 1.9452, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 9.905263157894737e-05, | |
| "loss": 2.3028, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 9.852631578947368e-05, | |
| "loss": 1.724, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 9.8e-05, | |
| "loss": 1.6638, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 9.747368421052632e-05, | |
| "loss": 1.9702, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 9.694736842105264e-05, | |
| "loss": 2.0713, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 9.642105263157896e-05, | |
| "loss": 1.5564, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 9.589473684210528e-05, | |
| "loss": 1.7359, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 9.536842105263159e-05, | |
| "loss": 2.1635, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 9.48421052631579e-05, | |
| "loss": 1.888, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 1.9660418033599854, | |
| "eval_runtime": 56.2317, | |
| "eval_samples_per_second": 29.876, | |
| "eval_steps_per_second": 29.876, | |
| "eval_wer": 0.8417752050168837, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 9.431578947368421e-05, | |
| "loss": 1.5669, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 9.378947368421053e-05, | |
| "loss": 1.804, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 9.326315789473684e-05, | |
| "loss": 2.0758, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 9.273684210526317e-05, | |
| "loss": 1.5747, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 9.221052631578948e-05, | |
| "loss": 1.6976, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.16842105263158e-05, | |
| "loss": 2.0368, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 9.11578947368421e-05, | |
| "loss": 1.8833, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 9.063157894736842e-05, | |
| "loss": 1.4821, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 9.010526315789473e-05, | |
| "loss": 1.5729, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 8.957894736842106e-05, | |
| "loss": 2.0066, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_loss": 1.984747290611267, | |
| "eval_runtime": 56.4354, | |
| "eval_samples_per_second": 29.769, | |
| "eval_steps_per_second": 29.769, | |
| "eval_wer": 0.8530769760871063, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 8.905263157894737e-05, | |
| "loss": 1.5368, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 8.852631578947368e-05, | |
| "loss": 1.4722, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.7548, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 8.747368421052632e-05, | |
| "loss": 1.9574, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.694736842105264e-05, | |
| "loss": 1.4809, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 8.642105263157895e-05, | |
| "loss": 1.5359, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 8.589473684210528e-05, | |
| "loss": 1.9546, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 8.536842105263159e-05, | |
| "loss": 1.7378, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 8.48421052631579e-05, | |
| "loss": 1.4299, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 8.431578947368422e-05, | |
| "loss": 1.7044, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 1.9760315418243408, | |
| "eval_runtime": 56.2185, | |
| "eval_samples_per_second": 29.883, | |
| "eval_steps_per_second": 29.883, | |
| "eval_wer": 0.852732409895941, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 8.378947368421053e-05, | |
| "loss": 1.9622, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 8.326315789473684e-05, | |
| "loss": 1.3434, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 8.273684210526315e-05, | |
| "loss": 1.2997, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 8.221052631578948e-05, | |
| "loss": 1.6188, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 8.16842105263158e-05, | |
| "loss": 1.8035, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 8.115789473684211e-05, | |
| "loss": 1.3289, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 8.063157894736842e-05, | |
| "loss": 1.4101, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 8.010526315789473e-05, | |
| "loss": 1.7912, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 7.957894736842106e-05, | |
| "loss": 1.4832, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 7.905263157894737e-05, | |
| "loss": 1.3168, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 2.0708327293395996, | |
| "eval_runtime": 56.483, | |
| "eval_samples_per_second": 29.743, | |
| "eval_steps_per_second": 29.743, | |
| "eval_wer": 0.8326786575701193, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 7.852631578947369e-05, | |
| "loss": 1.5461, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 1.9242, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 7.747368421052633e-05, | |
| "loss": 1.3536, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 7.694736842105264e-05, | |
| "loss": 1.3392, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.642105263157895e-05, | |
| "loss": 1.6485, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 7.589473684210526e-05, | |
| "loss": 1.5835, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.536842105263158e-05, | |
| "loss": 1.0981, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 7.48421052631579e-05, | |
| "loss": 1.2099, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 7.431578947368422e-05, | |
| "loss": 1.7216, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 7.378947368421053e-05, | |
| "loss": 1.2143, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 2.060051441192627, | |
| "eval_runtime": 56.4793, | |
| "eval_samples_per_second": 29.745, | |
| "eval_steps_per_second": 29.745, | |
| "eval_wer": 0.8419130314933498, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 7.326315789473684e-05, | |
| "loss": 1.1191, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 7.273684210526316e-05, | |
| "loss": 1.3588, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 7.221052631578947e-05, | |
| "loss": 1.7362, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 7.16842105263158e-05, | |
| "loss": 1.2034, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 7.115789473684211e-05, | |
| "loss": 1.273, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 7.063157894736842e-05, | |
| "loss": 1.6301, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 7.010526315789473e-05, | |
| "loss": 1.4584, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 6.957894736842106e-05, | |
| "loss": 1.1206, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 6.905263157894737e-05, | |
| "loss": 1.2503, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 6.852631578947369e-05, | |
| "loss": 1.6189, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_loss": 2.0960006713867188, | |
| "eval_runtime": 56.5192, | |
| "eval_samples_per_second": 29.724, | |
| "eval_steps_per_second": 29.724, | |
| "eval_wer": 0.8298532148025636, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.0777, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 6.747368421052633e-05, | |
| "loss": 0.9296, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 6.694736842105264e-05, | |
| "loss": 1.204, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 6.642105263157895e-05, | |
| "loss": 1.5133, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 6.589473684210526e-05, | |
| "loss": 0.9943, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 6.536842105263158e-05, | |
| "loss": 1.0309, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 6.484210526315789e-05, | |
| "loss": 1.4495, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 6.431578947368422e-05, | |
| "loss": 1.1653, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 6.378947368421053e-05, | |
| "loss": 0.9316, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 6.326315789473684e-05, | |
| "loss": 1.13, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 2.2540431022644043, | |
| "eval_runtime": 56.8535, | |
| "eval_samples_per_second": 29.55, | |
| "eval_steps_per_second": 29.55, | |
| "eval_wer": 0.8408104196816208, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 6.273684210526316e-05, | |
| "loss": 1.5653, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 6.221052631578947e-05, | |
| "loss": 0.9847, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 6.168421052631578e-05, | |
| "loss": 1.0703, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.115789473684211e-05, | |
| "loss": 1.3565, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 6.063157894736843e-05, | |
| "loss": 1.2267, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 6.010526315789474e-05, | |
| "loss": 0.8131, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 5.9578947368421055e-05, | |
| "loss": 0.8947, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 5.905263157894737e-05, | |
| "loss": 1.3365, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 5.852631578947369e-05, | |
| "loss": 0.9521, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.8001, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 2.4259722232818604, | |
| "eval_runtime": 56.4759, | |
| "eval_samples_per_second": 29.747, | |
| "eval_steps_per_second": 29.747, | |
| "eval_wer": 0.8306112604231273, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 5.747368421052633e-05, | |
| "loss": 1.048, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 5.694736842105264e-05, | |
| "loss": 1.3414, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 5.642105263157895e-05, | |
| "loss": 0.8109, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 5.5894736842105266e-05, | |
| "loss": 0.8485, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 5.536842105263158e-05, | |
| "loss": 1.2154, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 5.484210526315789e-05, | |
| "loss": 1.1199, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 5.431578947368421e-05, | |
| "loss": 0.8175, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 5.378947368421053e-05, | |
| "loss": 0.9612, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 5.326315789473685e-05, | |
| "loss": 1.2567, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 5.2736842105263164e-05, | |
| "loss": 0.7769, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "eval_loss": 2.4182372093200684, | |
| "eval_runtime": 56.5923, | |
| "eval_samples_per_second": 29.686, | |
| "eval_steps_per_second": 29.686, | |
| "eval_wer": 0.8445317345462063, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 5.221052631578948e-05, | |
| "loss": 0.7036, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 5.168421052631579e-05, | |
| "loss": 0.8825, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 5.11578947368421e-05, | |
| "loss": 1.0916, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 5.0631578947368416e-05, | |
| "loss": 0.7172, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 5.010526315789474e-05, | |
| "loss": 0.7482, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 4.9578947368421055e-05, | |
| "loss": 1.1713, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 4.9052631578947375e-05, | |
| "loss": 0.9084, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 4.852631578947369e-05, | |
| "loss": 0.6577, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9152, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 4.747368421052632e-05, | |
| "loss": 1.2165, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 2.3665781021118164, | |
| "eval_runtime": 56.109, | |
| "eval_samples_per_second": 29.942, | |
| "eval_steps_per_second": 29.942, | |
| "eval_wer": 0.8284060367996692, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 4.6947368421052634e-05, | |
| "loss": 0.7083, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 4.642105263157895e-05, | |
| "loss": 0.6976, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.589473684210526e-05, | |
| "loss": 0.9437, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 4.536842105263158e-05, | |
| "loss": 0.8991, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 4.48421052631579e-05, | |
| "loss": 0.5336, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.431578947368421e-05, | |
| "loss": 0.6817, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 4.378947368421053e-05, | |
| "loss": 1.0623, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.3263157894736845e-05, | |
| "loss": 0.7178, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.273684210526316e-05, | |
| "loss": 0.615, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 4.221052631578948e-05, | |
| "loss": 0.8026, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 2.7118234634399414, | |
| "eval_runtime": 56.582, | |
| "eval_samples_per_second": 29.691, | |
| "eval_steps_per_second": 29.691, | |
| "eval_wer": 0.8662394045896217, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 4.168421052631579e-05, | |
| "loss": 0.965, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 4.1157894736842104e-05, | |
| "loss": 0.58, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 4.0631578947368424e-05, | |
| "loss": 0.6671, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 4.010526315789474e-05, | |
| "loss": 0.967, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 3.9578947368421056e-05, | |
| "loss": 0.8289, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 3.905263157894737e-05, | |
| "loss": 0.5458, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 3.852631578947369e-05, | |
| "loss": 0.6513, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.9857, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 3.7473684210526315e-05, | |
| "loss": 0.5503, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 3.6947368421052635e-05, | |
| "loss": 0.5148, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "eval_loss": 2.7957205772399902, | |
| "eval_runtime": 56.2713, | |
| "eval_samples_per_second": 29.855, | |
| "eval_steps_per_second": 29.855, | |
| "eval_wer": 0.8525945834194749, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 3.642105263157895e-05, | |
| "loss": 0.7051, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 3.589473684210527e-05, | |
| "loss": 0.817, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 3.536842105263158e-05, | |
| "loss": 0.5092, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 3.48421052631579e-05, | |
| "loss": 0.5318, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 3.431578947368421e-05, | |
| "loss": 0.8995, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 3.3789473684210526e-05, | |
| "loss": 0.6037, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 3.3263157894736846e-05, | |
| "loss": 0.52, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 3.273684210526316e-05, | |
| "loss": 0.6745, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 3.221052631578947e-05, | |
| "loss": 0.9086, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 3.168421052631579e-05, | |
| "loss": 0.4921, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 2.824350118637085, | |
| "eval_runtime": 56.324, | |
| "eval_samples_per_second": 29.827, | |
| "eval_steps_per_second": 29.827, | |
| "eval_wer": 0.8346082282406451, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 3.115789473684211e-05, | |
| "loss": 0.5383, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.0631578947368425e-05, | |
| "loss": 0.7778, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 3.0105263157894737e-05, | |
| "loss": 0.6716, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 2.9578947368421057e-05, | |
| "loss": 0.4281, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 2.905263157894737e-05, | |
| "loss": 0.511, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.8526315789473683e-05, | |
| "loss": 0.7935, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.4858, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 2.747368421052632e-05, | |
| "loss": 0.4521, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 2.6947368421052632e-05, | |
| "loss": 0.6002, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 2.6421052631578945e-05, | |
| "loss": 0.7629, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 2.8944132328033447, | |
| "eval_runtime": 56.1499, | |
| "eval_samples_per_second": 29.92, | |
| "eval_steps_per_second": 29.92, | |
| "eval_wer": 0.8370201915788023, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 2.5894736842105265e-05, | |
| "loss": 0.4264, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 2.536842105263158e-05, | |
| "loss": 0.4922, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 2.4842105263157898e-05, | |
| "loss": 0.7604, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 2.431578947368421e-05, | |
| "loss": 0.5849, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 2.3789473684210527e-05, | |
| "loss": 0.4361, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 2.3263157894736844e-05, | |
| "loss": 0.5268, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.273684210526316e-05, | |
| "loss": 0.701, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 2.2210526315789476e-05, | |
| "loss": 0.3879, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 2.168421052631579e-05, | |
| "loss": 0.4206, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.1157894736842106e-05, | |
| "loss": 0.5762, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 3.033527374267578, | |
| "eval_runtime": 56.4911, | |
| "eval_samples_per_second": 29.739, | |
| "eval_steps_per_second": 29.739, | |
| "eval_wer": 0.8367445386258701, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.0631578947368422e-05, | |
| "loss": 0.6278, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.010526315789474e-05, | |
| "loss": 0.3739, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 1.957894736842105e-05, | |
| "loss": 0.4282, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 1.9052631578947368e-05, | |
| "loss": 0.6777, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 1.8526315789473687e-05, | |
| "loss": 0.4568, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.3895, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 1.7473684210526317e-05, | |
| "loss": 0.5135, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 1.694736842105263e-05, | |
| "loss": 0.6763, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 1.642105263157895e-05, | |
| "loss": 0.3896, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 1.5894736842105266e-05, | |
| "loss": 0.4076, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "eval_loss": 3.077630043029785, | |
| "eval_runtime": 56.4096, | |
| "eval_samples_per_second": 29.782, | |
| "eval_steps_per_second": 29.782, | |
| "eval_wer": 0.8357797532906072, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.536842105263158e-05, | |
| "loss": 0.6128, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 1.4842105263157895e-05, | |
| "loss": 0.4903, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 1.431578947368421e-05, | |
| "loss": 0.3345, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 18.21, | |
| "learning_rate": 1.3789473684210526e-05, | |
| "loss": 0.4081, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 1.3263157894736844e-05, | |
| "loss": 0.6347, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 1.2736842105263157e-05, | |
| "loss": 0.3937, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 1.2210526315789474e-05, | |
| "loss": 0.3617, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 1.168421052631579e-05, | |
| "loss": 0.5184, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 1.1157894736842106e-05, | |
| "loss": 0.5485, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 1.0631578947368421e-05, | |
| "loss": 0.3395, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 3.1572251319885254, | |
| "eval_runtime": 56.3473, | |
| "eval_samples_per_second": 29.815, | |
| "eval_steps_per_second": 29.815, | |
| "eval_wer": 0.826062986699745, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 1.0105263157894738e-05, | |
| "loss": 0.4157, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 9.578947368421052e-06, | |
| "loss": 0.5832, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 9.05263157894737e-06, | |
| "loss": 0.4316, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 8.526315789473685e-06, | |
| "loss": 0.3585, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.4486, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 7.473684210526316e-06, | |
| "loss": 0.5915, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 6.9473684210526315e-06, | |
| "loss": 0.3305, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "learning_rate": 6.421052631578947e-06, | |
| "loss": 0.3461, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 5.8947368421052634e-06, | |
| "loss": 0.5152, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 5.36842105263158e-06, | |
| "loss": 0.4862, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "eval_loss": 3.1318912506103516, | |
| "eval_runtime": 56.0611, | |
| "eval_samples_per_second": 29.967, | |
| "eval_steps_per_second": 29.967, | |
| "eval_wer": 0.8413617255874853, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 4.842105263157895e-06, | |
| "loss": 0.321, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 4.315789473684211e-06, | |
| "loss": 0.3679, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 3.789473684210527e-06, | |
| "loss": 0.5582, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 3.2631578947368423e-06, | |
| "loss": 0.3519, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "learning_rate": 2.7368421052631583e-06, | |
| "loss": 0.3334, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 19.72, | |
| "learning_rate": 2.2105263157894738e-06, | |
| "loss": 0.458, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.6842105263157895e-06, | |
| "loss": 0.545, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 1.1578947368421055e-06, | |
| "loss": 0.3147, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 6.315789473684211e-07, | |
| "loss": 0.3622, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.0526315789473685e-07, | |
| "loss": 0.5061, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.1593191623687744, | |
| "eval_runtime": 56.4849, | |
| "eval_samples_per_second": 29.742, | |
| "eval_steps_per_second": 29.742, | |
| "eval_wer": 0.8363999724347047, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2900, | |
| "total_flos": 2.6571394113263534e+18, | |
| "train_loss": 1.5437560312501315, | |
| "train_runtime": 4725.355, | |
| "train_samples_per_second": 19.554, | |
| "train_steps_per_second": 0.614 | |
| } | |
| ], | |
| "max_steps": 2900, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.6571394113263534e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |