| { | |
| "best_metric": 5.654235369046776, | |
| "best_model_checkpoint": "./checkpoint-2500", | |
| "epoch": 5.020060180541625, | |
| "global_step": 2500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.2000000000000006e-07, | |
| "loss": 2.983, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.200000000000001e-07, | |
| "loss": 2.4843, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.42e-06, | |
| "loss": 1.6336, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9200000000000003e-06, | |
| "loss": 1.1537, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 2.42e-06, | |
| "loss": 0.8284, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 2.92e-06, | |
| "loss": 0.2948, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.4200000000000007e-06, | |
| "loss": 0.1898, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3.920000000000001e-06, | |
| "loss": 0.1784, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.42e-06, | |
| "loss": 0.1658, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.92e-06, | |
| "loss": 0.1603, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 5.420000000000001e-06, | |
| "loss": 0.1459, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 5.92e-06, | |
| "loss": 0.1448, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 6.42e-06, | |
| "loss": 0.1413, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 6.92e-06, | |
| "loss": 0.1349, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 7.420000000000001e-06, | |
| "loss": 0.1372, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 7.92e-06, | |
| "loss": 0.1304, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 8.42e-06, | |
| "loss": 0.128, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 8.920000000000001e-06, | |
| "loss": 0.1228, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 9.42e-06, | |
| "loss": 0.1198, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 9.920000000000002e-06, | |
| "loss": 0.1176, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.10957513749599457, | |
| "eval_runtime": 2168.1929, | |
| "eval_samples_per_second": 1.78, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 10.749416177199066, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.953333333333333e-06, | |
| "loss": 0.0829, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 9.89777777777778e-06, | |
| "loss": 0.0721, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 9.842222222222223e-06, | |
| "loss": 0.0747, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 9.786666666666667e-06, | |
| "loss": 0.0781, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 9.731111111111113e-06, | |
| "loss": 0.0769, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 9.675555555555555e-06, | |
| "loss": 0.0755, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 9.620000000000001e-06, | |
| "loss": 0.0805, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 9.564444444444445e-06, | |
| "loss": 0.0733, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 9.508888888888889e-06, | |
| "loss": 0.0696, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 9.453333333333335e-06, | |
| "loss": 0.0682, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 9.397777777777779e-06, | |
| "loss": 0.0727, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.342222222222223e-06, | |
| "loss": 0.0724, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 9.286666666666667e-06, | |
| "loss": 0.0688, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 9.231111111111111e-06, | |
| "loss": 0.0672, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 9.175555555555557e-06, | |
| "loss": 0.0705, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 9.12e-06, | |
| "loss": 0.0714, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 9.064444444444447e-06, | |
| "loss": 0.0635, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 9.008888888888889e-06, | |
| "loss": 0.0639, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 8.953333333333335e-06, | |
| "loss": 0.0728, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 8.897777777777779e-06, | |
| "loss": 0.0605, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_loss": 0.08087032288312912, | |
| "eval_runtime": 2054.0453, | |
| "eval_samples_per_second": 1.879, | |
| "eval_steps_per_second": 0.059, | |
| "eval_wer": 7.904606892647371, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 8.842222222222223e-06, | |
| "loss": 0.0319, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 8.786666666666668e-06, | |
| "loss": 0.0327, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 8.73111111111111e-06, | |
| "loss": 0.0352, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.675555555555556e-06, | |
| "loss": 0.0363, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 8.62e-06, | |
| "loss": 0.0327, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.564444444444445e-06, | |
| "loss": 0.0329, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 8.50888888888889e-06, | |
| "loss": 0.0305, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 8.453333333333334e-06, | |
| "loss": 0.0294, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 8.397777777777778e-06, | |
| "loss": 0.0304, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 8.342222222222222e-06, | |
| "loss": 0.0334, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 8.286666666666668e-06, | |
| "loss": 0.0326, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 8.231111111111112e-06, | |
| "loss": 0.0312, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 8.175555555555556e-06, | |
| "loss": 0.0291, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 8.120000000000002e-06, | |
| "loss": 0.0303, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 8.064444444444444e-06, | |
| "loss": 0.0308, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 8.00888888888889e-06, | |
| "loss": 0.0282, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.86, | |
| "learning_rate": 7.953333333333334e-06, | |
| "loss": 0.0317, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 7.897777777777778e-06, | |
| "loss": 0.028, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 7.842222222222224e-06, | |
| "loss": 0.0331, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 7.786666666666666e-06, | |
| "loss": 0.0241, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "eval_loss": 0.07452740520238876, | |
| "eval_runtime": 2048.6454, | |
| "eval_samples_per_second": 1.884, | |
| "eval_steps_per_second": 0.059, | |
| "eval_wer": 6.864340810982945, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.731111111111112e-06, | |
| "loss": 0.0167, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 7.675555555555556e-06, | |
| "loss": 0.0163, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 7.620000000000001e-06, | |
| "loss": 0.0149, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 7.564444444444446e-06, | |
| "loss": 0.0155, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 7.50888888888889e-06, | |
| "loss": 0.0145, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 7.453333333333334e-06, | |
| "loss": 0.0157, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 7.3977777777777786e-06, | |
| "loss": 0.0161, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 7.342222222222223e-06, | |
| "loss": 0.0154, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 7.2866666666666675e-06, | |
| "loss": 0.0149, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 7.231111111111112e-06, | |
| "loss": 0.0157, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 7.1755555555555556e-06, | |
| "loss": 0.0147, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "learning_rate": 7.1200000000000004e-06, | |
| "loss": 0.0153, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 7.0644444444444445e-06, | |
| "loss": 0.0155, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 7.008888888888889e-06, | |
| "loss": 0.0147, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "learning_rate": 6.953333333333334e-06, | |
| "loss": 0.015, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 6.897777777777779e-06, | |
| "loss": 0.0151, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 6.842222222222222e-06, | |
| "loss": 0.015, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 6.786666666666667e-06, | |
| "loss": 0.015, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 6.731111111111111e-06, | |
| "loss": 0.0164, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 6.675555555555556e-06, | |
| "loss": 0.0124, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "eval_loss": 0.06861147284507751, | |
| "eval_runtime": 2053.7472, | |
| "eval_samples_per_second": 1.879, | |
| "eval_steps_per_second": 0.059, | |
| "eval_wer": 5.767461609227938, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.620000000000001e-06, | |
| "loss": 0.0087, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 4.12, | |
| "learning_rate": 6.564444444444446e-06, | |
| "loss": 0.0087, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "learning_rate": 6.508888888888889e-06, | |
| "loss": 0.0079, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 6.453333333333334e-06, | |
| "loss": 0.007, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 6.397777777777778e-06, | |
| "loss": 0.0066, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 6.342222222222223e-06, | |
| "loss": 0.008, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 6.286666666666668e-06, | |
| "loss": 0.0083, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 6.231111111111111e-06, | |
| "loss": 0.0075, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 6.175555555555556e-06, | |
| "loss": 0.0071, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "learning_rate": 6.120000000000001e-06, | |
| "loss": 0.0066, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 6.064444444444445e-06, | |
| "loss": 0.0065, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 6.00888888888889e-06, | |
| "loss": 0.0076, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 5.9533333333333345e-06, | |
| "loss": 0.0072, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 5.897777777777778e-06, | |
| "loss": 0.0061, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 5.8422222222222226e-06, | |
| "loss": 0.0074, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 5.7866666666666674e-06, | |
| "loss": 0.007, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 5.7311111111111115e-06, | |
| "loss": 0.0082, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 5.675555555555556e-06, | |
| "loss": 0.0073, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 5.620000000000001e-06, | |
| "loss": 0.0073, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 5.5644444444444444e-06, | |
| "loss": 0.0058, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "eval_loss": 0.06784354150295258, | |
| "eval_runtime": 2168.2344, | |
| "eval_samples_per_second": 1.78, | |
| "eval_steps_per_second": 0.056, | |
| "eval_wer": 5.654235369046776, | |
| "step": 2500 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 11, | |
| "total_flos": 1.633018882277376e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |