| { | |
| "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": 2.4000000000000003e-07, | |
| "loss": 9.5954, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 5.4e-07, | |
| "loss": 8.4532, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.4e-07, | |
| "loss": 8.0035, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 1.11e-06, | |
| "loss": 8.8373, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.41e-06, | |
| "loss": 8.4238, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.7100000000000001e-06, | |
| "loss": 8.0434, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 2.0100000000000002e-06, | |
| "loss": 7.5531, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 2.31e-06, | |
| "loss": 8.6097, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.61e-06, | |
| "loss": 7.6526, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 2.91e-06, | |
| "loss": 6.9958, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 6.717099189758301, | |
| "eval_runtime": 55.8689, | |
| "eval_samples_per_second": 30.07, | |
| "eval_steps_per_second": 30.07, | |
| "eval_wer": 1.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 3.21e-06, | |
| "loss": 6.8539, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 3.5100000000000003e-06, | |
| "loss": 5.8905, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.81e-06, | |
| "loss": 4.7961, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 4.1100000000000005e-06, | |
| "loss": 3.9155, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 4.41e-06, | |
| "loss": 3.5858, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 4.71e-06, | |
| "loss": 3.2806, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 5.01e-06, | |
| "loss": 3.1627, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 5.31e-06, | |
| "loss": 3.1167, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 5.6100000000000005e-06, | |
| "loss": 3.0981, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 5.91e-06, | |
| "loss": 3.0453, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_loss": 3.037445545196533, | |
| "eval_runtime": 55.8686, | |
| "eval_samples_per_second": 30.071, | |
| "eval_steps_per_second": 30.071, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 6.21e-06, | |
| "loss": 3.0244, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 6.51e-06, | |
| "loss": 3.0413, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 6.81e-06, | |
| "loss": 3.0155, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 7.11e-06, | |
| "loss": 2.9987, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 7.41e-06, | |
| "loss": 2.9953, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 7.71e-06, | |
| "loss": 3.0347, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 8.010000000000001e-06, | |
| "loss": 2.9781, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 8.310000000000002e-06, | |
| "loss": 2.9874, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 8.609999999999999e-06, | |
| "loss": 3.0158, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.91e-06, | |
| "loss": 2.9989, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 2.9806668758392334, | |
| "eval_runtime": 55.9321, | |
| "eval_samples_per_second": 30.036, | |
| "eval_steps_per_second": 30.036, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 9.21e-06, | |
| "loss": 2.961, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 9.51e-06, | |
| "loss": 2.966, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 9.810000000000001e-06, | |
| "loss": 2.9862, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.0110000000000001e-05, | |
| "loss": 2.9594, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 1.041e-05, | |
| "loss": 2.9571, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 1.0709999999999999e-05, | |
| "loss": 2.9593, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "learning_rate": 1.101e-05, | |
| "loss": 2.9717, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 1.131e-05, | |
| "loss": 2.9464, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 1.161e-05, | |
| "loss": 2.9459, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 1.1910000000000001e-05, | |
| "loss": 2.969, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 2.957939863204956, | |
| "eval_runtime": 55.8923, | |
| "eval_samples_per_second": 30.058, | |
| "eval_steps_per_second": 30.058, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 1.221e-05, | |
| "loss": 2.9376, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 1.251e-05, | |
| "loss": 2.9216, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.281e-05, | |
| "loss": 2.9236, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 1.311e-05, | |
| "loss": 2.9544, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 1.341e-05, | |
| "loss": 2.9085, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 1.3710000000000001e-05, | |
| "loss": 2.9052, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 1.4010000000000001e-05, | |
| "loss": 2.9162, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 1.431e-05, | |
| "loss": 2.9261, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 1.461e-05, | |
| "loss": 2.8997, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 1.491e-05, | |
| "loss": 2.903, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "eval_loss": 2.907166004180908, | |
| "eval_runtime": 56.2579, | |
| "eval_samples_per_second": 29.862, | |
| "eval_steps_per_second": 29.862, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "learning_rate": 1.521e-05, | |
| "loss": 2.9234, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 1.5510000000000002e-05, | |
| "loss": 2.8869, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 1.5810000000000003e-05, | |
| "loss": 2.8841, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 1.611e-05, | |
| "loss": 2.9118, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 1.641e-05, | |
| "loss": 2.9236, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 1.671e-05, | |
| "loss": 2.8824, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.93, | |
| "learning_rate": 1.7009999999999998e-05, | |
| "loss": 2.8745, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.731e-05, | |
| "loss": 2.9052, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1.761e-05, | |
| "loss": 2.8945, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 1.791e-05, | |
| "loss": 2.8565, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "eval_loss": 2.8804337978363037, | |
| "eval_runtime": 56.0547, | |
| "eval_samples_per_second": 29.971, | |
| "eval_steps_per_second": 29.971, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.821e-05, | |
| "loss": 2.8649, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 1.851e-05, | |
| "loss": 2.8968, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.881e-05, | |
| "loss": 2.8642, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 1.911e-05, | |
| "loss": 2.8483, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 1.9410000000000002e-05, | |
| "loss": 2.8636, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 1.9710000000000003e-05, | |
| "loss": 2.8843, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 2.0010000000000003e-05, | |
| "loss": 2.8345, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 4.69, | |
| "learning_rate": 2.031e-05, | |
| "loss": 2.8296, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.76, | |
| "learning_rate": 2.061e-05, | |
| "loss": 2.8429, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 2.0909999999999998e-05, | |
| "loss": 2.8195, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "eval_loss": 2.7915687561035156, | |
| "eval_runtime": 56.0017, | |
| "eval_samples_per_second": 29.999, | |
| "eval_steps_per_second": 29.999, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 2.121e-05, | |
| "loss": 2.7545, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 2.151e-05, | |
| "loss": 2.7348, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 2.181e-05, | |
| "loss": 2.7843, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "learning_rate": 2.211e-05, | |
| "loss": 2.6651, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 2.241e-05, | |
| "loss": 2.5925, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.24, | |
| "learning_rate": 2.271e-05, | |
| "loss": 2.5786, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 5.31, | |
| "learning_rate": 2.301e-05, | |
| "loss": 2.533, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 2.3310000000000002e-05, | |
| "loss": 2.3808, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 2.3610000000000003e-05, | |
| "loss": 2.3189, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 2.3910000000000003e-05, | |
| "loss": 2.3134, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_loss": 2.1455929279327393, | |
| "eval_runtime": 56.2645, | |
| "eval_samples_per_second": 29.859, | |
| "eval_steps_per_second": 29.859, | |
| "eval_wer": 1.0004134794293984, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.59, | |
| "learning_rate": 2.4210000000000004e-05, | |
| "loss": 2.1507, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 2.4509999999999997e-05, | |
| "loss": 2.0229, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 5.72, | |
| "learning_rate": 2.4809999999999998e-05, | |
| "loss": 2.0275, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 2.511e-05, | |
| "loss": 2.039, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 5.86, | |
| "learning_rate": 2.541e-05, | |
| "loss": 1.7749, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 2.571e-05, | |
| "loss": 1.7627, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 2.601e-05, | |
| "loss": 1.8399, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 2.631e-05, | |
| "loss": 1.7037, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 2.661e-05, | |
| "loss": 1.5494, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "learning_rate": 2.691e-05, | |
| "loss": 1.5475, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.21, | |
| "eval_loss": 1.46632719039917, | |
| "eval_runtime": 56.3055, | |
| "eval_samples_per_second": 29.837, | |
| "eval_steps_per_second": 29.837, | |
| "eval_wer": 0.9549307421955757, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.28, | |
| "learning_rate": 2.7210000000000002e-05, | |
| "loss": 1.7119, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 2.7510000000000003e-05, | |
| "loss": 1.4623, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 2.7810000000000003e-05, | |
| "loss": 1.4332, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "learning_rate": 2.8110000000000004e-05, | |
| "loss": 1.4821, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 6.55, | |
| "learning_rate": 2.841e-05, | |
| "loss": 1.4974, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.62, | |
| "learning_rate": 2.871e-05, | |
| "loss": 1.2691, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 6.69, | |
| "learning_rate": 2.901e-05, | |
| "loss": 1.2886, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 6.76, | |
| "learning_rate": 2.931e-05, | |
| "loss": 1.4779, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 2.961e-05, | |
| "loss": 1.2628, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "learning_rate": 2.991e-05, | |
| "loss": 1.1295, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.9, | |
| "eval_loss": 1.1139757633209229, | |
| "eval_runtime": 55.8538, | |
| "eval_samples_per_second": 30.079, | |
| "eval_steps_per_second": 30.079, | |
| "eval_wer": 0.7226931293501482, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "learning_rate": 2.988947368421053e-05, | |
| "loss": 1.219, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 2.9731578947368423e-05, | |
| "loss": 1.3547, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "learning_rate": 2.9573684210526316e-05, | |
| "loss": 1.0705, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "learning_rate": 2.9415789473684213e-05, | |
| "loss": 1.045, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 7.24, | |
| "learning_rate": 2.9257894736842106e-05, | |
| "loss": 1.1838, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.31, | |
| "learning_rate": 2.91e-05, | |
| "loss": 1.1164, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 2.8942105263157896e-05, | |
| "loss": 0.9137, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 7.45, | |
| "learning_rate": 2.8784210526315792e-05, | |
| "loss": 1.0233, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "learning_rate": 2.8626315789473686e-05, | |
| "loss": 1.2129, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 2.846842105263158e-05, | |
| "loss": 1.0181, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "eval_loss": 0.9257634282112122, | |
| "eval_runtime": 55.9356, | |
| "eval_samples_per_second": 30.035, | |
| "eval_steps_per_second": 30.035, | |
| "eval_wer": 0.6497140100613328, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.66, | |
| "learning_rate": 2.8310526315789475e-05, | |
| "loss": 0.9005, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 2.815263157894737e-05, | |
| "loss": 0.9942, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 2.7994736842105262e-05, | |
| "loss": 1.1054, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 7.86, | |
| "learning_rate": 2.783684210526316e-05, | |
| "loss": 0.8651, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "learning_rate": 2.7678947368421055e-05, | |
| "loss": 0.9205, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.752105263157895e-05, | |
| "loss": 1.0901, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 2.7363157894736845e-05, | |
| "loss": 0.9183, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 2.7205263157894738e-05, | |
| "loss": 0.7821, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "learning_rate": 2.704736842105263e-05, | |
| "loss": 0.8509, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "learning_rate": 2.6889473684210525e-05, | |
| "loss": 1.0252, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.28, | |
| "eval_loss": 0.8430129885673523, | |
| "eval_runtime": 55.9457, | |
| "eval_samples_per_second": 30.029, | |
| "eval_steps_per_second": 30.029, | |
| "eval_wer": 0.625456550203294, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.34, | |
| "learning_rate": 2.673157894736842e-05, | |
| "loss": 0.77, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 8.41, | |
| "learning_rate": 2.6573684210526318e-05, | |
| "loss": 0.7903, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 2.641578947368421e-05, | |
| "loss": 0.9209, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 8.55, | |
| "learning_rate": 2.6257894736842108e-05, | |
| "loss": 0.9417, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 8.62, | |
| "learning_rate": 2.61e-05, | |
| "loss": 0.7393, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.69, | |
| "learning_rate": 2.5942105263157894e-05, | |
| "loss": 0.7823, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 8.76, | |
| "learning_rate": 2.578421052631579e-05, | |
| "loss": 0.988, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "learning_rate": 2.5626315789473684e-05, | |
| "loss": 0.7872, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "learning_rate": 2.546842105263158e-05, | |
| "loss": 0.7158, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "learning_rate": 2.5310526315789474e-05, | |
| "loss": 0.835, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 0.8063345551490784, | |
| "eval_runtime": 56.07, | |
| "eval_samples_per_second": 29.963, | |
| "eval_steps_per_second": 29.963, | |
| "eval_wer": 0.6031975742540142, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.515263157894737e-05, | |
| "loss": 0.9458, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 2.4994736842105264e-05, | |
| "loss": 0.6545, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "learning_rate": 2.4836842105263157e-05, | |
| "loss": 0.6883, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 9.24, | |
| "learning_rate": 2.4678947368421053e-05, | |
| "loss": 0.8223, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 2.452105263157895e-05, | |
| "loss": 0.8211, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.38, | |
| "learning_rate": 2.4363157894736843e-05, | |
| "loss": 0.6561, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 9.45, | |
| "learning_rate": 2.420526315789474e-05, | |
| "loss": 0.7167, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 2.4047368421052633e-05, | |
| "loss": 0.8687, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 2.3889473684210526e-05, | |
| "loss": 0.6825, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "learning_rate": 2.373157894736842e-05, | |
| "loss": 0.662, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.66, | |
| "eval_loss": 0.7595294713973999, | |
| "eval_runtime": 56.1578, | |
| "eval_samples_per_second": 29.916, | |
| "eval_steps_per_second": 29.916, | |
| "eval_wer": 0.5931362414719867, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.72, | |
| "learning_rate": 2.3573684210526316e-05, | |
| "loss": 0.7777, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 2.3415789473684213e-05, | |
| "loss": 0.8767, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 9.86, | |
| "learning_rate": 2.3257894736842106e-05, | |
| "loss": 0.6377, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 9.93, | |
| "learning_rate": 2.3100000000000002e-05, | |
| "loss": 0.6553, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.2942105263157896e-05, | |
| "loss": 0.8173, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 2.278421052631579e-05, | |
| "loss": 0.6973, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 10.14, | |
| "learning_rate": 2.2626315789473685e-05, | |
| "loss": 0.5723, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 10.21, | |
| "learning_rate": 2.246842105263158e-05, | |
| "loss": 0.6482, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 10.28, | |
| "learning_rate": 2.2310526315789475e-05, | |
| "loss": 0.852, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "learning_rate": 2.215263157894737e-05, | |
| "loss": 0.5558, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.34, | |
| "eval_loss": 0.7321563959121704, | |
| "eval_runtime": 56.2309, | |
| "eval_samples_per_second": 29.877, | |
| "eval_steps_per_second": 29.877, | |
| "eval_wer": 0.5819033836399973, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.41, | |
| "learning_rate": 2.1994736842105265e-05, | |
| "loss": 0.5892, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "learning_rate": 2.183684210526316e-05, | |
| "loss": 0.7143, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 10.55, | |
| "learning_rate": 2.167894736842105e-05, | |
| "loss": 0.7776, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 2.1521052631578948e-05, | |
| "loss": 0.5764, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 10.69, | |
| "learning_rate": 2.136315789473684e-05, | |
| "loss": 0.6483, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "learning_rate": 2.1205263157894738e-05, | |
| "loss": 0.7903, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 10.83, | |
| "learning_rate": 2.1047368421052635e-05, | |
| "loss": 0.6301, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 10.9, | |
| "learning_rate": 2.0889473684210528e-05, | |
| "loss": 0.5636, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "learning_rate": 2.073157894736842e-05, | |
| "loss": 0.6343, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 2.0573684210526314e-05, | |
| "loss": 0.7596, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "eval_loss": 0.7120490670204163, | |
| "eval_runtime": 56.1181, | |
| "eval_samples_per_second": 29.937, | |
| "eval_steps_per_second": 29.937, | |
| "eval_wer": 0.5708083522844738, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 11.1, | |
| "learning_rate": 2.041578947368421e-05, | |
| "loss": 0.5381, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 11.17, | |
| "learning_rate": 2.0257894736842104e-05, | |
| "loss": 0.5223, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 11.24, | |
| "learning_rate": 2.01e-05, | |
| "loss": 0.686, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 11.31, | |
| "learning_rate": 1.9942105263157897e-05, | |
| "loss": 0.6802, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 11.38, | |
| "learning_rate": 1.978421052631579e-05, | |
| "loss": 0.5173, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.45, | |
| "learning_rate": 1.9626315789473684e-05, | |
| "loss": 0.5713, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 11.52, | |
| "learning_rate": 1.946842105263158e-05, | |
| "loss": 0.753, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 11.59, | |
| "learning_rate": 1.9310526315789474e-05, | |
| "loss": 0.5405, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 11.66, | |
| "learning_rate": 1.915263157894737e-05, | |
| "loss": 0.5048, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "learning_rate": 1.8994736842105263e-05, | |
| "loss": 0.6169, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.72, | |
| "eval_loss": 0.7072769999504089, | |
| "eval_runtime": 55.9632, | |
| "eval_samples_per_second": 30.02, | |
| "eval_steps_per_second": 30.02, | |
| "eval_wer": 0.5606091930259803, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.79, | |
| "learning_rate": 1.883684210526316e-05, | |
| "loss": 0.7515, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 11.86, | |
| "learning_rate": 1.8678947368421053e-05, | |
| "loss": 0.4808, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 1.8521052631578946e-05, | |
| "loss": 0.5669, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.8363157894736843e-05, | |
| "loss": 0.726, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 12.07, | |
| "learning_rate": 1.8205263157894736e-05, | |
| "loss": 0.569, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 1.8047368421052633e-05, | |
| "loss": 0.4807, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 1.788947368421053e-05, | |
| "loss": 0.5481, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 1.7731578947368423e-05, | |
| "loss": 0.7132, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 12.34, | |
| "learning_rate": 1.7573684210526316e-05, | |
| "loss": 0.505, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "learning_rate": 1.741578947368421e-05, | |
| "loss": 0.4565, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.41, | |
| "eval_loss": 0.712365448474884, | |
| "eval_runtime": 56.2962, | |
| "eval_samples_per_second": 29.842, | |
| "eval_steps_per_second": 29.842, | |
| "eval_wer": 0.5586107091172214, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.48, | |
| "learning_rate": 1.7257894736842106e-05, | |
| "loss": 0.6345, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 12.55, | |
| "learning_rate": 1.71e-05, | |
| "loss": 0.6555, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 1.6942105263157896e-05, | |
| "loss": 0.4412, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 12.69, | |
| "learning_rate": 1.6784210526315792e-05, | |
| "loss": 0.5018, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 1.6626315789473685e-05, | |
| "loss": 0.6562, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.83, | |
| "learning_rate": 1.646842105263158e-05, | |
| "loss": 0.5214, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 1.6310526315789475e-05, | |
| "loss": 0.4588, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 12.97, | |
| "learning_rate": 1.615263157894737e-05, | |
| "loss": 0.5569, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 1.599473684210526e-05, | |
| "loss": 0.6319, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "learning_rate": 1.5836842105263158e-05, | |
| "loss": 0.4554, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.1, | |
| "eval_loss": 0.6879737973213196, | |
| "eval_runtime": 56.2465, | |
| "eval_samples_per_second": 29.869, | |
| "eval_steps_per_second": 29.869, | |
| "eval_wer": 0.5500654675763215, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.5678947368421055e-05, | |
| "loss": 0.4659, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 13.24, | |
| "learning_rate": 1.5521052631578948e-05, | |
| "loss": 0.5649, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "learning_rate": 1.536315789473684e-05, | |
| "loss": 0.5556, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 13.38, | |
| "learning_rate": 1.5205263157894736e-05, | |
| "loss": 0.4317, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 1.5047368421052631e-05, | |
| "loss": 0.4837, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.52, | |
| "learning_rate": 1.4889473684210526e-05, | |
| "loss": 0.685, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "learning_rate": 1.4731578947368421e-05, | |
| "loss": 0.4845, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 1.4573684210526317e-05, | |
| "loss": 0.4278, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 13.72, | |
| "learning_rate": 1.441578947368421e-05, | |
| "loss": 0.5824, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "learning_rate": 1.4257894736842106e-05, | |
| "loss": 0.6216, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.79, | |
| "eval_loss": 0.6783236265182495, | |
| "eval_runtime": 56.2386, | |
| "eval_samples_per_second": 29.873, | |
| "eval_steps_per_second": 29.873, | |
| "eval_wer": 0.5494452484322239, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 1.4099999999999999e-05, | |
| "loss": 0.4211, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 13.93, | |
| "learning_rate": 1.3942105263157895e-05, | |
| "loss": 0.4556, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.378421052631579e-05, | |
| "loss": 0.6032, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 1.3626315789473684e-05, | |
| "loss": 0.4931, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 14.14, | |
| "learning_rate": 1.346842105263158e-05, | |
| "loss": 0.3818, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 14.21, | |
| "learning_rate": 1.3310526315789473e-05, | |
| "loss": 0.4653, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 14.28, | |
| "learning_rate": 1.3152631578947368e-05, | |
| "loss": 0.6482, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 14.34, | |
| "learning_rate": 1.2994736842105265e-05, | |
| "loss": 0.4267, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 14.41, | |
| "learning_rate": 1.2836842105263158e-05, | |
| "loss": 0.4512, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "learning_rate": 1.2678947368421053e-05, | |
| "loss": 0.5393, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.48, | |
| "eval_loss": 0.7067343592643738, | |
| "eval_runtime": 55.8427, | |
| "eval_samples_per_second": 30.085, | |
| "eval_steps_per_second": 30.085, | |
| "eval_wer": 0.5498587278616223, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 1.2521052631578946e-05, | |
| "loss": 0.5632, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 14.62, | |
| "learning_rate": 1.2363157894736843e-05, | |
| "loss": 0.3933, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 14.69, | |
| "learning_rate": 1.2205263157894738e-05, | |
| "loss": 0.4898, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 14.76, | |
| "learning_rate": 1.2047368421052631e-05, | |
| "loss": 0.6235, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 14.83, | |
| "learning_rate": 1.1889473684210528e-05, | |
| "loss": 0.4552, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 14.9, | |
| "learning_rate": 1.173157894736842e-05, | |
| "loss": 0.3651, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "learning_rate": 1.1573684210526316e-05, | |
| "loss": 0.4551, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.141578947368421e-05, | |
| "loss": 0.5899, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 1.1257894736842106e-05, | |
| "loss": 0.3866, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "learning_rate": 1.11e-05, | |
| "loss": 0.4095, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.17, | |
| "eval_loss": 0.7013991475105286, | |
| "eval_runtime": 55.9588, | |
| "eval_samples_per_second": 30.022, | |
| "eval_steps_per_second": 30.022, | |
| "eval_wer": 0.5437943628971126, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 1.0942105263157894e-05, | |
| "loss": 0.5369, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 15.31, | |
| "learning_rate": 1.078421052631579e-05, | |
| "loss": 0.4926, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 1.0626315789473685e-05, | |
| "loss": 0.3958, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 1.0468421052631578e-05, | |
| "loss": 0.4025, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 15.52, | |
| "learning_rate": 1.0310526315789475e-05, | |
| "loss": 0.6068, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 1.0152631578947368e-05, | |
| "loss": 0.4053, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 9.994736842105263e-06, | |
| "loss": 0.3958, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 15.72, | |
| "learning_rate": 9.836842105263158e-06, | |
| "loss": 0.5262, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 15.79, | |
| "learning_rate": 9.678947368421053e-06, | |
| "loss": 0.5737, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 9.521052631578948e-06, | |
| "loss": 0.3551, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "eval_loss": 0.6999756097793579, | |
| "eval_runtime": 55.808, | |
| "eval_samples_per_second": 30.103, | |
| "eval_steps_per_second": 30.103, | |
| "eval_wer": 0.5426228378471505, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 15.93, | |
| "learning_rate": 9.363157894736841e-06, | |
| "loss": 0.4163, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 9.205263157894738e-06, | |
| "loss": 0.5751, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 16.07, | |
| "learning_rate": 9.047368421052633e-06, | |
| "loss": 0.4375, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 16.14, | |
| "learning_rate": 8.889473684210526e-06, | |
| "loss": 0.3596, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 16.21, | |
| "learning_rate": 8.731578947368422e-06, | |
| "loss": 0.4274, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 8.573684210526316e-06, | |
| "loss": 0.5825, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 8.41578947368421e-06, | |
| "loss": 0.3741, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 16.41, | |
| "learning_rate": 8.257894736842105e-06, | |
| "loss": 0.3959, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 16.48, | |
| "learning_rate": 8.1e-06, | |
| "loss": 0.4807, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "learning_rate": 7.942105263157895e-06, | |
| "loss": 0.5112, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.55, | |
| "eval_loss": 0.6866209506988525, | |
| "eval_runtime": 55.9115, | |
| "eval_samples_per_second": 30.047, | |
| "eval_steps_per_second": 30.047, | |
| "eval_wer": 0.5426228378471505, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 16.62, | |
| "learning_rate": 7.784210526315789e-06, | |
| "loss": 0.3423, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 16.69, | |
| "learning_rate": 7.626315789473685e-06, | |
| "loss": 0.4187, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 16.76, | |
| "learning_rate": 7.468421052631579e-06, | |
| "loss": 0.5704, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 7.310526315789473e-06, | |
| "loss": 0.4254, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "learning_rate": 7.152631578947369e-06, | |
| "loss": 0.3743, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 16.97, | |
| "learning_rate": 6.994736842105263e-06, | |
| "loss": 0.4354, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 6.836842105263158e-06, | |
| "loss": 0.5345, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 17.1, | |
| "learning_rate": 6.678947368421053e-06, | |
| "loss": 0.3484, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 17.17, | |
| "learning_rate": 6.521052631578947e-06, | |
| "loss": 0.3749, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "learning_rate": 6.363157894736843e-06, | |
| "loss": 0.5139, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.24, | |
| "eval_loss": 0.7133733034133911, | |
| "eval_runtime": 55.9844, | |
| "eval_samples_per_second": 30.008, | |
| "eval_steps_per_second": 30.008, | |
| "eval_wer": 0.5445524085176763, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 17.31, | |
| "learning_rate": 6.205263157894737e-06, | |
| "loss": 0.4901, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 17.38, | |
| "learning_rate": 6.047368421052632e-06, | |
| "loss": 0.3151, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 17.45, | |
| "learning_rate": 5.889473684210527e-06, | |
| "loss": 0.3978, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 17.52, | |
| "learning_rate": 5.731578947368421e-06, | |
| "loss": 0.5495, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 17.59, | |
| "learning_rate": 5.573684210526316e-06, | |
| "loss": 0.3808, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 17.66, | |
| "learning_rate": 5.4157894736842105e-06, | |
| "loss": 0.3644, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 5.257894736842105e-06, | |
| "loss": 0.4576, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 17.79, | |
| "learning_rate": 5.1e-06, | |
| "loss": 0.5217, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 17.86, | |
| "learning_rate": 4.942105263157894e-06, | |
| "loss": 0.3407, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "learning_rate": 4.78421052631579e-06, | |
| "loss": 0.3638, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 17.93, | |
| "eval_loss": 0.7130064368247986, | |
| "eval_runtime": 56.0896, | |
| "eval_samples_per_second": 29.952, | |
| "eval_steps_per_second": 29.952, | |
| "eval_wer": 0.5434497967059472, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 4.626315789473684e-06, | |
| "loss": 0.5318, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 4.468421052631579e-06, | |
| "loss": 0.4136, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 18.14, | |
| "learning_rate": 4.310526315789474e-06, | |
| "loss": 0.3302, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 18.21, | |
| "learning_rate": 4.152631578947368e-06, | |
| "loss": 0.3974, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 18.28, | |
| "learning_rate": 3.994736842105264e-06, | |
| "loss": 0.5619, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 18.34, | |
| "learning_rate": 3.836842105263158e-06, | |
| "loss": 0.355, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 3.678947368421053e-06, | |
| "loss": 0.3625, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 18.48, | |
| "learning_rate": 3.5210526315789473e-06, | |
| "loss": 0.4751, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 18.55, | |
| "learning_rate": 3.363157894736842e-06, | |
| "loss": 0.4726, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 3.2052631578947367e-06, | |
| "loss": 0.3327, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "eval_loss": 0.6980041265487671, | |
| "eval_runtime": 56.0391, | |
| "eval_samples_per_second": 29.979, | |
| "eval_steps_per_second": 29.979, | |
| "eval_wer": 0.5377299979326029, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 18.69, | |
| "learning_rate": 3.0473684210526316e-06, | |
| "loss": 0.3934, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 18.76, | |
| "learning_rate": 2.8894736842105265e-06, | |
| "loss": 0.5291, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 18.83, | |
| "learning_rate": 2.731578947368421e-06, | |
| "loss": 0.3689, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 18.9, | |
| "learning_rate": 2.573684210526316e-06, | |
| "loss": 0.3522, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "learning_rate": 2.4157894736842104e-06, | |
| "loss": 0.4337, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 19.03, | |
| "learning_rate": 2.2578947368421053e-06, | |
| "loss": 0.5182, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 19.1, | |
| "learning_rate": 2.1000000000000002e-06, | |
| "loss": 0.3249, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 19.17, | |
| "learning_rate": 1.9421052631578947e-06, | |
| "loss": 0.355, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 19.24, | |
| "learning_rate": 1.7842105263157896e-06, | |
| "loss": 0.4947, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 1.6263157894736843e-06, | |
| "loss": 0.4385, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "eval_loss": 0.7016746997833252, | |
| "eval_runtime": 56.3283, | |
| "eval_samples_per_second": 29.825, | |
| "eval_steps_per_second": 29.825, | |
| "eval_wer": 0.5390393494590311, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 19.38, | |
| "learning_rate": 1.4684210526315788e-06, | |
| "loss": 0.3328, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 19.45, | |
| "learning_rate": 1.3105263157894737e-06, | |
| "loss": 0.3865, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 19.52, | |
| "learning_rate": 1.1526315789473684e-06, | |
| "loss": 0.5166, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 9.947368421052633e-07, | |
| "loss": 0.3516, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 19.66, | |
| "learning_rate": 8.368421052631579e-07, | |
| "loss": 0.3287, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 19.72, | |
| "learning_rate": 6.789473684210527e-07, | |
| "loss": 0.4436, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 19.79, | |
| "learning_rate": 5.210526315789473e-07, | |
| "loss": 0.503, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 19.86, | |
| "learning_rate": 3.6315789473684213e-07, | |
| "loss": 0.3404, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 19.93, | |
| "learning_rate": 2.0526315789473685e-07, | |
| "loss": 0.352, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 4.736842105263158e-08, | |
| "loss": 0.4986, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.7013729810714722, | |
| "eval_runtime": 55.9153, | |
| "eval_samples_per_second": 30.045, | |
| "eval_steps_per_second": 30.045, | |
| "eval_wer": 0.5373854317414375, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 2900, | |
| "total_flos": 2.6571394113263534e+18, | |
| "train_loss": 1.51724585755118, | |
| "train_runtime": 4772.7285, | |
| "train_samples_per_second": 19.36, | |
| "train_steps_per_second": 0.608 | |
| } | |
| ], | |
| "max_steps": 2900, | |
| "num_train_epochs": 20, | |
| "total_flos": 2.6571394113263534e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |