| { | |
| "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": 9.653, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.8e-06, | |
| "loss": 8.4639, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.8000000000000003e-06, | |
| "loss": 7.9446, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.7e-06, | |
| "loss": 8.6256, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.7e-06, | |
| "loss": 8.0756, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.7000000000000005e-06, | |
| "loss": 7.4432, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.700000000000001e-06, | |
| "loss": 6.5524, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.7e-06, | |
| "loss": 6.5869, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.7e-06, | |
| "loss": 5.2277, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.7e-06, | |
| "loss": 4.2247, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 3.8606975078582764, | |
| "eval_runtime": 48.7411, | |
| "eval_samples_per_second": 34.468, | |
| "eval_steps_per_second": 34.468, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.0700000000000001e-05, | |
| "loss": 3.6454, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.1700000000000001e-05, | |
| "loss": 3.3306, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.27e-05, | |
| "loss": 3.1763, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3700000000000001e-05, | |
| "loss": 3.0708, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.47e-05, | |
| "loss": 3.0836, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.5700000000000002e-05, | |
| "loss": 2.999, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.6700000000000003e-05, | |
| "loss": 2.9717, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.77e-05, | |
| "loss": 2.9846, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.87e-05, | |
| "loss": 2.9912, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.97e-05, | |
| "loss": 2.9444, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.950913667678833, | |
| "eval_runtime": 49.1765, | |
| "eval_samples_per_second": 34.163, | |
| "eval_steps_per_second": 34.163, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.07e-05, | |
| "loss": 2.9386, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.1700000000000002e-05, | |
| "loss": 2.9601, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.2700000000000003e-05, | |
| "loss": 2.9395, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.37e-05, | |
| "loss": 2.9159, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.47e-05, | |
| "loss": 2.9124, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.57e-05, | |
| "loss": 2.9412, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 2.6700000000000002e-05, | |
| "loss": 2.8826, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 2.7700000000000002e-05, | |
| "loss": 2.8915, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.87e-05, | |
| "loss": 2.9096, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.97e-05, | |
| "loss": 2.8858, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.844599723815918, | |
| "eval_runtime": 49.0987, | |
| "eval_samples_per_second": 34.217, | |
| "eval_steps_per_second": 34.217, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.07e-05, | |
| "loss": 2.8025, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.1700000000000005e-05, | |
| "loss": 2.7678, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 3.27e-05, | |
| "loss": 2.7655, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.3700000000000006e-05, | |
| "loss": 2.6538, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.4699999999999996e-05, | |
| "loss": 2.5774, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.57e-05, | |
| "loss": 2.5132, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.6700000000000004e-05, | |
| "loss": 2.4977, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.77e-05, | |
| "loss": 2.3494, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.8700000000000006e-05, | |
| "loss": 2.2912, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.97e-05, | |
| "loss": 2.2804, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 2.1984939575195312, | |
| "eval_runtime": 49.2048, | |
| "eval_samples_per_second": 34.143, | |
| "eval_steps_per_second": 34.143, | |
| "eval_wer": 1.0013782647646612, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.07e-05, | |
| "loss": 2.1559, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.17e-05, | |
| "loss": 2.013, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.27e-05, | |
| "loss": 1.9527, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.3700000000000005e-05, | |
| "loss": 1.9805, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.47e-05, | |
| "loss": 1.7592, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.5700000000000006e-05, | |
| "loss": 1.6803, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.6700000000000003e-05, | |
| "loss": 1.674, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.77e-05, | |
| "loss": 1.6908, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.87e-05, | |
| "loss": 1.4964, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.97e-05, | |
| "loss": 1.505, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 1.4971957206726074, | |
| "eval_runtime": 49.3444, | |
| "eval_samples_per_second": 34.046, | |
| "eval_steps_per_second": 34.046, | |
| "eval_wer": 0.9609261939218524, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.0700000000000006e-05, | |
| "loss": 1.5943, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.17e-05, | |
| "loss": 1.3954, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.270000000000001e-05, | |
| "loss": 1.3033, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.3700000000000004e-05, | |
| "loss": 1.4023, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 5.470000000000001e-05, | |
| "loss": 1.4554, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 5.5700000000000005e-05, | |
| "loss": 1.2369, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 5.6699999999999996e-05, | |
| "loss": 1.2277, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.77e-05, | |
| "loss": 1.3168, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 5.87e-05, | |
| "loss": 1.2854, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.97e-05, | |
| "loss": 1.06, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 1.2013553380966187, | |
| "eval_runtime": 49.3577, | |
| "eval_samples_per_second": 34.037, | |
| "eval_steps_per_second": 34.037, | |
| "eval_wer": 0.805802494659224, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 6.07e-05, | |
| "loss": 1.1026, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.170000000000001e-05, | |
| "loss": 1.2475, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.27e-05, | |
| "loss": 1.0605, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 6.37e-05, | |
| "loss": 1.0361, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 6.47e-05, | |
| "loss": 1.1079, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 6.570000000000001e-05, | |
| "loss": 1.2017, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.670000000000001e-05, | |
| "loss": 0.9803, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.77e-05, | |
| "loss": 1.0294, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 6.87e-05, | |
| "loss": 1.1393, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 6.97e-05, | |
| "loss": 1.0166, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_loss": 1.0605226755142212, | |
| "eval_runtime": 49.5346, | |
| "eval_samples_per_second": 33.916, | |
| "eval_steps_per_second": 33.916, | |
| "eval_wer": 0.7535662600785611, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.07e-05, | |
| "loss": 0.8619, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 7.17e-05, | |
| "loss": 0.9549, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.27e-05, | |
| "loss": 1.1162, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 7.37e-05, | |
| "loss": 0.8634, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 7.47e-05, | |
| "loss": 0.8374, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 7.570000000000001e-05, | |
| "loss": 0.9316, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 7.670000000000001e-05, | |
| "loss": 0.9827, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 7.77e-05, | |
| "loss": 0.7866, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.87e-05, | |
| "loss": 0.816, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 7.970000000000001e-05, | |
| "loss": 0.966, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_loss": 0.9963059425354004, | |
| "eval_runtime": 49.4475, | |
| "eval_samples_per_second": 33.975, | |
| "eval_steps_per_second": 33.975, | |
| "eval_wer": 0.7100820067534973, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 8.070000000000001e-05, | |
| "loss": 0.8691, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8.17e-05, | |
| "loss": 0.748, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 8.27e-05, | |
| "loss": 0.8441, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 8.37e-05, | |
| "loss": 1.0217, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 8.47e-05, | |
| "loss": 0.7328, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.57e-05, | |
| "loss": 0.7625, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.67e-05, | |
| "loss": 0.883, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 8.77e-05, | |
| "loss": 0.8376, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 8.87e-05, | |
| "loss": 0.6518, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 8.970000000000001e-05, | |
| "loss": 0.6857, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 0.9442995190620422, | |
| "eval_runtime": 49.4918, | |
| "eval_samples_per_second": 33.945, | |
| "eval_steps_per_second": 33.945, | |
| "eval_wer": 0.6898215147129764, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 9.070000000000001e-05, | |
| "loss": 0.8764, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 9.17e-05, | |
| "loss": 0.7431, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 9.27e-05, | |
| "loss": 0.6688, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 9.370000000000001e-05, | |
| "loss": 0.7162, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 9.47e-05, | |
| "loss": 0.8465, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 9.57e-05, | |
| "loss": 0.6191, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 9.67e-05, | |
| "loss": 0.6149, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 9.77e-05, | |
| "loss": 0.8114, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 9.87e-05, | |
| "loss": 0.7832, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 9.970000000000001e-05, | |
| "loss": 0.5859, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 0.9042919278144836, | |
| "eval_runtime": 49.5927, | |
| "eval_samples_per_second": 33.876, | |
| "eval_steps_per_second": 33.876, | |
| "eval_wer": 0.6795534422162497, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 9.963157894736843e-05, | |
| "loss": 0.6366, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 9.910526315789475e-05, | |
| "loss": 0.8343, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 9.857894736842106e-05, | |
| "loss": 0.5977, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 9.805263157894737e-05, | |
| "loss": 0.5143, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 9.752631578947369e-05, | |
| "loss": 0.5725, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 9.7e-05, | |
| "loss": 0.7447, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 9.647368421052631e-05, | |
| "loss": 0.4958, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 9.594736842105264e-05, | |
| "loss": 0.5129, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 9.542105263157895e-05, | |
| "loss": 0.6971, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 9.489473684210527e-05, | |
| "loss": 0.6812, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 0.9094838500022888, | |
| "eval_runtime": 49.4365, | |
| "eval_samples_per_second": 33.983, | |
| "eval_steps_per_second": 33.983, | |
| "eval_wer": 0.6715595065812142, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 9.436842105263158e-05, | |
| "loss": 0.4968, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 9.384210526315789e-05, | |
| "loss": 0.5241, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 9.331578947368422e-05, | |
| "loss": 0.7348, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 9.278947368421053e-05, | |
| "loss": 0.5148, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 9.226315789473686e-05, | |
| "loss": 0.518, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.173684210526317e-05, | |
| "loss": 0.6009, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 9.121052631578948e-05, | |
| "loss": 0.6677, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 9.06842105263158e-05, | |
| "loss": 0.4171, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 9.015789473684211e-05, | |
| "loss": 0.4129, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 8.963157894736842e-05, | |
| "loss": 0.6088, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_loss": 0.9421910047531128, | |
| "eval_runtime": 50.1077, | |
| "eval_samples_per_second": 33.528, | |
| "eval_steps_per_second": 33.528, | |
| "eval_wer": 0.6677003652401626, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 8.910526315789474e-05, | |
| "loss": 0.4936, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 8.857894736842106e-05, | |
| "loss": 0.4166, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 8.805263157894737e-05, | |
| "loss": 0.4619, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 8.752631578947369e-05, | |
| "loss": 0.6438, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.7e-05, | |
| "loss": 0.4375, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 8.647368421052631e-05, | |
| "loss": 0.3933, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 8.594736842105263e-05, | |
| "loss": 0.5451, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 8.542105263157894e-05, | |
| "loss": 0.5696, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 8.489473684210527e-05, | |
| "loss": 0.3988, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 8.436842105263158e-05, | |
| "loss": 0.4162, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 0.9547951817512512, | |
| "eval_runtime": 49.7551, | |
| "eval_samples_per_second": 33.765, | |
| "eval_steps_per_second": 33.765, | |
| "eval_wer": 0.6657018813314037, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 8.38421052631579e-05, | |
| "loss": 0.581, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 8.331578947368422e-05, | |
| "loss": 0.3721, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 8.278947368421053e-05, | |
| "loss": 0.3251, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 8.226315789473684e-05, | |
| "loss": 0.3486, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 8.173684210526317e-05, | |
| "loss": 0.5894, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 8.121052631578948e-05, | |
| "loss": 0.3616, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 8.06842105263158e-05, | |
| "loss": 0.3189, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 8.015789473684211e-05, | |
| "loss": 0.4903, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 7.963157894736842e-05, | |
| "loss": 0.4564, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 7.910526315789474e-05, | |
| "loss": 0.3411, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 0.9900805950164795, | |
| "eval_runtime": 49.6853, | |
| "eval_samples_per_second": 33.813, | |
| "eval_steps_per_second": 33.813, | |
| "eval_wer": 0.6688718902901247, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 7.857894736842105e-05, | |
| "loss": 0.3415, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 7.805263157894738e-05, | |
| "loss": 0.5735, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 7.752631578947369e-05, | |
| "loss": 0.3583, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 7.7e-05, | |
| "loss": 0.2923, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.647368421052631e-05, | |
| "loss": 0.3396, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 7.594736842105263e-05, | |
| "loss": 0.493, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.542105263157895e-05, | |
| "loss": 0.2637, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 7.489473684210527e-05, | |
| "loss": 0.2342, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 7.43684210526316e-05, | |
| "loss": 0.4601, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 7.38421052631579e-05, | |
| "loss": 0.3323, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.9995749592781067, | |
| "eval_runtime": 49.7001, | |
| "eval_samples_per_second": 33.803, | |
| "eval_steps_per_second": 33.803, | |
| "eval_wer": 0.6637723106608779, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 7.331578947368422e-05, | |
| "loss": 0.2467, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 7.278947368421053e-05, | |
| "loss": 0.2583, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 7.226315789473685e-05, | |
| "loss": 0.5144, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 7.173684210526316e-05, | |
| "loss": 0.2957, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 7.121052631578947e-05, | |
| "loss": 0.2613, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 7.06842105263158e-05, | |
| "loss": 0.3751, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 7.015789473684211e-05, | |
| "loss": 0.4468, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 6.963157894736842e-05, | |
| "loss": 0.2497, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 6.910526315789474e-05, | |
| "loss": 0.2478, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 6.857894736842105e-05, | |
| "loss": 0.431, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_loss": 1.0520888566970825, | |
| "eval_runtime": 49.5966, | |
| "eval_samples_per_second": 33.873, | |
| "eval_steps_per_second": 33.873, | |
| "eval_wer": 0.6708014609606505, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 6.805263157894736e-05, | |
| "loss": 0.2673, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 6.752631578947368e-05, | |
| "loss": 0.1873, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 6.7e-05, | |
| "loss": 0.1937, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 6.647368421052632e-05, | |
| "loss": 0.48, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 6.594736842105264e-05, | |
| "loss": 0.2313, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 6.542105263157895e-05, | |
| "loss": 0.1876, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 6.489473684210527e-05, | |
| "loss": 0.3463, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 6.436842105263158e-05, | |
| "loss": 0.3318, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 6.384210526315791e-05, | |
| "loss": 0.1915, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 6.331578947368422e-05, | |
| "loss": 0.2029, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 1.0945649147033691, | |
| "eval_runtime": 49.6143, | |
| "eval_samples_per_second": 33.861, | |
| "eval_steps_per_second": 33.861, | |
| "eval_wer": 0.6792777892633175, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 6.278947368421053e-05, | |
| "loss": 0.4609, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 6.226315789473685e-05, | |
| "loss": 0.2301, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 6.173684210526316e-05, | |
| "loss": 0.2117, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.121052631578947e-05, | |
| "loss": 0.2205, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 6.0684210526315785e-05, | |
| "loss": 0.396, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 6.015789473684211e-05, | |
| "loss": 0.1706, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 5.9631578947368425e-05, | |
| "loss": 0.1486, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 5.9105263157894744e-05, | |
| "loss": 0.3378, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 5.857894736842106e-05, | |
| "loss": 0.2552, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 5.805263157894737e-05, | |
| "loss": 0.1424, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 1.1287890672683716, | |
| "eval_runtime": 49.6023, | |
| "eval_samples_per_second": 33.869, | |
| "eval_steps_per_second": 33.869, | |
| "eval_wer": 0.6712149403900489, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 5.752631578947368e-05, | |
| "loss": 0.1614, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 0.3994, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 5.647368421052632e-05, | |
| "loss": 0.1802, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 5.5947368421052636e-05, | |
| "loss": 0.1493, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 5.542105263157895e-05, | |
| "loss": 0.2477, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 5.489473684210527e-05, | |
| "loss": 0.3343, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 5.436842105263158e-05, | |
| "loss": 0.1604, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 5.3842105263157895e-05, | |
| "loss": 0.1511, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 5.331578947368421e-05, | |
| "loss": 0.3277, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 5.2789473684210534e-05, | |
| "loss": 0.1922, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "eval_loss": 1.1456407308578491, | |
| "eval_runtime": 49.7045, | |
| "eval_samples_per_second": 33.8, | |
| "eval_steps_per_second": 33.8, | |
| "eval_wer": 0.6739714699193715, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 5.226315789473685e-05, | |
| "loss": 0.1204, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 5.173684210526316e-05, | |
| "loss": 0.1144, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 5.121052631578947e-05, | |
| "loss": 0.373, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 5.068421052631579e-05, | |
| "loss": 0.1471, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 5.0157894736842106e-05, | |
| "loss": 0.1108, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 4.9631578947368426e-05, | |
| "loss": 0.2611, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 4.910526315789474e-05, | |
| "loss": 0.2552, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 4.857894736842106e-05, | |
| "loss": 0.1166, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 4.805263157894737e-05, | |
| "loss": 0.124, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 4.7526315789473684e-05, | |
| "loss": 0.326, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 1.2077220678329468, | |
| "eval_runtime": 49.5752, | |
| "eval_samples_per_second": 33.888, | |
| "eval_steps_per_second": 33.888, | |
| "eval_wer": 0.6914754324305699, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 4.7e-05, | |
| "loss": 0.1467, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 4.647368421052632e-05, | |
| "loss": 0.1127, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.594736842105264e-05, | |
| "loss": 0.1298, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 4.542105263157895e-05, | |
| "loss": 0.3033, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 4.489473684210527e-05, | |
| "loss": 0.0986, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.436842105263158e-05, | |
| "loss": 0.0835, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 4.3842105263157895e-05, | |
| "loss": 0.2537, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.3315789473684215e-05, | |
| "loss": 0.1715, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.278947368421053e-05, | |
| "loss": 0.1022, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 4.226315789473684e-05, | |
| "loss": 0.0892, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 1.2525088787078857, | |
| "eval_runtime": 49.6787, | |
| "eval_samples_per_second": 33.817, | |
| "eval_steps_per_second": 33.817, | |
| "eval_wer": 0.6795534422162497, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 4.1736842105263154e-05, | |
| "loss": 0.2858, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 4.1210526315789474e-05, | |
| "loss": 0.1159, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 4.0684210526315794e-05, | |
| "loss": 0.0943, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 4.015789473684211e-05, | |
| "loss": 0.1748, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 3.9631578947368426e-05, | |
| "loss": 0.2513, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 3.910526315789474e-05, | |
| "loss": 0.0915, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 3.857894736842105e-05, | |
| "loss": 0.0812, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3.8052631578947365e-05, | |
| "loss": 0.2623, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 3.7526315789473685e-05, | |
| "loss": 0.1312, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.0769, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "eval_loss": 1.2313077449798584, | |
| "eval_runtime": 49.7254, | |
| "eval_samples_per_second": 33.786, | |
| "eval_steps_per_second": 33.786, | |
| "eval_wer": 0.6736269037282062, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 3.647368421052632e-05, | |
| "loss": 0.0657, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 3.594736842105264e-05, | |
| "loss": 0.2479, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 3.542105263157895e-05, | |
| "loss": 0.1003, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 3.4894736842105264e-05, | |
| "loss": 0.0655, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 3.436842105263158e-05, | |
| "loss": 0.1793, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 3.3842105263157896e-05, | |
| "loss": 0.162, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 3.331578947368421e-05, | |
| "loss": 0.0824, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 3.278947368421052e-05, | |
| "loss": 0.0805, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 3.226315789473684e-05, | |
| "loss": 0.2688, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 3.173684210526316e-05, | |
| "loss": 0.0927, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 1.300140619277954, | |
| "eval_runtime": 49.6316, | |
| "eval_samples_per_second": 33.849, | |
| "eval_steps_per_second": 33.849, | |
| "eval_wer": 0.6863758528013232, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 3.1210526315789475e-05, | |
| "loss": 0.0762, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.0684210526315795e-05, | |
| "loss": 0.0758, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 3.0157894736842108e-05, | |
| "loss": 0.2356, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 2.963157894736842e-05, | |
| "loss": 0.0667, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 2.910526315789474e-05, | |
| "loss": 0.0607, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.8578947368421057e-05, | |
| "loss": 0.1887, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 2.805263157894737e-05, | |
| "loss": 0.1209, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 2.7526315789473683e-05, | |
| "loss": 0.0688, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 0.0572, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 2.647368421052632e-05, | |
| "loss": 0.232, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 1.349038004875183, | |
| "eval_runtime": 49.7195, | |
| "eval_samples_per_second": 33.79, | |
| "eval_steps_per_second": 33.79, | |
| "eval_wer": 0.6962993591068845, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 2.5947368421052632e-05, | |
| "loss": 0.0775, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 2.542105263157895e-05, | |
| "loss": 0.0604, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 2.4894736842105264e-05, | |
| "loss": 0.1308, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 2.436842105263158e-05, | |
| "loss": 0.1677, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 2.3842105263157897e-05, | |
| "loss": 0.0647, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 2.331578947368421e-05, | |
| "loss": 0.0554, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.2789473684210527e-05, | |
| "loss": 0.1842, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 2.2263157894736843e-05, | |
| "loss": 0.0917, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 2.173684210526316e-05, | |
| "loss": 0.0469, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.1210526315789476e-05, | |
| "loss": 0.0485, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 1.3268318176269531, | |
| "eval_runtime": 49.6454, | |
| "eval_samples_per_second": 33.84, | |
| "eval_steps_per_second": 33.84, | |
| "eval_wer": 0.6763145200192957, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.068421052631579e-05, | |
| "loss": 0.2096, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.0157894736842105e-05, | |
| "loss": 0.0561, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 1.963157894736842e-05, | |
| "loss": 0.0476, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 1.9105263157894738e-05, | |
| "loss": 0.1363, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 1.8578947368421054e-05, | |
| "loss": 0.1129, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 1.8052631578947367e-05, | |
| "loss": 0.0508, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 1.7526315789473683e-05, | |
| "loss": 0.0426, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 0.1895, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 1.6473684210526316e-05, | |
| "loss": 0.0756, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 1.5947368421052633e-05, | |
| "loss": 0.0487, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "eval_loss": 1.3375909328460693, | |
| "eval_runtime": 49.815, | |
| "eval_samples_per_second": 33.725, | |
| "eval_steps_per_second": 33.725, | |
| "eval_wer": 0.6779684377368893, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.5421052631578946e-05, | |
| "loss": 0.0497, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 1.4894736842105264e-05, | |
| "loss": 0.1801, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 1.4368421052631582e-05, | |
| "loss": 0.047, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 18.21, | |
| "learning_rate": 1.3842105263157895e-05, | |
| "loss": 0.044, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 1.3315789473684213e-05, | |
| "loss": 0.1409, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 1.2789473684210526e-05, | |
| "loss": 0.0887, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 1.2263157894736844e-05, | |
| "loss": 0.0456, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 1.1736842105263158e-05, | |
| "loss": 0.0403, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 1.1210526315789475e-05, | |
| "loss": 0.161, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 1.068421052631579e-05, | |
| "loss": 0.0607, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 1.3700547218322754, | |
| "eval_runtime": 49.648, | |
| "eval_samples_per_second": 33.838, | |
| "eval_steps_per_second": 33.838, | |
| "eval_wer": 0.6894769485218111, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 1.0157894736842106e-05, | |
| "loss": 0.0388, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 9.631578947368422e-06, | |
| "loss": 0.1065, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 9.105263157894737e-06, | |
| "loss": 0.1294, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 8.578947368421053e-06, | |
| "loss": 0.0507, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 8.052631578947368e-06, | |
| "loss": 0.0382, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 7.526315789473684e-06, | |
| "loss": 0.1584, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 0.0715, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "learning_rate": 6.473684210526316e-06, | |
| "loss": 0.0443, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 5.947368421052632e-06, | |
| "loss": 0.041, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 5.421052631578947e-06, | |
| "loss": 0.1618, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "eval_loss": 1.3657355308532715, | |
| "eval_runtime": 49.611, | |
| "eval_samples_per_second": 33.863, | |
| "eval_steps_per_second": 33.863, | |
| "eval_wer": 0.6795534422162497, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 4.894736842105263e-06, | |
| "loss": 0.0475, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 4.368421052631579e-06, | |
| "loss": 0.0417, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 3.842105263157895e-06, | |
| "loss": 0.0996, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 3.315789473684211e-06, | |
| "loss": 0.0923, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "learning_rate": 2.7894736842105266e-06, | |
| "loss": 0.0363, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 19.72, | |
| "learning_rate": 2.2631578947368426e-06, | |
| "loss": 0.0363, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.7368421052631579e-06, | |
| "loss": 0.1545, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 1.2105263157894738e-06, | |
| "loss": 0.0575, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 6.842105263157896e-07, | |
| "loss": 0.0372, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.5789473684210527e-07, | |
| "loss": 0.0415, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 1.3688374757766724, | |
| "eval_runtime": 49.8972, | |
| "eval_samples_per_second": 33.669, | |
| "eval_steps_per_second": 33.669, | |
| "eval_wer": 0.6817586658397078, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2900, | |
| "total_flos": 6.619750382049396e+17, | |
| "train_loss": 0.938711966728342, | |
| "train_runtime": 4007.4157, | |
| "train_samples_per_second": 23.057, | |
| "train_steps_per_second": 0.724 | |
| } | |
| ], | |
| "max_steps": 2900, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.619750382049396e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |