| { | |
| "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": 7.000000000000001e-07, | |
| "loss": 9.0475, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.7000000000000002e-06, | |
| "loss": 8.4641, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 2.7e-06, | |
| "loss": 8.0099, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 3.7e-06, | |
| "loss": 8.1466, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 4.7e-06, | |
| "loss": 8.2443, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.7000000000000005e-06, | |
| "loss": 7.8136, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 6.6e-06, | |
| "loss": 7.1746, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 7.6e-06, | |
| "loss": 7.4995, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 8.599999999999999e-06, | |
| "loss": 6.8225, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 9.600000000000001e-06, | |
| "loss": 5.9793, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 5.453220844268799, | |
| "eval_runtime": 64.7883, | |
| "eval_samples_per_second": 25.931, | |
| "eval_steps_per_second": 25.931, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.06e-05, | |
| "loss": 5.2503, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.16e-05, | |
| "loss": 4.5578, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.2600000000000001e-05, | |
| "loss": 3.8024, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 1.3600000000000002e-05, | |
| "loss": 3.408, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.4599999999999999e-05, | |
| "loss": 3.2535, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.56e-05, | |
| "loss": 3.0941, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.66e-05, | |
| "loss": 2.9925, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.76e-05, | |
| "loss": 2.975, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.86e-05, | |
| "loss": 2.9489, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 2.9066, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 2.9070022106170654, | |
| "eval_runtime": 64.4624, | |
| "eval_samples_per_second": 26.062, | |
| "eval_steps_per_second": 26.062, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 2.06e-05, | |
| "loss": 2.8941, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 2.16e-05, | |
| "loss": 2.8987, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 2.26e-05, | |
| "loss": 2.8822, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.36e-05, | |
| "loss": 2.8357, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 2.46e-05, | |
| "loss": 2.8095, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 2.5600000000000002e-05, | |
| "loss": 2.7864, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 2.6600000000000003e-05, | |
| "loss": 2.6018, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 2.7600000000000003e-05, | |
| "loss": 2.5299, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.86e-05, | |
| "loss": 2.4706, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 2.96e-05, | |
| "loss": 2.2562, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.0323026180267334, | |
| "eval_runtime": 64.5834, | |
| "eval_samples_per_second": 26.013, | |
| "eval_steps_per_second": 26.013, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.06e-05, | |
| "loss": 1.9614, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.16e-05, | |
| "loss": 1.9041, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 3.26e-05, | |
| "loss": 1.8966, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 3.3600000000000004e-05, | |
| "loss": 1.6169, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.46e-05, | |
| "loss": 1.5506, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 3.56e-05, | |
| "loss": 1.6174, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 3.66e-05, | |
| "loss": 1.5146, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 3.76e-05, | |
| "loss": 1.2849, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.86e-05, | |
| "loss": 1.3535, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 3.960000000000001e-05, | |
| "loss": 1.5273, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 1.1509779691696167, | |
| "eval_runtime": 64.9222, | |
| "eval_samples_per_second": 25.877, | |
| "eval_steps_per_second": 25.877, | |
| "eval_wer": 0.8000826958858797, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4.0600000000000004e-05, | |
| "loss": 1.2342, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.16e-05, | |
| "loss": 1.1628, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 4.26e-05, | |
| "loss": 1.2844, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.36e-05, | |
| "loss": 1.3822, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.46e-05, | |
| "loss": 1.0556, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.5600000000000004e-05, | |
| "loss": 1.0931, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 4.660000000000001e-05, | |
| "loss": 1.2242, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.76e-05, | |
| "loss": 1.1836, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 4.86e-05, | |
| "loss": 1.0046, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.96e-05, | |
| "loss": 1.1085, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 0.9521023035049438, | |
| "eval_runtime": 64.7354, | |
| "eval_samples_per_second": 25.952, | |
| "eval_steps_per_second": 25.952, | |
| "eval_wer": 0.7052580800771828, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 5.0600000000000003e-05, | |
| "loss": 1.2836, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 5.16e-05, | |
| "loss": 0.9545, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.2600000000000005e-05, | |
| "loss": 0.9409, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.360000000000001e-05, | |
| "loss": 1.1573, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 5.4600000000000006e-05, | |
| "loss": 1.1732, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 5.560000000000001e-05, | |
| "loss": 0.9246, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 5.66e-05, | |
| "loss": 0.9752, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5.76e-05, | |
| "loss": 1.168, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 5.86e-05, | |
| "loss": 0.9875, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.96e-05, | |
| "loss": 0.813, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 0.8617281317710876, | |
| "eval_runtime": 64.7641, | |
| "eval_samples_per_second": 25.94, | |
| "eval_steps_per_second": 25.94, | |
| "eval_wer": 0.6701812418165529, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 6.06e-05, | |
| "loss": 0.9281, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.16e-05, | |
| "loss": 1.1197, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 6.26e-05, | |
| "loss": 0.8355, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 6.36e-05, | |
| "loss": 0.8804, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 6.460000000000001e-05, | |
| "loss": 1.0027, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 6.560000000000001e-05, | |
| "loss": 1.0293, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.66e-05, | |
| "loss": 0.7928, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 6.76e-05, | |
| "loss": 0.8941, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 6.860000000000001e-05, | |
| "loss": 1.0863, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 6.96e-05, | |
| "loss": 0.8434, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_loss": 0.8068093657493591, | |
| "eval_runtime": 64.8904, | |
| "eval_samples_per_second": 25.89, | |
| "eval_steps_per_second": 25.89, | |
| "eval_wer": 0.63930811108814, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 7.06e-05, | |
| "loss": 0.7419, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 7.16e-05, | |
| "loss": 0.8879, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.26e-05, | |
| "loss": 1.0322, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 7.36e-05, | |
| "loss": 0.7242, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 7.46e-05, | |
| "loss": 0.7617, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 7.560000000000001e-05, | |
| "loss": 0.9264, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 7.66e-05, | |
| "loss": 0.874, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 7.76e-05, | |
| "loss": 0.6869, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 7.860000000000001e-05, | |
| "loss": 0.784, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 7.960000000000001e-05, | |
| "loss": 0.9631, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_loss": 0.7862501740455627, | |
| "eval_runtime": 64.9413, | |
| "eval_samples_per_second": 25.87, | |
| "eval_steps_per_second": 25.87, | |
| "eval_wer": 0.6247674178209635, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 8.060000000000001e-05, | |
| "loss": 0.7428, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8.16e-05, | |
| "loss": 0.6836, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 8.26e-05, | |
| "loss": 0.8492, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 8.36e-05, | |
| "loss": 0.9554, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 8.46e-05, | |
| "loss": 0.6538, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 8.560000000000001e-05, | |
| "loss": 0.7178, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 8.66e-05, | |
| "loss": 0.9451, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 8.76e-05, | |
| "loss": 0.7119, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 8.86e-05, | |
| "loss": 0.6209, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 8.960000000000001e-05, | |
| "loss": 0.707, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 0.7476431131362915, | |
| "eval_runtime": 65.0738, | |
| "eval_samples_per_second": 25.817, | |
| "eval_steps_per_second": 25.817, | |
| "eval_wer": 0.5972710357659706, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 9.06e-05, | |
| "loss": 0.9038, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 9.16e-05, | |
| "loss": 0.6738, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 9.260000000000001e-05, | |
| "loss": 0.6604, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 9.360000000000001e-05, | |
| "loss": 0.7715, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 9.46e-05, | |
| "loss": 0.8073, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 9.56e-05, | |
| "loss": 0.5744, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 9.66e-05, | |
| "loss": 0.6418, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 9.76e-05, | |
| "loss": 0.9057, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 9.86e-05, | |
| "loss": 0.6742, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 9.960000000000001e-05, | |
| "loss": 0.5568, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 0.7349711060523987, | |
| "eval_runtime": 65.1738, | |
| "eval_samples_per_second": 25.777, | |
| "eval_steps_per_second": 25.777, | |
| "eval_wer": 0.5910688443249948, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 9.968421052631578e-05, | |
| "loss": 0.6777, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 9.915789473684211e-05, | |
| "loss": 0.8442, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 9.863157894736842e-05, | |
| "loss": 0.556, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 9.810526315789475e-05, | |
| "loss": 0.5375, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 9.757894736842106e-05, | |
| "loss": 0.7289, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 9.705263157894738e-05, | |
| "loss": 0.6911, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 9.652631578947369e-05, | |
| "loss": 0.4931, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.5905, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 9.547368421052632e-05, | |
| "loss": 0.8086, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 9.494736842105264e-05, | |
| "loss": 0.6171, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 0.7171187400817871, | |
| "eval_runtime": 65.2384, | |
| "eval_samples_per_second": 25.752, | |
| "eval_steps_per_second": 25.752, | |
| "eval_wer": 0.5841086072634553, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 9.442105263157895e-05, | |
| "loss": 0.5214, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 9.389473684210527e-05, | |
| "loss": 0.6501, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 9.336842105263158e-05, | |
| "loss": 0.7331, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 9.28421052631579e-05, | |
| "loss": 0.4932, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 9.231578947368421e-05, | |
| "loss": 0.5721, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 9.178947368421052e-05, | |
| "loss": 0.7402, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 9.126315789473685e-05, | |
| "loss": 0.5805, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 9.073684210526316e-05, | |
| "loss": 0.4518, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 9.021052631578947e-05, | |
| "loss": 0.5277, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 8.96842105263158e-05, | |
| "loss": 0.7011, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_loss": 0.7318134903907776, | |
| "eval_runtime": 65.2494, | |
| "eval_samples_per_second": 25.747, | |
| "eval_steps_per_second": 25.747, | |
| "eval_wer": 0.5797670732547723, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 8.915789473684211e-05, | |
| "loss": 0.4753, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 8.863157894736842e-05, | |
| "loss": 0.4837, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 8.810526315789475e-05, | |
| "loss": 0.6082, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 8.757894736842106e-05, | |
| "loss": 0.6443, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 8.705263157894738e-05, | |
| "loss": 0.4504, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 8.652631578947369e-05, | |
| "loss": 0.4878, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 8.6e-05, | |
| "loss": 0.6947, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 8.547368421052632e-05, | |
| "loss": 0.5122, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 8.494736842105263e-05, | |
| "loss": 0.4533, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 8.442105263157896e-05, | |
| "loss": 0.5546, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 0.74466872215271, | |
| "eval_runtime": 65.1965, | |
| "eval_samples_per_second": 25.768, | |
| "eval_steps_per_second": 25.768, | |
| "eval_wer": 0.5766659775342844, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 8.389473684210527e-05, | |
| "loss": 0.6444, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 8.336842105263158e-05, | |
| "loss": 0.3923, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 8.28421052631579e-05, | |
| "loss": 0.4217, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 8.231578947368421e-05, | |
| "loss": 0.5629, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 8.178947368421052e-05, | |
| "loss": 0.561, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 8.126315789473685e-05, | |
| "loss": 0.4181, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 8.073684210526316e-05, | |
| "loss": 0.4596, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 8.021052631578949e-05, | |
| "loss": 0.6366, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 7.96842105263158e-05, | |
| "loss": 0.4342, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 7.915789473684211e-05, | |
| "loss": 0.4278, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 0.7480624318122864, | |
| "eval_runtime": 65.275, | |
| "eval_samples_per_second": 25.737, | |
| "eval_steps_per_second": 25.737, | |
| "eval_wer": 0.5649507270346633, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 7.863157894736843e-05, | |
| "loss": 0.5455, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 7.810526315789474e-05, | |
| "loss": 0.6034, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 7.757894736842105e-05, | |
| "loss": 0.4034, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 7.705263157894738e-05, | |
| "loss": 0.4176, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 7.652631578947369e-05, | |
| "loss": 0.5848, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 7.6e-05, | |
| "loss": 0.4504, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 7.547368421052632e-05, | |
| "loss": 0.3659, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 7.494736842105263e-05, | |
| "loss": 0.4167, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 7.442105263157894e-05, | |
| "loss": 0.6001, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 7.389473684210527e-05, | |
| "loss": 0.3576, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.744252622127533, | |
| "eval_runtime": 65.0561, | |
| "eval_samples_per_second": 25.824, | |
| "eval_steps_per_second": 25.824, | |
| "eval_wer": 0.5712907449521053, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 7.336842105263158e-05, | |
| "loss": 0.3586, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 7.28421052631579e-05, | |
| "loss": 0.4736, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 7.231578947368421e-05, | |
| "loss": 0.5345, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 7.178947368421054e-05, | |
| "loss": 0.3812, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 7.126315789473685e-05, | |
| "loss": 0.434, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 7.073684210526316e-05, | |
| "loss": 0.6021, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 7.021052631578949e-05, | |
| "loss": 0.4438, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 6.96842105263158e-05, | |
| "loss": 0.3657, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 6.915789473684211e-05, | |
| "loss": 0.422, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 6.863157894736843e-05, | |
| "loss": 0.5506, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_loss": 0.7573680877685547, | |
| "eval_runtime": 65.03, | |
| "eval_samples_per_second": 25.834, | |
| "eval_steps_per_second": 25.834, | |
| "eval_wer": 0.5663979050375577, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 6.810526315789474e-05, | |
| "loss": 0.3659, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 6.757894736842105e-05, | |
| "loss": 0.3548, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 6.705263157894737e-05, | |
| "loss": 0.4603, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 6.652631578947369e-05, | |
| "loss": 0.4728, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 6.6e-05, | |
| "loss": 0.3358, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 6.547368421052632e-05, | |
| "loss": 0.3769, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 6.494736842105263e-05, | |
| "loss": 0.5439, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 6.442105263157894e-05, | |
| "loss": 0.3554, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 6.389473684210526e-05, | |
| "loss": 0.3352, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 6.336842105263158e-05, | |
| "loss": 0.4127, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 0.804313600063324, | |
| "eval_runtime": 65.3559, | |
| "eval_samples_per_second": 25.705, | |
| "eval_steps_per_second": 25.705, | |
| "eval_wer": 0.5630900696023706, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 6.28421052631579e-05, | |
| "loss": 0.5517, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 6.231578947368422e-05, | |
| "loss": 0.3276, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 6.178947368421054e-05, | |
| "loss": 0.402, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 6.126315789473685e-05, | |
| "loss": 0.5447, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 6.073684210526316e-05, | |
| "loss": 0.3825, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 6.0210526315789475e-05, | |
| "loss": 0.3153, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 5.968421052631579e-05, | |
| "loss": 0.376, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 5.9157894736842114e-05, | |
| "loss": 0.516, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 5.863157894736843e-05, | |
| "loss": 0.3347, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 5.810526315789474e-05, | |
| "loss": 0.3251, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 0.7738037705421448, | |
| "eval_runtime": 64.871, | |
| "eval_samples_per_second": 25.898, | |
| "eval_steps_per_second": 25.898, | |
| "eval_wer": 0.5549583074908689, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 5.7578947368421053e-05, | |
| "loss": 0.4643, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 5.7052631578947366e-05, | |
| "loss": 0.4798, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 5.652631578947368e-05, | |
| "loss": 0.2974, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.3234, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 5.547368421052632e-05, | |
| "loss": 0.5002, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 5.494736842105264e-05, | |
| "loss": 0.3643, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 5.442105263157895e-05, | |
| "loss": 0.2987, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 5.3894736842105265e-05, | |
| "loss": 0.3989, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 5.336842105263158e-05, | |
| "loss": 0.4696, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 5.284210526315789e-05, | |
| "loss": 0.3119, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "eval_loss": 0.7829384207725525, | |
| "eval_runtime": 64.82, | |
| "eval_samples_per_second": 25.918, | |
| "eval_steps_per_second": 25.918, | |
| "eval_wer": 0.5515815588174489, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 5.231578947368422e-05, | |
| "loss": 0.2997, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 5.178947368421053e-05, | |
| "loss": 0.4126, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 5.126315789473684e-05, | |
| "loss": 0.382, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 5.073684210526316e-05, | |
| "loss": 0.2903, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 5.0210526315789476e-05, | |
| "loss": 0.3354, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 4.9684210526315796e-05, | |
| "loss": 0.4861, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 4.915789473684211e-05, | |
| "loss": 0.3228, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 4.863157894736842e-05, | |
| "loss": 0.285, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 4.8105263157894735e-05, | |
| "loss": 0.4209, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 4.7578947368421054e-05, | |
| "loss": 0.4371, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.8024688959121704, | |
| "eval_runtime": 65.0217, | |
| "eval_samples_per_second": 25.838, | |
| "eval_steps_per_second": 25.838, | |
| "eval_wer": 0.5556474398731996, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 4.705263157894737e-05, | |
| "loss": 0.2753, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 4.652631578947369e-05, | |
| "loss": 0.3165, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.4676, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 4.547368421052632e-05, | |
| "loss": 0.3263, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 4.494736842105263e-05, | |
| "loss": 0.2557, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 4.442105263157895e-05, | |
| "loss": 0.3331, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 4.3894736842105266e-05, | |
| "loss": 0.4421, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 4.336842105263158e-05, | |
| "loss": 0.2897, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 4.284210526315789e-05, | |
| "loss": 0.2944, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 4.231578947368421e-05, | |
| "loss": 0.3772, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 0.8450560569763184, | |
| "eval_runtime": 64.8033, | |
| "eval_samples_per_second": 25.925, | |
| "eval_steps_per_second": 25.925, | |
| "eval_wer": 0.555923092826132, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 4.178947368421053e-05, | |
| "loss": 0.4119, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 4.1263157894736844e-05, | |
| "loss": 0.2626, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 4.0736842105263164e-05, | |
| "loss": 0.3358, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 4.021052631578948e-05, | |
| "loss": 0.4507, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 3.968421052631579e-05, | |
| "loss": 0.3014, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 3.91578947368421e-05, | |
| "loss": 0.2568, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 3.863157894736842e-05, | |
| "loss": 0.325, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 3.8105263157894735e-05, | |
| "loss": 0.4206, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 3.7578947368421055e-05, | |
| "loss": 0.2509, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 3.7052631578947375e-05, | |
| "loss": 0.2942, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "eval_loss": 0.8300282955169678, | |
| "eval_runtime": 64.8139, | |
| "eval_samples_per_second": 25.92, | |
| "eval_steps_per_second": 25.92, | |
| "eval_wer": 0.5556474398731996, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 3.652631578947369e-05, | |
| "loss": 0.4212, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.3379, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 3.547368421052632e-05, | |
| "loss": 0.2652, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 3.4947368421052634e-05, | |
| "loss": 0.2921, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 3.442105263157895e-05, | |
| "loss": 0.4361, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 3.389473684210526e-05, | |
| "loss": 0.2628, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 3.336842105263158e-05, | |
| "loss": 0.2929, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 3.28421052631579e-05, | |
| "loss": 0.3727, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 3.231578947368421e-05, | |
| "loss": 0.4145, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 3.178947368421053e-05, | |
| "loss": 0.2503, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 0.8417431712150574, | |
| "eval_runtime": 64.6266, | |
| "eval_samples_per_second": 25.995, | |
| "eval_steps_per_second": 25.995, | |
| "eval_wer": 0.5541313486320723, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 3.1263157894736845e-05, | |
| "loss": 0.2859, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.073684210526316e-05, | |
| "loss": 0.4291, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 3.021052631578947e-05, | |
| "loss": 0.2987, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 2.968421052631579e-05, | |
| "loss": 0.2357, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 2.9157894736842107e-05, | |
| "loss": 0.3266, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.863157894736842e-05, | |
| "loss": 0.4187, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 2.810526315789474e-05, | |
| "loss": 0.2507, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 2.7578947368421053e-05, | |
| "loss": 0.2876, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 2.705263157894737e-05, | |
| "loss": 0.3606, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 2.652631578947369e-05, | |
| "loss": 0.3671, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 0.8567938208580017, | |
| "eval_runtime": 65.0478, | |
| "eval_samples_per_second": 25.827, | |
| "eval_steps_per_second": 25.827, | |
| "eval_wer": 0.552821997105644, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.2398, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 2.5473684210526315e-05, | |
| "loss": 0.292, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 2.4947368421052635e-05, | |
| "loss": 0.397, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 2.4421052631578948e-05, | |
| "loss": 0.2781, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 2.3894736842105264e-05, | |
| "loss": 0.2663, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 2.336842105263158e-05, | |
| "loss": 0.3187, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.2842105263157897e-05, | |
| "loss": 0.392, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 2.2315789473684213e-05, | |
| "loss": 0.2491, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 2.1789473684210526e-05, | |
| "loss": 0.2545, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 2.1263157894736842e-05, | |
| "loss": 0.3867, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 0.8520529866218567, | |
| "eval_runtime": 64.8285, | |
| "eval_samples_per_second": 25.915, | |
| "eval_steps_per_second": 25.915, | |
| "eval_wer": 0.5510302529115844, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 2.073684210526316e-05, | |
| "loss": 0.3364, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 2.0210526315789475e-05, | |
| "loss": 0.2243, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 1.968421052631579e-05, | |
| "loss": 0.2874, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 1.9157894736842104e-05, | |
| "loss": 0.4084, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 1.863157894736842e-05, | |
| "loss": 0.2578, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 1.810526315789474e-05, | |
| "loss": 0.26, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 1.7578947368421054e-05, | |
| "loss": 0.3482, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 1.705263157894737e-05, | |
| "loss": 0.3769, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 1.6526315789473683e-05, | |
| "loss": 0.243, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.2614, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "eval_loss": 0.8478676676750183, | |
| "eval_runtime": 65.0427, | |
| "eval_samples_per_second": 25.829, | |
| "eval_steps_per_second": 25.829, | |
| "eval_wer": 0.5522706911997795, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.5473684210526316e-05, | |
| "loss": 0.4247, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 1.4947368421052632e-05, | |
| "loss": 0.2766, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 1.4421052631578948e-05, | |
| "loss": 0.2311, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 18.21, | |
| "learning_rate": 1.3894736842105263e-05, | |
| "loss": 0.2888, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 1.336842105263158e-05, | |
| "loss": 0.3954, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 1.2842105263157894e-05, | |
| "loss": 0.2502, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 1.231578947368421e-05, | |
| "loss": 0.2557, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 1.1789473684210527e-05, | |
| "loss": 0.3494, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 1.1263157894736842e-05, | |
| "loss": 0.3442, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 1.073684210526316e-05, | |
| "loss": 0.2441, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 0.8557784557342529, | |
| "eval_runtime": 65.0177, | |
| "eval_samples_per_second": 25.839, | |
| "eval_steps_per_second": 25.839, | |
| "eval_wer": 0.5494452484322239, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 1.0210526315789474e-05, | |
| "loss": 0.2735, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 9.68421052631579e-06, | |
| "loss": 0.3609, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 9.157894736842105e-06, | |
| "loss": 0.2519, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 8.631578947368422e-06, | |
| "loss": 0.2632, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 8.105263157894736e-06, | |
| "loss": 0.3227, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 7.578947368421054e-06, | |
| "loss": 0.3494, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 7.052631578947369e-06, | |
| "loss": 0.2321, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "learning_rate": 6.526315789473685e-06, | |
| "loss": 0.2746, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 6e-06, | |
| "loss": 0.383, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 5.4736842105263165e-06, | |
| "loss": 0.3059, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "eval_loss": 0.855342447757721, | |
| "eval_runtime": 64.851, | |
| "eval_samples_per_second": 25.906, | |
| "eval_steps_per_second": 25.906, | |
| "eval_wer": 0.547446764523465, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 4.947368421052632e-06, | |
| "loss": 0.2364, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 4.4210526315789476e-06, | |
| "loss": 0.2962, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 3.894736842105264e-06, | |
| "loss": 0.3882, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 3.368421052631579e-06, | |
| "loss": 0.2342, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "learning_rate": 2.842105263157895e-06, | |
| "loss": 0.2409, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 19.72, | |
| "learning_rate": 2.315789473684211e-06, | |
| "loss": 0.322, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 1.7894736842105262e-06, | |
| "loss": 0.3532, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 1.2631578947368422e-06, | |
| "loss": 0.2274, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 7.368421052631579e-07, | |
| "loss": 0.2696, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2.105263157894737e-07, | |
| "loss": 0.3734, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.8561776280403137, | |
| "eval_runtime": 65.0405, | |
| "eval_samples_per_second": 25.83, | |
| "eval_steps_per_second": 25.83, | |
| "eval_wer": 0.5484115498587279, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2900, | |
| "total_flos": 2.6574132447000617e+18, | |
| "train_loss": 0.9877855575495753, | |
| "train_runtime": 5510.014, | |
| "train_samples_per_second": 16.769, | |
| "train_steps_per_second": 0.526 | |
| } | |
| ], | |
| "max_steps": 2900, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.6574132447000617e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |