| { | |
| "best_metric": 11.972265023112481, | |
| "best_model_checkpoint": "whisper-small-pt-cv11-v6/checkpoint-4000", | |
| "epoch": 10.128913443830571, | |
| "global_step": 5500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.0500000000000001e-07, | |
| "loss": 2.8945, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 2.3000000000000002e-07, | |
| "loss": 2.8283, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 3.55e-07, | |
| "loss": 2.5065, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 4.800000000000001e-07, | |
| "loss": 2.1153, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 6.05e-07, | |
| "loss": 1.8013, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.3e-07, | |
| "loss": 1.5409, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 8.550000000000002e-07, | |
| "loss": 1.3225, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 9.800000000000001e-07, | |
| "loss": 1.0784, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.105e-06, | |
| "loss": 0.849, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.23e-06, | |
| "loss": 0.7216, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.3550000000000002e-06, | |
| "loss": 0.7346, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.48e-06, | |
| "loss": 0.6889, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 1.605e-06, | |
| "loss": 0.65, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 1.73e-06, | |
| "loss": 0.6041, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.8550000000000002e-06, | |
| "loss": 0.577, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.5672, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 2.105e-06, | |
| "loss": 0.5452, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.2300000000000002e-06, | |
| "loss": 0.6018, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 2.355e-06, | |
| "loss": 0.5373, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 2.4800000000000004e-06, | |
| "loss": 0.5235, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_cer": 6.12047526941144, | |
| "eval_loss": 0.36049944162368774, | |
| "eval_runtime": 195.7624, | |
| "eval_samples_per_second": 5.108, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 15.408320493066254, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 2.6050000000000005e-06, | |
| "loss": 0.4965, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2.7300000000000005e-06, | |
| "loss": 0.4609, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.855e-06, | |
| "loss": 0.4644, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.9800000000000003e-06, | |
| "loss": 0.4705, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 3.1050000000000003e-06, | |
| "loss": 0.4124, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3.2300000000000004e-06, | |
| "loss": 0.432, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.3550000000000005e-06, | |
| "loss": 0.4382, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.48e-06, | |
| "loss": 0.4476, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3.6050000000000002e-06, | |
| "loss": 0.4038, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 3.7300000000000003e-06, | |
| "loss": 0.4147, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 3.855e-06, | |
| "loss": 0.3953, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 0.4013, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 4.1050000000000005e-06, | |
| "loss": 0.4129, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 4.23e-06, | |
| "loss": 0.4124, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.355000000000001e-06, | |
| "loss": 0.3788, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.48e-06, | |
| "loss": 0.3987, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 4.605000000000001e-06, | |
| "loss": 0.3795, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 4.7300000000000005e-06, | |
| "loss": 0.379, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 4.855e-06, | |
| "loss": 0.3712, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 4.980000000000001e-06, | |
| "loss": 0.3839, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_cer": 5.60099474993092, | |
| "eval_loss": 0.30335938930511475, | |
| "eval_runtime": 195.7525, | |
| "eval_samples_per_second": 5.108, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 14.28351309707242, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 4.9883333333333336e-06, | |
| "loss": 0.3648, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 4.974444444444445e-06, | |
| "loss": 0.347, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.9605555555555564e-06, | |
| "loss": 0.373, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 4.946666666666667e-06, | |
| "loss": 0.3188, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 4.932777777777778e-06, | |
| "loss": 0.3181, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.918888888888889e-06, | |
| "loss": 0.3131, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 4.9050000000000005e-06, | |
| "loss": 0.298, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 4.891111111111111e-06, | |
| "loss": 0.2885, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "learning_rate": 4.8772222222222225e-06, | |
| "loss": 0.3091, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.863333333333334e-06, | |
| "loss": 0.3007, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 4.849444444444445e-06, | |
| "loss": 0.2927, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.835555555555556e-06, | |
| "loss": 0.3095, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.821666666666667e-06, | |
| "loss": 0.2895, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 4.807777777777779e-06, | |
| "loss": 0.2864, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 4.793888888888889e-06, | |
| "loss": 0.2875, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.78e-06, | |
| "loss": 0.2752, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 4.766111111111111e-06, | |
| "loss": 0.2973, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 4.752222222222223e-06, | |
| "loss": 0.2919, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 4.7383333333333334e-06, | |
| "loss": 0.3035, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 4.724444444444445e-06, | |
| "loss": 0.2828, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_cer": 5.172699640784747, | |
| "eval_loss": 0.28523197770118713, | |
| "eval_runtime": 195.9181, | |
| "eval_samples_per_second": 5.104, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 13.49768875192604, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 4.710555555555556e-06, | |
| "loss": 0.2898, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 4.696666666666667e-06, | |
| "loss": 0.2805, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 4.682777777777778e-06, | |
| "loss": 0.2673, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 4.66888888888889e-06, | |
| "loss": 0.274, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 4.655e-06, | |
| "loss": 0.2953, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.641111111111111e-06, | |
| "loss": 0.2478, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.627222222222222e-06, | |
| "loss": 0.2312, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "learning_rate": 4.613333333333334e-06, | |
| "loss": 0.2277, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "learning_rate": 4.599444444444444e-06, | |
| "loss": 0.232, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 4.585555555555556e-06, | |
| "loss": 0.2403, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 4.571666666666667e-06, | |
| "loss": 0.226, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 3.31, | |
| "learning_rate": 4.557777777777778e-06, | |
| "loss": 0.2107, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 4.543888888888889e-06, | |
| "loss": 0.2237, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.530000000000001e-06, | |
| "loss": 0.2459, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 4.516111111111111e-06, | |
| "loss": 0.2305, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 4.502222222222223e-06, | |
| "loss": 0.2265, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 4.488333333333333e-06, | |
| "loss": 0.2331, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 3.59, | |
| "learning_rate": 4.474444444444445e-06, | |
| "loss": 0.2387, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 4.460555555555556e-06, | |
| "loss": 0.2268, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 4.446666666666667e-06, | |
| "loss": 0.2367, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_cer": 5.227963525835866, | |
| "eval_loss": 0.2767547070980072, | |
| "eval_runtime": 194.9939, | |
| "eval_samples_per_second": 5.128, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 12.912172573189522, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 4.432777777777778e-06, | |
| "loss": 0.2241, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 4.41888888888889e-06, | |
| "loss": 0.2241, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 3.82, | |
| "learning_rate": 4.405e-06, | |
| "loss": 0.2183, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.391111111111112e-06, | |
| "loss": 0.2318, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "learning_rate": 4.377222222222223e-06, | |
| "loss": 0.2315, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 4.363333333333334e-06, | |
| "loss": 0.2166, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 4.349444444444445e-06, | |
| "loss": 0.2207, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 4.3355555555555565e-06, | |
| "loss": 0.1828, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "learning_rate": 4.321666666666667e-06, | |
| "loss": 0.1753, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 4.307777777777778e-06, | |
| "loss": 0.1906, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 4.293888888888889e-06, | |
| "loss": 0.1791, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 4.2800000000000005e-06, | |
| "loss": 0.173, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 4.266111111111111e-06, | |
| "loss": 0.184, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 4.33, | |
| "learning_rate": 4.2522222222222225e-06, | |
| "loss": 0.1735, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 4.37, | |
| "learning_rate": 4.238333333333334e-06, | |
| "loss": 0.1839, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 4.42, | |
| "learning_rate": 4.2244444444444446e-06, | |
| "loss": 0.1697, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 4.47, | |
| "learning_rate": 4.210555555555556e-06, | |
| "loss": 0.1966, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 4.1966666666666674e-06, | |
| "loss": 0.1836, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 4.182777777777778e-06, | |
| "loss": 0.1822, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 4.168888888888889e-06, | |
| "loss": 0.1832, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "eval_cer": 4.9157225752970435, | |
| "eval_loss": 0.27277910709381104, | |
| "eval_runtime": 194.976, | |
| "eval_samples_per_second": 5.129, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 12.249614791987673, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 4.155e-06, | |
| "loss": 0.1783, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 4.1411111111111115e-06, | |
| "loss": 0.1759, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 4.127222222222222e-06, | |
| "loss": 0.1739, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 4.79, | |
| "learning_rate": 4.1133333333333335e-06, | |
| "loss": 0.1805, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "learning_rate": 4.099444444444445e-06, | |
| "loss": 0.1886, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.085555555555556e-06, | |
| "loss": 0.1899, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 4.93, | |
| "learning_rate": 4.071666666666667e-06, | |
| "loss": 0.181, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 4.057777777777778e-06, | |
| "loss": 0.1858, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 4.04388888888889e-06, | |
| "loss": 0.1746, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.03e-06, | |
| "loss": 0.1467, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 5.11, | |
| "learning_rate": 4.016111111111111e-06, | |
| "loss": 0.1446, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.002222222222222e-06, | |
| "loss": 0.1512, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 3.988333333333334e-06, | |
| "loss": 0.1543, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 5.25, | |
| "learning_rate": 3.974444444444444e-06, | |
| "loss": 0.141, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 3.960555555555556e-06, | |
| "loss": 0.1395, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 5.34, | |
| "learning_rate": 3.946666666666667e-06, | |
| "loss": 0.1517, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "learning_rate": 3.932777777777778e-06, | |
| "loss": 0.1514, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "learning_rate": 3.918888888888889e-06, | |
| "loss": 0.1443, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "learning_rate": 3.905000000000001e-06, | |
| "loss": 0.15, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "learning_rate": 3.891111111111111e-06, | |
| "loss": 0.1549, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "eval_cer": 4.838353136225477, | |
| "eval_loss": 0.27296122908592224, | |
| "eval_runtime": 194.5355, | |
| "eval_samples_per_second": 5.14, | |
| "eval_steps_per_second": 0.324, | |
| "eval_wer": 12.064714946070879, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "learning_rate": 3.877222222222223e-06, | |
| "loss": 0.1592, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 3.863333333333333e-06, | |
| "loss": 0.1526, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 3.849444444444445e-06, | |
| "loss": 0.1578, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 3.835555555555555e-06, | |
| "loss": 0.1463, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 3.821666666666667e-06, | |
| "loss": 0.1554, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 3.8077777777777782e-06, | |
| "loss": 0.1614, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.7938888888888892e-06, | |
| "loss": 0.1535, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 5.89, | |
| "learning_rate": 3.7800000000000002e-06, | |
| "loss": 0.1522, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 3.7661111111111117e-06, | |
| "loss": 0.1657, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 3.7522222222222227e-06, | |
| "loss": 0.1514, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 3.7383333333333337e-06, | |
| "loss": 0.1317, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.724444444444445e-06, | |
| "loss": 0.1221, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 3.710555555555556e-06, | |
| "loss": 0.1407, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 3.6966666666666667e-06, | |
| "loss": 0.1224, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 3.6827777777777777e-06, | |
| "loss": 0.123, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "learning_rate": 3.668888888888889e-06, | |
| "loss": 0.1205, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 6.31, | |
| "learning_rate": 3.655e-06, | |
| "loss": 0.1292, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 3.641111111111111e-06, | |
| "loss": 0.1312, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 3.6272222222222226e-06, | |
| "loss": 0.1254, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 3.6133333333333336e-06, | |
| "loss": 0.1318, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "eval_cer": 4.813484387952473, | |
| "eval_loss": 0.2756599187850952, | |
| "eval_runtime": 197.5294, | |
| "eval_samples_per_second": 5.063, | |
| "eval_steps_per_second": 0.319, | |
| "eval_wer": 12.09553158705701, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 3.599444444444445e-06, | |
| "loss": 0.1312, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 6.54, | |
| "learning_rate": 3.585555555555556e-06, | |
| "loss": 0.135, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 6.58, | |
| "learning_rate": 3.571666666666667e-06, | |
| "loss": 0.118, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 6.63, | |
| "learning_rate": 3.5577777777777785e-06, | |
| "loss": 0.128, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 3.543888888888889e-06, | |
| "loss": 0.1296, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 3.53e-06, | |
| "loss": 0.1352, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 3.516111111111111e-06, | |
| "loss": 0.1311, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 6.81, | |
| "learning_rate": 3.5022222222222225e-06, | |
| "loss": 0.1251, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 6.86, | |
| "learning_rate": 3.4883333333333335e-06, | |
| "loss": 0.1213, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "learning_rate": 3.4744444444444445e-06, | |
| "loss": 0.124, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 6.95, | |
| "learning_rate": 3.460555555555556e-06, | |
| "loss": 0.1357, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.446666666666667e-06, | |
| "loss": 0.1282, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 3.4327777777777784e-06, | |
| "loss": 0.1048, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 3.4188888888888894e-06, | |
| "loss": 0.1007, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 7.14, | |
| "learning_rate": 3.4050000000000004e-06, | |
| "loss": 0.1021, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 3.391111111111111e-06, | |
| "loss": 0.1029, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 3.3772222222222225e-06, | |
| "loss": 0.1108, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3.3633333333333335e-06, | |
| "loss": 0.113, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 3.3494444444444445e-06, | |
| "loss": 0.1007, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 3.335555555555556e-06, | |
| "loss": 0.1077, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "eval_cer": 4.827300359215253, | |
| "eval_loss": 0.27384495735168457, | |
| "eval_runtime": 195.4102, | |
| "eval_samples_per_second": 5.117, | |
| "eval_steps_per_second": 0.322, | |
| "eval_wer": 11.972265023112481, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 3.321666666666667e-06, | |
| "loss": 0.1084, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 3.307777777777778e-06, | |
| "loss": 0.1014, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 3.2938888888888894e-06, | |
| "loss": 0.1175, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 7.55, | |
| "learning_rate": 3.2800000000000004e-06, | |
| "loss": 0.1098, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 3.2661111111111114e-06, | |
| "loss": 0.1114, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 7.64, | |
| "learning_rate": 3.252222222222223e-06, | |
| "loss": 0.1039, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 7.69, | |
| "learning_rate": 3.2383333333333334e-06, | |
| "loss": 0.1066, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 7.73, | |
| "learning_rate": 3.2244444444444444e-06, | |
| "loss": 0.1149, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 3.210555555555556e-06, | |
| "loss": 0.0999, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 3.196666666666667e-06, | |
| "loss": 0.1106, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "learning_rate": 3.182777777777778e-06, | |
| "loss": 0.1153, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 3.1688888888888893e-06, | |
| "loss": 0.1044, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 3.1550000000000003e-06, | |
| "loss": 0.1105, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 3.1411111111111113e-06, | |
| "loss": 0.1094, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 3.1272222222222227e-06, | |
| "loss": 0.0883, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 3.1133333333333337e-06, | |
| "loss": 0.0944, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 8.15, | |
| "learning_rate": 3.0994444444444447e-06, | |
| "loss": 0.0922, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 3.085555555555556e-06, | |
| "loss": 0.0917, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 8.24, | |
| "learning_rate": 3.0716666666666668e-06, | |
| "loss": 0.0848, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.0577777777777778e-06, | |
| "loss": 0.0969, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "eval_cer": 4.921248963802155, | |
| "eval_loss": 0.27840641140937805, | |
| "eval_runtime": 194.3399, | |
| "eval_samples_per_second": 5.146, | |
| "eval_steps_per_second": 0.324, | |
| "eval_wer": 12.157164869029275, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 8.33, | |
| "learning_rate": 3.043888888888889e-06, | |
| "loss": 0.0907, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 8.38, | |
| "learning_rate": 3.0300000000000002e-06, | |
| "loss": 0.093, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "learning_rate": 3.0161111111111112e-06, | |
| "loss": 0.1022, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 3.0022222222222227e-06, | |
| "loss": 0.0985, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "learning_rate": 2.9883333333333337e-06, | |
| "loss": 0.0954, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 2.9744444444444447e-06, | |
| "loss": 0.0849, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "learning_rate": 2.960555555555556e-06, | |
| "loss": 0.0931, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 2.946666666666667e-06, | |
| "loss": 0.0965, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "learning_rate": 2.932777777777778e-06, | |
| "loss": 0.0922, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "learning_rate": 2.9188888888888887e-06, | |
| "loss": 0.0998, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 8.79, | |
| "learning_rate": 2.905e-06, | |
| "loss": 0.0906, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 8.84, | |
| "learning_rate": 2.891111111111111e-06, | |
| "loss": 0.0967, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 2.877222222222222e-06, | |
| "loss": 0.0972, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 8.93, | |
| "learning_rate": 2.8633333333333336e-06, | |
| "loss": 0.0936, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "learning_rate": 2.8494444444444446e-06, | |
| "loss": 0.0942, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 2.835555555555556e-06, | |
| "loss": 0.0882, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 2.821666666666667e-06, | |
| "loss": 0.0832, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 9.12, | |
| "learning_rate": 2.807777777777778e-06, | |
| "loss": 0.0792, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 2.7938888888888895e-06, | |
| "loss": 0.084, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "learning_rate": 2.7800000000000005e-06, | |
| "loss": 0.0813, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.21, | |
| "eval_cer": 5.02072395689417, | |
| "eval_loss": 0.28047436475753784, | |
| "eval_runtime": 194.5917, | |
| "eval_samples_per_second": 5.139, | |
| "eval_steps_per_second": 0.324, | |
| "eval_wer": 12.311248073959938, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 9.25, | |
| "learning_rate": 2.766111111111111e-06, | |
| "loss": 0.0767, | |
| "step": 5025 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.752222222222222e-06, | |
| "loss": 0.0812, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "learning_rate": 2.7383333333333335e-06, | |
| "loss": 0.0757, | |
| "step": 5075 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "learning_rate": 2.7244444444444445e-06, | |
| "loss": 0.0829, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 9.44, | |
| "learning_rate": 2.7105555555555555e-06, | |
| "loss": 0.0779, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "learning_rate": 2.696666666666667e-06, | |
| "loss": 0.0858, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 9.53, | |
| "learning_rate": 2.682777777777778e-06, | |
| "loss": 0.0809, | |
| "step": 5175 | |
| }, | |
| { | |
| "epoch": 9.58, | |
| "learning_rate": 2.6688888888888894e-06, | |
| "loss": 0.0814, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 9.62, | |
| "learning_rate": 2.6550000000000004e-06, | |
| "loss": 0.0821, | |
| "step": 5225 | |
| }, | |
| { | |
| "epoch": 9.67, | |
| "learning_rate": 2.6411111111111114e-06, | |
| "loss": 0.0787, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 9.71, | |
| "learning_rate": 2.627222222222223e-06, | |
| "loss": 0.0797, | |
| "step": 5275 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.6133333333333334e-06, | |
| "loss": 0.0785, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 9.81, | |
| "learning_rate": 2.5994444444444444e-06, | |
| "loss": 0.083, | |
| "step": 5325 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "learning_rate": 2.5855555555555555e-06, | |
| "loss": 0.0818, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 2.571666666666667e-06, | |
| "loss": 0.0799, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "learning_rate": 2.557777777777778e-06, | |
| "loss": 0.0812, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 2.543888888888889e-06, | |
| "loss": 0.0834, | |
| "step": 5425 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 2.5300000000000003e-06, | |
| "loss": 0.075, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 10.08, | |
| "learning_rate": 2.5161111111111113e-06, | |
| "loss": 0.0682, | |
| "step": 5475 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 2.5022222222222224e-06, | |
| "loss": 0.0751, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "eval_cer": 4.8494059132357, | |
| "eval_loss": 0.28306886553764343, | |
| "eval_runtime": 195.1376, | |
| "eval_samples_per_second": 5.125, | |
| "eval_steps_per_second": 0.323, | |
| "eval_wer": 12.080123266563945, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "step": 5500, | |
| "total_flos": 5.079103045632e+19, | |
| "train_loss": 0.2771375383897261, | |
| "train_runtime": 18059.6987, | |
| "train_samples_per_second": 17.719, | |
| "train_steps_per_second": 0.554 | |
| } | |
| ], | |
| "max_steps": 10000, | |
| "num_train_epochs": 19, | |
| "total_flos": 5.079103045632e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |