| { | |
| "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": 9e-07, | |
| "loss": 9.7696, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9e-06, | |
| "loss": 8.5301, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.9e-06, | |
| "loss": 7.8804, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.8e-06, | |
| "loss": 8.3082, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.800000000000001e-06, | |
| "loss": 7.4843, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.8e-06, | |
| "loss": 6.4623, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.800000000000001e-06, | |
| "loss": 5.0935, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.8e-06, | |
| "loss": 4.1987, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.8e-06, | |
| "loss": 3.3439, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.800000000000001e-06, | |
| "loss": 3.1621, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 3.1101927757263184, | |
| "eval_runtime": 55.8678, | |
| "eval_samples_per_second": 30.071, | |
| "eval_steps_per_second": 30.071, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.08e-05, | |
| "loss": 3.1218, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.18e-05, | |
| "loss": 3.0538, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.2800000000000001e-05, | |
| "loss": 3.0235, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3800000000000002e-05, | |
| "loss": 3.0046, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.48e-05, | |
| "loss": 3.0439, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.58e-05, | |
| "loss": 3.0008, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.6800000000000002e-05, | |
| "loss": 2.9794, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.78e-05, | |
| "loss": 2.9939, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.88e-05, | |
| "loss": 2.9932, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.9800000000000004e-05, | |
| "loss": 2.9592, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.960341215133667, | |
| "eval_runtime": 55.5906, | |
| "eval_samples_per_second": 30.221, | |
| "eval_steps_per_second": 30.221, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.08e-05, | |
| "loss": 2.9513, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.18e-05, | |
| "loss": 2.9676, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.2800000000000002e-05, | |
| "loss": 2.9471, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.38e-05, | |
| "loss": 2.9278, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.48e-05, | |
| "loss": 2.92, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.58e-05, | |
| "loss": 2.9487, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 2.6800000000000004e-05, | |
| "loss": 2.8995, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 2.7800000000000005e-05, | |
| "loss": 2.9073, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.88e-05, | |
| "loss": 2.9357, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.98e-05, | |
| "loss": 2.9116, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.8920886516571045, | |
| "eval_runtime": 55.9266, | |
| "eval_samples_per_second": 30.039, | |
| "eval_steps_per_second": 30.039, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.08e-05, | |
| "loss": 2.8585, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.18e-05, | |
| "loss": 2.8498, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 3.2800000000000004e-05, | |
| "loss": 2.8704, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.38e-05, | |
| "loss": 2.7755, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.48e-05, | |
| "loss": 2.6958, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.58e-05, | |
| "loss": 2.6205, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.68e-05, | |
| "loss": 2.5635, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.7800000000000004e-05, | |
| "loss": 2.3608, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.88e-05, | |
| "loss": 2.1984, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.9800000000000005e-05, | |
| "loss": 2.1332, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 1.971829891204834, | |
| "eval_runtime": 55.7409, | |
| "eval_samples_per_second": 30.139, | |
| "eval_steps_per_second": 30.139, | |
| "eval_wer": 0.995796292467783, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.08e-05, | |
| "loss": 1.9219, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.18e-05, | |
| "loss": 1.7055, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.2800000000000004e-05, | |
| "loss": 1.5857, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.38e-05, | |
| "loss": 1.6118, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.4800000000000005e-05, | |
| "loss": 1.2915, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.58e-05, | |
| "loss": 1.1766, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.6800000000000006e-05, | |
| "loss": 1.1746, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.78e-05, | |
| "loss": 1.1026, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.88e-05, | |
| "loss": 0.8672, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.9800000000000004e-05, | |
| "loss": 0.8477, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 0.7813268303871155, | |
| "eval_runtime": 56.0095, | |
| "eval_samples_per_second": 29.995, | |
| "eval_steps_per_second": 29.995, | |
| "eval_wer": 0.5237406105712907, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.08e-05, | |
| "loss": 0.9733, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.1800000000000005e-05, | |
| "loss": 0.6984, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.28e-05, | |
| "loss": 0.65, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.380000000000001e-05, | |
| "loss": 0.7546, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 5.4800000000000004e-05, | |
| "loss": 0.7855, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 5.580000000000001e-05, | |
| "loss": 0.5968, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 5.68e-05, | |
| "loss": 0.6043, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.7799999999999995e-05, | |
| "loss": 0.6865, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 5.88e-05, | |
| "loss": 0.567, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.9800000000000003e-05, | |
| "loss": 0.4251, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 0.5166197419166565, | |
| "eval_runtime": 55.6677, | |
| "eval_samples_per_second": 30.179, | |
| "eval_steps_per_second": 30.179, | |
| "eval_wer": 0.39824960374888013, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 6.08e-05, | |
| "loss": 0.4704, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.18e-05, | |
| "loss": 0.5981, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.280000000000001e-05, | |
| "loss": 0.4192, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 6.38e-05, | |
| "loss": 0.4157, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 6.48e-05, | |
| "loss": 0.4709, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 6.58e-05, | |
| "loss": 0.506, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.680000000000001e-05, | |
| "loss": 0.3589, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.780000000000001e-05, | |
| "loss": 0.3787, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 6.879999999999999e-05, | |
| "loss": 0.5322, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 6.98e-05, | |
| "loss": 0.3743, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_loss": 0.4399681091308594, | |
| "eval_runtime": 56.2233, | |
| "eval_samples_per_second": 29.881, | |
| "eval_steps_per_second": 29.881, | |
| "eval_wer": 0.35779753290607125, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.08e-05, | |
| "loss": 0.3068, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 7.18e-05, | |
| "loss": 0.3595, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.280000000000001e-05, | |
| "loss": 0.4887, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 7.38e-05, | |
| "loss": 0.2713, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 7.48e-05, | |
| "loss": 0.2848, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 7.58e-05, | |
| "loss": 0.3825, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 7.680000000000001e-05, | |
| "loss": 0.3228, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 7.780000000000001e-05, | |
| "loss": 0.2446, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.88e-05, | |
| "loss": 0.2864, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 7.98e-05, | |
| "loss": 0.4194, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_loss": 0.40769341588020325, | |
| "eval_runtime": 56.215, | |
| "eval_samples_per_second": 29.885, | |
| "eval_steps_per_second": 29.885, | |
| "eval_wer": 0.33698573495968576, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 8.080000000000001e-05, | |
| "loss": 0.2523, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8.18e-05, | |
| "loss": 0.2337, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 8.28e-05, | |
| "loss": 0.3113, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 8.38e-05, | |
| "loss": 0.373, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 8.48e-05, | |
| "loss": 0.2086, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.58e-05, | |
| "loss": 0.2483, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.680000000000001e-05, | |
| "loss": 0.3528, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 8.78e-05, | |
| "loss": 0.225, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 8.88e-05, | |
| "loss": 0.1819, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 8.98e-05, | |
| "loss": 0.23, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 0.40175914764404297, | |
| "eval_runtime": 56.0809, | |
| "eval_samples_per_second": 29.957, | |
| "eval_steps_per_second": 29.957, | |
| "eval_wer": 0.3142443663427745, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 9.080000000000001e-05, | |
| "loss": 0.3483, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 9.180000000000001e-05, | |
| "loss": 0.1811, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 9.28e-05, | |
| "loss": 0.1956, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 9.38e-05, | |
| "loss": 0.2701, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 9.48e-05, | |
| "loss": 0.2568, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 9.58e-05, | |
| "loss": 0.1576, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 9.680000000000001e-05, | |
| "loss": 0.2026, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 9.78e-05, | |
| "loss": 0.354, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 9.88e-05, | |
| "loss": 0.2064, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 9.98e-05, | |
| "loss": 0.1554, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 0.362291544675827, | |
| "eval_runtime": 56.0865, | |
| "eval_samples_per_second": 29.954, | |
| "eval_steps_per_second": 29.954, | |
| "eval_wer": 0.2994969333608986, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 9.957894736842106e-05, | |
| "loss": 0.2273, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 9.905263157894737e-05, | |
| "loss": 0.2759, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 9.852631578947368e-05, | |
| "loss": 0.1415, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 9.8e-05, | |
| "loss": 0.1495, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 9.747368421052632e-05, | |
| "loss": 0.2374, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 9.694736842105264e-05, | |
| "loss": 0.205, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 9.642105263157896e-05, | |
| "loss": 0.1435, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 9.589473684210528e-05, | |
| "loss": 0.1814, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 9.536842105263159e-05, | |
| "loss": 0.2679, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 9.48421052631579e-05, | |
| "loss": 0.1511, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 0.343284547328949, | |
| "eval_runtime": 56.3106, | |
| "eval_samples_per_second": 29.835, | |
| "eval_steps_per_second": 29.835, | |
| "eval_wer": 0.26972641444421475, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 9.431578947368421e-05, | |
| "loss": 0.1261, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 9.378947368421053e-05, | |
| "loss": 0.1857, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 9.326315789473684e-05, | |
| "loss": 0.241, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 9.273684210526317e-05, | |
| "loss": 0.1263, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 9.221052631578948e-05, | |
| "loss": 0.1407, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.16842105263158e-05, | |
| "loss": 0.2252, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 9.11578947368421e-05, | |
| "loss": 0.1381, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 9.063157894736842e-05, | |
| "loss": 0.1075, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 9.010526315789473e-05, | |
| "loss": 0.1296, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 8.957894736842106e-05, | |
| "loss": 0.1983, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_loss": 0.3538705110549927, | |
| "eval_runtime": 56.1491, | |
| "eval_samples_per_second": 29.92, | |
| "eval_steps_per_second": 29.92, | |
| "eval_wer": 0.2715181586382744, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 8.905263157894737e-05, | |
| "loss": 0.1206, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 8.852631578947368e-05, | |
| "loss": 0.1165, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.1739, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 8.747368421052632e-05, | |
| "loss": 0.1857, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.694736842105264e-05, | |
| "loss": 0.0872, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 8.642105263157895e-05, | |
| "loss": 0.1306, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 8.589473684210528e-05, | |
| "loss": 0.2083, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 8.536842105263159e-05, | |
| "loss": 0.1227, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 8.48421052631579e-05, | |
| "loss": 0.0972, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 8.431578947368422e-05, | |
| "loss": 0.1443, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 0.3622087836265564, | |
| "eval_runtime": 55.9308, | |
| "eval_samples_per_second": 30.037, | |
| "eval_steps_per_second": 30.037, | |
| "eval_wer": 0.25511680793880503, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 8.378947368421053e-05, | |
| "loss": 0.1891, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 8.326315789473684e-05, | |
| "loss": 0.0701, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 8.273684210526315e-05, | |
| "loss": 0.0991, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 8.221052631578948e-05, | |
| "loss": 0.1548, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 8.16842105263158e-05, | |
| "loss": 0.132, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 8.115789473684211e-05, | |
| "loss": 0.0846, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 8.063157894736842e-05, | |
| "loss": 0.1054, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 8.010526315789473e-05, | |
| "loss": 0.1609, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 7.957894736842106e-05, | |
| "loss": 0.0877, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 7.905263157894737e-05, | |
| "loss": 0.0971, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 0.35799121856689453, | |
| "eval_runtime": 55.9651, | |
| "eval_samples_per_second": 30.019, | |
| "eval_steps_per_second": 30.019, | |
| "eval_wer": 0.251877885741851, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 7.852631578947369e-05, | |
| "loss": 0.1346, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 0.1553, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 7.747368421052633e-05, | |
| "loss": 0.0879, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 7.694736842105264e-05, | |
| "loss": 0.0909, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.642105263157895e-05, | |
| "loss": 0.1714, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 7.589473684210526e-05, | |
| "loss": 0.0766, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.536842105263158e-05, | |
| "loss": 0.0693, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 7.48421052631579e-05, | |
| "loss": 0.0947, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 7.431578947368422e-05, | |
| "loss": 0.1642, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 7.378947368421053e-05, | |
| "loss": 0.0764, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.35290759801864624, | |
| "eval_runtime": 56.302, | |
| "eval_samples_per_second": 29.839, | |
| "eval_steps_per_second": 29.839, | |
| "eval_wer": 0.2437461236303494, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 7.326315789473684e-05, | |
| "loss": 0.0819, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 7.273684210526316e-05, | |
| "loss": 0.1125, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 7.221052631578947e-05, | |
| "loss": 0.1252, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 7.16842105263158e-05, | |
| "loss": 0.0672, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 7.115789473684211e-05, | |
| "loss": 0.0797, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 7.063157894736842e-05, | |
| "loss": 0.1408, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 7.010526315789473e-05, | |
| "loss": 0.0758, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 6.957894736842106e-05, | |
| "loss": 0.069, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 6.905263157894737e-05, | |
| "loss": 0.0953, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 6.852631578947369e-05, | |
| "loss": 0.1203, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_loss": 0.34554988145828247, | |
| "eval_runtime": 56.3041, | |
| "eval_samples_per_second": 29.838, | |
| "eval_steps_per_second": 29.838, | |
| "eval_wer": 0.24305699124801874, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 0.0573, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 6.747368421052633e-05, | |
| "loss": 0.0592, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 6.694736842105264e-05, | |
| "loss": 0.1146, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 6.642105263157895e-05, | |
| "loss": 0.0912, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 6.589473684210526e-05, | |
| "loss": 0.0698, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 6.536842105263158e-05, | |
| "loss": 0.0845, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 6.484210526315789e-05, | |
| "loss": 0.1377, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 6.431578947368422e-05, | |
| "loss": 0.0615, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 6.378947368421053e-05, | |
| "loss": 0.0577, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 6.326315789473684e-05, | |
| "loss": 0.0881, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 0.3647613823413849, | |
| "eval_runtime": 56.4856, | |
| "eval_samples_per_second": 29.742, | |
| "eval_steps_per_second": 29.742, | |
| "eval_wer": 0.24147198676865825, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 6.273684210526316e-05, | |
| "loss": 0.1276, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 6.221052631578947e-05, | |
| "loss": 0.0503, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 6.168421052631578e-05, | |
| "loss": 0.0747, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.115789473684211e-05, | |
| "loss": 0.1326, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 6.063157894736843e-05, | |
| "loss": 0.0642, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 6.010526315789474e-05, | |
| "loss": 0.0548, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 5.9578947368421055e-05, | |
| "loss": 0.0628, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 5.905263157894737e-05, | |
| "loss": 0.1033, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 5.852631578947369e-05, | |
| "loss": 0.06, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 5.8e-05, | |
| "loss": 0.0521, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 0.35644999146461487, | |
| "eval_runtime": 56.2186, | |
| "eval_samples_per_second": 29.883, | |
| "eval_steps_per_second": 29.883, | |
| "eval_wer": 0.23203087313072843, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 5.747368421052633e-05, | |
| "loss": 0.1038, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 5.694736842105264e-05, | |
| "loss": 0.0943, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 5.642105263157895e-05, | |
| "loss": 0.0463, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 5.5894736842105266e-05, | |
| "loss": 0.0514, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 5.536842105263158e-05, | |
| "loss": 0.1006, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 5.484210526315789e-05, | |
| "loss": 0.0524, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 5.431578947368421e-05, | |
| "loss": 0.0533, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 5.378947368421053e-05, | |
| "loss": 0.0683, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 5.326315789473685e-05, | |
| "loss": 0.0983, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 5.2736842105263164e-05, | |
| "loss": 0.0434, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "eval_loss": 0.34853529930114746, | |
| "eval_runtime": 55.7329, | |
| "eval_samples_per_second": 30.144, | |
| "eval_steps_per_second": 30.144, | |
| "eval_wer": 0.2270002067397147, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 5.221052631578948e-05, | |
| "loss": 0.0451, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 5.168421052631579e-05, | |
| "loss": 0.0706, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 5.11578947368421e-05, | |
| "loss": 0.0723, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 5.0631578947368416e-05, | |
| "loss": 0.0433, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 5.010526315789474e-05, | |
| "loss": 0.0722, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 4.9578947368421055e-05, | |
| "loss": 0.0909, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 4.9052631578947375e-05, | |
| "loss": 0.056, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 4.852631578947369e-05, | |
| "loss": 0.0521, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.0793, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 4.747368421052632e-05, | |
| "loss": 0.0864, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.3516710102558136, | |
| "eval_runtime": 56.0246, | |
| "eval_samples_per_second": 29.987, | |
| "eval_steps_per_second": 29.987, | |
| "eval_wer": 0.22279649920749775, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 4.6947368421052634e-05, | |
| "loss": 0.0422, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 4.642105263157895e-05, | |
| "loss": 0.0489, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.589473684210526e-05, | |
| "loss": 0.0975, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 4.536842105263158e-05, | |
| "loss": 0.0459, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 4.48421052631579e-05, | |
| "loss": 0.05, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.431578947368421e-05, | |
| "loss": 0.057, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 4.378947368421053e-05, | |
| "loss": 0.0887, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.3263157894736845e-05, | |
| "loss": 0.033, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.273684210526316e-05, | |
| "loss": 0.0533, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 4.221052631578948e-05, | |
| "loss": 0.0651, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 0.35055288672447205, | |
| "eval_runtime": 56.3016, | |
| "eval_samples_per_second": 29.839, | |
| "eval_steps_per_second": 29.839, | |
| "eval_wer": 0.2285162979808421, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 4.168421052631579e-05, | |
| "loss": 0.0905, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 4.1157894736842104e-05, | |
| "loss": 0.0387, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 4.0631578947368424e-05, | |
| "loss": 0.0607, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 4.010526315789474e-05, | |
| "loss": 0.0808, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 3.9578947368421056e-05, | |
| "loss": 0.0396, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 3.905263157894737e-05, | |
| "loss": 0.036, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 3.852631578947369e-05, | |
| "loss": 0.0439, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3.8e-05, | |
| "loss": 0.0885, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 3.7473684210526315e-05, | |
| "loss": 0.0313, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 3.6947368421052635e-05, | |
| "loss": 0.0423, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "eval_loss": 0.34284868836402893, | |
| "eval_runtime": 56.494, | |
| "eval_samples_per_second": 29.738, | |
| "eval_steps_per_second": 29.738, | |
| "eval_wer": 0.2246571566397905, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 3.642105263157895e-05, | |
| "loss": 0.0724, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 3.589473684210527e-05, | |
| "loss": 0.0547, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 3.536842105263158e-05, | |
| "loss": 0.0358, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 3.48421052631579e-05, | |
| "loss": 0.0557, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 3.431578947368421e-05, | |
| "loss": 0.085, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 3.3789473684210526e-05, | |
| "loss": 0.0429, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 3.3263157894736846e-05, | |
| "loss": 0.0397, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 3.273684210526316e-05, | |
| "loss": 0.0613, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 3.221052631578947e-05, | |
| "loss": 0.0634, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 3.168421052631579e-05, | |
| "loss": 0.0302, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 0.3372061848640442, | |
| "eval_runtime": 56.0069, | |
| "eval_samples_per_second": 29.996, | |
| "eval_steps_per_second": 29.996, | |
| "eval_wer": 0.21983322996347598, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 3.115789473684211e-05, | |
| "loss": 0.0465, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.0631578947368425e-05, | |
| "loss": 0.0689, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 3.0105263157894737e-05, | |
| "loss": 0.042, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 2.9578947368421057e-05, | |
| "loss": 0.0352, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 2.905263157894737e-05, | |
| "loss": 0.0379, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.8526315789473683e-05, | |
| "loss": 0.0766, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 0.0382, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 2.747368421052632e-05, | |
| "loss": 0.0417, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 2.6947368421052632e-05, | |
| "loss": 0.0459, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 2.6421052631578945e-05, | |
| "loss": 0.0548, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 0.34956642985343933, | |
| "eval_runtime": 56.0254, | |
| "eval_samples_per_second": 29.986, | |
| "eval_steps_per_second": 29.986, | |
| "eval_wer": 0.21962649024877678, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 2.5894736842105265e-05, | |
| "loss": 0.0317, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 2.536842105263158e-05, | |
| "loss": 0.0362, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 2.4842105263157898e-05, | |
| "loss": 0.064, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 2.431578947368421e-05, | |
| "loss": 0.0436, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 2.3789473684210527e-05, | |
| "loss": 0.0367, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 2.3263157894736844e-05, | |
| "loss": 0.0443, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.273684210526316e-05, | |
| "loss": 0.0785, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 2.2210526315789476e-05, | |
| "loss": 0.0316, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 2.168421052631579e-05, | |
| "loss": 0.0411, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.1157894736842106e-05, | |
| "loss": 0.0674, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 0.3407049775123596, | |
| "eval_runtime": 55.8959, | |
| "eval_samples_per_second": 30.056, | |
| "eval_steps_per_second": 30.056, | |
| "eval_wer": 0.21659430776652194, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.0631578947368422e-05, | |
| "loss": 0.0392, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.010526315789474e-05, | |
| "loss": 0.0248, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 1.957894736842105e-05, | |
| "loss": 0.0336, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 1.9052631578947368e-05, | |
| "loss": 0.056, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 1.8526315789473687e-05, | |
| "loss": 0.0397, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 1.8e-05, | |
| "loss": 0.0307, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 1.7473684210526317e-05, | |
| "loss": 0.0417, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 1.694736842105263e-05, | |
| "loss": 0.0646, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 1.642105263157895e-05, | |
| "loss": 0.0231, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 1.5894736842105266e-05, | |
| "loss": 0.0291, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "eval_loss": 0.3512323200702667, | |
| "eval_runtime": 56.0829, | |
| "eval_samples_per_second": 29.956, | |
| "eval_steps_per_second": 29.956, | |
| "eval_wer": 0.2170767004341534, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.536842105263158e-05, | |
| "loss": 0.0675, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 1.4842105263157895e-05, | |
| "loss": 0.0357, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 1.431578947368421e-05, | |
| "loss": 0.0254, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 18.21, | |
| "learning_rate": 1.3789473684210526e-05, | |
| "loss": 0.0414, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 1.3263157894736844e-05, | |
| "loss": 0.0572, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 1.2736842105263157e-05, | |
| "loss": 0.0272, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 1.2210526315789474e-05, | |
| "loss": 0.0359, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 1.168421052631579e-05, | |
| "loss": 0.0543, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 1.1157894736842106e-05, | |
| "loss": 0.0579, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 1.0631578947368421e-05, | |
| "loss": 0.0298, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 0.3363025188446045, | |
| "eval_runtime": 56.3537, | |
| "eval_samples_per_second": 29.812, | |
| "eval_steps_per_second": 29.812, | |
| "eval_wer": 0.21576734890772517, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 1.0105263157894738e-05, | |
| "loss": 0.0423, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 9.578947368421052e-06, | |
| "loss": 0.0574, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 9.05263157894737e-06, | |
| "loss": 0.0267, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 8.526315789473685e-06, | |
| "loss": 0.035, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0429, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 7.473684210526316e-06, | |
| "loss": 0.0634, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 6.9473684210526315e-06, | |
| "loss": 0.0315, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "learning_rate": 6.421052631578947e-06, | |
| "loss": 0.0238, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 5.8947368421052634e-06, | |
| "loss": 0.0556, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 5.36842105263158e-06, | |
| "loss": 0.0419, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "eval_loss": 0.34925296902656555, | |
| "eval_runtime": 56.3051, | |
| "eval_samples_per_second": 29.837, | |
| "eval_steps_per_second": 29.837, | |
| "eval_wer": 0.21452691061953, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 4.842105263157895e-06, | |
| "loss": 0.0274, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 4.315789473684211e-06, | |
| "loss": 0.0365, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 3.789473684210527e-06, | |
| "loss": 0.0517, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 3.2631578947368423e-06, | |
| "loss": 0.0274, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "learning_rate": 2.7368421052631583e-06, | |
| "loss": 0.0162, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 19.72, | |
| "learning_rate": 2.2105263157894738e-06, | |
| "loss": 0.0378, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.6842105263157895e-06, | |
| "loss": 0.0544, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 1.1578947368421055e-06, | |
| "loss": 0.0249, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 6.315789473684211e-07, | |
| "loss": 0.0284, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.0526315789473685e-07, | |
| "loss": 0.046, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.3457382619380951, | |
| "eval_runtime": 56.4432, | |
| "eval_samples_per_second": 29.764, | |
| "eval_steps_per_second": 29.764, | |
| "eval_wer": 0.2151471297636276, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2900, | |
| "total_flos": 2.6571394113263534e+18, | |
| "train_loss": 0.6826766253185683, | |
| "train_runtime": 4678.6424, | |
| "train_samples_per_second": 19.749, | |
| "train_steps_per_second": 0.62 | |
| } | |
| ], | |
| "max_steps": 2900, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.6571394113263534e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |