| { | |
| "best_metric": 32.421838974453564, | |
| "best_model_checkpoint": "whisper-tiny-id/checkpoint-1500", | |
| "epoch": 49.504950495049506, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.8e-07, | |
| "loss": 2.0546, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 1.8800000000000002e-06, | |
| "loss": 1.7433, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 2.88e-06, | |
| "loss": 1.4134, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 3.88e-06, | |
| "loss": 1.1316, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 4.880000000000001e-06, | |
| "loss": 0.9869, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 5.8800000000000005e-06, | |
| "loss": 0.8961, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 6.88e-06, | |
| "loss": 0.8559, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 7.88e-06, | |
| "loss": 0.8192, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.880000000000001e-06, | |
| "loss": 0.7152, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 9.88e-06, | |
| "loss": 0.6923, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "learning_rate": 1.0880000000000001e-05, | |
| "loss": 0.6661, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 1.188e-05, | |
| "loss": 0.6392, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "learning_rate": 1.2880000000000002e-05, | |
| "loss": 0.5459, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.3880000000000001e-05, | |
| "loss": 0.5288, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.71, | |
| "learning_rate": 1.4880000000000002e-05, | |
| "loss": 0.5133, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 1.588e-05, | |
| "loss": 0.4964, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 1.688e-05, | |
| "loss": 0.4037, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 1.788e-05, | |
| "loss": 0.3798, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "learning_rate": 1.8880000000000002e-05, | |
| "loss": 0.3928, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 1.9880000000000003e-05, | |
| "loss": 0.3823, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_loss": 0.5251474380493164, | |
| "eval_runtime": 248.9641, | |
| "eval_samples_per_second": 14.532, | |
| "eval_steps_per_second": 0.458, | |
| "eval_wer": 33.47320852162686, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 1.9902222222222222e-05, | |
| "loss": 0.2984, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 1.9791111111111113e-05, | |
| "loss": 0.2636, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 1.968e-05, | |
| "loss": 0.2798, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 1.956888888888889e-05, | |
| "loss": 0.2654, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 1.945777777777778e-05, | |
| "loss": 0.2041, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 6.44, | |
| "learning_rate": 1.934666666666667e-05, | |
| "loss": 0.181, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.68, | |
| "learning_rate": 1.9235555555555557e-05, | |
| "loss": 0.1862, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 6.93, | |
| "learning_rate": 1.9124444444444445e-05, | |
| "loss": 0.1821, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.18, | |
| "learning_rate": 1.9013333333333333e-05, | |
| "loss": 0.1318, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 1.8902222222222225e-05, | |
| "loss": 0.1151, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 1.8791111111111113e-05, | |
| "loss": 0.1222, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 7.92, | |
| "learning_rate": 1.8680000000000004e-05, | |
| "loss": 0.1242, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "learning_rate": 1.856888888888889e-05, | |
| "loss": 0.0858, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 1.845777777777778e-05, | |
| "loss": 0.0753, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.66, | |
| "learning_rate": 1.834666666666667e-05, | |
| "loss": 0.0759, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "learning_rate": 1.8235555555555556e-05, | |
| "loss": 0.0778, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.16, | |
| "learning_rate": 1.8124444444444448e-05, | |
| "loss": 0.0586, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 9.41, | |
| "learning_rate": 1.8013333333333333e-05, | |
| "loss": 0.0461, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 1.7902222222222224e-05, | |
| "loss": 0.0483, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "learning_rate": 1.7791111111111112e-05, | |
| "loss": 0.0495, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.9, | |
| "eval_loss": 0.5699537992477417, | |
| "eval_runtime": 259.8738, | |
| "eval_samples_per_second": 13.922, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 33.39020566263949, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.15, | |
| "learning_rate": 1.768e-05, | |
| "loss": 0.0384, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 1.756888888888889e-05, | |
| "loss": 0.0299, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 10.64, | |
| "learning_rate": 1.745777777777778e-05, | |
| "loss": 0.0307, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "learning_rate": 1.7346666666666668e-05, | |
| "loss": 0.0318, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.14, | |
| "learning_rate": 1.7235555555555556e-05, | |
| "loss": 0.0251, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 1.7124444444444447e-05, | |
| "loss": 0.0201, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 1.7013333333333335e-05, | |
| "loss": 0.021, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 11.88, | |
| "learning_rate": 1.6902222222222223e-05, | |
| "loss": 0.0209, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.13, | |
| "learning_rate": 1.6791111111111115e-05, | |
| "loss": 0.0183, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 12.38, | |
| "learning_rate": 1.668e-05, | |
| "loss": 0.0157, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 12.62, | |
| "learning_rate": 1.656888888888889e-05, | |
| "loss": 0.015, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 12.87, | |
| "learning_rate": 1.645777777777778e-05, | |
| "loss": 0.0146, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 1.6346666666666667e-05, | |
| "loss": 0.0134, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 13.37, | |
| "learning_rate": 1.623555555555556e-05, | |
| "loss": 0.0113, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 13.61, | |
| "learning_rate": 1.6124444444444443e-05, | |
| "loss": 0.0108, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 13.86, | |
| "learning_rate": 1.6013333333333335e-05, | |
| "loss": 0.0102, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.11, | |
| "learning_rate": 1.5902222222222223e-05, | |
| "loss": 0.0092, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 14.36, | |
| "learning_rate": 1.5791111111111114e-05, | |
| "loss": 0.008, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 1.5680000000000002e-05, | |
| "loss": 0.0076, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "learning_rate": 1.556888888888889e-05, | |
| "loss": 0.0077, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 14.85, | |
| "eval_loss": 0.6201720237731934, | |
| "eval_runtime": 249.1204, | |
| "eval_samples_per_second": 14.523, | |
| "eval_steps_per_second": 0.458, | |
| "eval_wer": 32.421838974453564, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.1, | |
| "learning_rate": 1.545777777777778e-05, | |
| "loss": 0.0066, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 15.35, | |
| "learning_rate": 1.5346666666666667e-05, | |
| "loss": 0.0057, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 1.5235555555555556e-05, | |
| "loss": 0.0057, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 15.84, | |
| "learning_rate": 1.5124444444444446e-05, | |
| "loss": 0.006, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.09, | |
| "learning_rate": 1.5013333333333336e-05, | |
| "loss": 0.0056, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 16.34, | |
| "learning_rate": 1.4902222222222222e-05, | |
| "loss": 0.0049, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 16.58, | |
| "learning_rate": 1.4791111111111112e-05, | |
| "loss": 0.0048, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 16.83, | |
| "learning_rate": 1.4680000000000002e-05, | |
| "loss": 0.0052, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.08, | |
| "learning_rate": 1.456888888888889e-05, | |
| "loss": 0.0046, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 17.33, | |
| "learning_rate": 1.445777777777778e-05, | |
| "loss": 0.0041, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "learning_rate": 1.434666666666667e-05, | |
| "loss": 0.004, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 17.82, | |
| "learning_rate": 1.4235555555555556e-05, | |
| "loss": 0.0044, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.07, | |
| "learning_rate": 1.4124444444444445e-05, | |
| "loss": 0.0043, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 18.32, | |
| "learning_rate": 1.4013333333333334e-05, | |
| "loss": 0.0039, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 18.56, | |
| "learning_rate": 1.3902222222222223e-05, | |
| "loss": 0.0035, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 18.81, | |
| "learning_rate": 1.3791111111111113e-05, | |
| "loss": 0.0035, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 19.06, | |
| "learning_rate": 1.3680000000000003e-05, | |
| "loss": 0.0034, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 19.31, | |
| "learning_rate": 1.356888888888889e-05, | |
| "loss": 0.003, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 19.55, | |
| "learning_rate": 1.3457777777777779e-05, | |
| "loss": 0.0031, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 1.3346666666666667e-05, | |
| "loss": 0.0031, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "eval_loss": 0.6616396903991699, | |
| "eval_runtime": 260.1585, | |
| "eval_samples_per_second": 13.907, | |
| "eval_steps_per_second": 0.438, | |
| "eval_wer": 32.537120723047124, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 20.05, | |
| "learning_rate": 1.3235555555555557e-05, | |
| "loss": 0.0031, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 20.3, | |
| "learning_rate": 1.3124444444444447e-05, | |
| "loss": 0.0028, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 20.54, | |
| "learning_rate": 1.3013333333333333e-05, | |
| "loss": 0.0027, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 20.79, | |
| "learning_rate": 1.2902222222222223e-05, | |
| "loss": 0.0028, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 21.04, | |
| "learning_rate": 1.2791111111111112e-05, | |
| "loss": 0.0028, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 21.29, | |
| "learning_rate": 1.268e-05, | |
| "loss": 0.0026, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 21.53, | |
| "learning_rate": 1.256888888888889e-05, | |
| "loss": 0.0024, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 21.78, | |
| "learning_rate": 1.245777777777778e-05, | |
| "loss": 0.0024, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 22.03, | |
| "learning_rate": 1.2346666666666666e-05, | |
| "loss": 0.0025, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 22.28, | |
| "learning_rate": 1.2235555555555556e-05, | |
| "loss": 0.0024, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 22.52, | |
| "learning_rate": 1.2124444444444446e-05, | |
| "loss": 0.0023, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "learning_rate": 1.2013333333333334e-05, | |
| "loss": 0.0022, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 1.1902222222222224e-05, | |
| "loss": 0.0023, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 23.27, | |
| "learning_rate": 1.1791111111111114e-05, | |
| "loss": 0.0021, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 23.51, | |
| "learning_rate": 1.168e-05, | |
| "loss": 0.002, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 23.76, | |
| "learning_rate": 1.156888888888889e-05, | |
| "loss": 0.0021, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 24.01, | |
| "learning_rate": 1.145777777777778e-05, | |
| "loss": 0.0021, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 24.26, | |
| "learning_rate": 1.1346666666666668e-05, | |
| "loss": 0.0019, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 24.5, | |
| "learning_rate": 1.1235555555555557e-05, | |
| "loss": 0.0019, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "learning_rate": 1.1124444444444444e-05, | |
| "loss": 0.0019, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 24.75, | |
| "eval_loss": 0.687311589717865, | |
| "eval_runtime": 253.1426, | |
| "eval_samples_per_second": 14.292, | |
| "eval_steps_per_second": 0.45, | |
| "eval_wer": 32.79535183989671, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 1.1013333333333333e-05, | |
| "loss": 0.002, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 25.25, | |
| "learning_rate": 1.0902222222222223e-05, | |
| "loss": 0.0017, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 25.5, | |
| "learning_rate": 1.0791111111111113e-05, | |
| "loss": 0.0019, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 25.74, | |
| "learning_rate": 1.0680000000000001e-05, | |
| "loss": 0.0018, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.99, | |
| "learning_rate": 1.0568888888888891e-05, | |
| "loss": 0.0018, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 26.24, | |
| "learning_rate": 1.0457777777777777e-05, | |
| "loss": 0.0016, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 26.49, | |
| "learning_rate": 1.0346666666666667e-05, | |
| "loss": 0.0017, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 26.73, | |
| "learning_rate": 1.0235555555555557e-05, | |
| "loss": 0.0016, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 26.98, | |
| "learning_rate": 1.0124444444444445e-05, | |
| "loss": 0.0017, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 27.23, | |
| "learning_rate": 1.0013333333333335e-05, | |
| "loss": 0.0016, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 27.48, | |
| "learning_rate": 9.902222222222223e-06, | |
| "loss": 0.0016, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 27.72, | |
| "learning_rate": 9.791111111111112e-06, | |
| "loss": 0.0015, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 27.97, | |
| "learning_rate": 9.68e-06, | |
| "loss": 0.0015, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 28.22, | |
| "learning_rate": 9.56888888888889e-06, | |
| "loss": 0.0015, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 28.47, | |
| "learning_rate": 9.457777777777778e-06, | |
| "loss": 0.0014, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 28.71, | |
| "learning_rate": 9.346666666666666e-06, | |
| "loss": 0.0014, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "learning_rate": 9.235555555555556e-06, | |
| "loss": 0.0015, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 29.21, | |
| "learning_rate": 9.124444444444444e-06, | |
| "loss": 0.0014, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 29.46, | |
| "learning_rate": 9.013333333333334e-06, | |
| "loss": 0.0014, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 29.7, | |
| "learning_rate": 8.902222222222224e-06, | |
| "loss": 0.0014, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 29.7, | |
| "eval_loss": 0.7055913805961609, | |
| "eval_runtime": 259.2244, | |
| "eval_samples_per_second": 13.957, | |
| "eval_steps_per_second": 0.44, | |
| "eval_wer": 33.57004519044545, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 29.95, | |
| "learning_rate": 8.791111111111112e-06, | |
| "loss": 0.0014, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 30.2, | |
| "learning_rate": 8.68e-06, | |
| "loss": 0.0013, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 30.45, | |
| "learning_rate": 8.56888888888889e-06, | |
| "loss": 0.0013, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 30.69, | |
| "learning_rate": 8.457777777777778e-06, | |
| "loss": 0.0014, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 30.94, | |
| "learning_rate": 8.346666666666668e-06, | |
| "loss": 0.0013, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 31.19, | |
| "learning_rate": 8.235555555555557e-06, | |
| "loss": 0.0012, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 31.44, | |
| "learning_rate": 8.124444444444445e-06, | |
| "loss": 0.0012, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 31.68, | |
| "learning_rate": 8.013333333333333e-06, | |
| "loss": 0.0014, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 31.93, | |
| "learning_rate": 7.902222222222223e-06, | |
| "loss": 0.0012, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 32.18, | |
| "learning_rate": 7.791111111111111e-06, | |
| "loss": 0.0012, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 32.43, | |
| "learning_rate": 7.680000000000001e-06, | |
| "loss": 0.0012, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 32.67, | |
| "learning_rate": 7.56888888888889e-06, | |
| "loss": 0.0011, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 32.92, | |
| "learning_rate": 7.457777777777778e-06, | |
| "loss": 0.0011, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 33.17, | |
| "learning_rate": 7.346666666666668e-06, | |
| "loss": 0.0011, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 33.42, | |
| "learning_rate": 7.235555555555556e-06, | |
| "loss": 0.0011, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 33.66, | |
| "learning_rate": 7.124444444444445e-06, | |
| "loss": 0.0011, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 33.91, | |
| "learning_rate": 7.0133333333333345e-06, | |
| "loss": 0.0011, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 34.16, | |
| "learning_rate": 6.902222222222223e-06, | |
| "loss": 0.0011, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 34.41, | |
| "learning_rate": 6.7911111111111115e-06, | |
| "loss": 0.001, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 34.65, | |
| "learning_rate": 6.680000000000001e-06, | |
| "loss": 0.0011, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 34.65, | |
| "eval_loss": 0.7204159498214722, | |
| "eval_runtime": 258.6907, | |
| "eval_samples_per_second": 13.986, | |
| "eval_steps_per_second": 0.441, | |
| "eval_wer": 33.79599741768883, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 34.9, | |
| "learning_rate": 6.568888888888889e-06, | |
| "loss": 0.0011, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 35.15, | |
| "learning_rate": 6.457777777777778e-06, | |
| "loss": 0.001, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 35.4, | |
| "learning_rate": 6.346666666666668e-06, | |
| "loss": 0.001, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 35.64, | |
| "learning_rate": 6.235555555555556e-06, | |
| "loss": 0.001, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 35.89, | |
| "learning_rate": 6.124444444444445e-06, | |
| "loss": 0.001, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 36.14, | |
| "learning_rate": 6.013333333333335e-06, | |
| "loss": 0.001, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 36.39, | |
| "learning_rate": 5.902222222222223e-06, | |
| "loss": 0.001, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 36.63, | |
| "learning_rate": 5.791111111111112e-06, | |
| "loss": 0.001, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 36.88, | |
| "learning_rate": 5.68e-06, | |
| "loss": 0.001, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 37.13, | |
| "learning_rate": 5.56888888888889e-06, | |
| "loss": 0.0009, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 37.38, | |
| "learning_rate": 5.4577777777777785e-06, | |
| "loss": 0.0009, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 37.62, | |
| "learning_rate": 5.346666666666667e-06, | |
| "loss": 0.0009, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 37.87, | |
| "learning_rate": 5.235555555555556e-06, | |
| "loss": 0.0009, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 38.12, | |
| "learning_rate": 5.124444444444445e-06, | |
| "loss": 0.0009, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 38.37, | |
| "learning_rate": 5.013333333333333e-06, | |
| "loss": 0.0009, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 38.61, | |
| "learning_rate": 4.902222222222222e-06, | |
| "loss": 0.0009, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 38.86, | |
| "learning_rate": 4.791111111111111e-06, | |
| "loss": 0.0009, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 39.11, | |
| "learning_rate": 4.680000000000001e-06, | |
| "loss": 0.0009, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 39.36, | |
| "learning_rate": 4.568888888888889e-06, | |
| "loss": 0.0009, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 39.6, | |
| "learning_rate": 4.457777777777778e-06, | |
| "loss": 0.0009, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 39.6, | |
| "eval_loss": 0.7326923608779907, | |
| "eval_runtime": 259.5488, | |
| "eval_samples_per_second": 13.94, | |
| "eval_steps_per_second": 0.439, | |
| "eval_wer": 33.772941067970116, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 39.85, | |
| "learning_rate": 4.346666666666667e-06, | |
| "loss": 0.0008, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 40.1, | |
| "learning_rate": 4.235555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 40.35, | |
| "learning_rate": 4.124444444444445e-06, | |
| "loss": 0.0008, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 40.59, | |
| "learning_rate": 4.013333333333334e-06, | |
| "loss": 0.0008, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 40.84, | |
| "learning_rate": 3.9022222222222225e-06, | |
| "loss": 0.0008, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 41.09, | |
| "learning_rate": 3.7911111111111114e-06, | |
| "loss": 0.0008, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 41.34, | |
| "learning_rate": 3.6800000000000003e-06, | |
| "loss": 0.0008, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 41.58, | |
| "learning_rate": 3.568888888888889e-06, | |
| "loss": 0.0008, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 41.83, | |
| "learning_rate": 3.457777777777778e-06, | |
| "loss": 0.0008, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 42.08, | |
| "learning_rate": 3.346666666666667e-06, | |
| "loss": 0.0008, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 42.33, | |
| "learning_rate": 3.2355555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 42.57, | |
| "learning_rate": 3.124444444444445e-06, | |
| "loss": 0.0008, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 42.82, | |
| "learning_rate": 3.013333333333334e-06, | |
| "loss": 0.0008, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 43.07, | |
| "learning_rate": 2.9022222222222223e-06, | |
| "loss": 0.0008, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 43.32, | |
| "learning_rate": 2.7911111111111113e-06, | |
| "loss": 0.0008, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 43.56, | |
| "learning_rate": 2.68e-06, | |
| "loss": 0.0008, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 43.81, | |
| "learning_rate": 2.568888888888889e-06, | |
| "loss": 0.0008, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 44.06, | |
| "learning_rate": 2.457777777777778e-06, | |
| "loss": 0.0008, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 44.31, | |
| "learning_rate": 2.346666666666667e-06, | |
| "loss": 0.0008, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 44.55, | |
| "learning_rate": 2.235555555555556e-06, | |
| "loss": 0.0008, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 44.55, | |
| "eval_loss": 0.7400460839271545, | |
| "eval_runtime": 254.3271, | |
| "eval_samples_per_second": 14.226, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 33.91127916628239, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 44.8, | |
| "learning_rate": 2.1244444444444443e-06, | |
| "loss": 0.0008, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 45.05, | |
| "learning_rate": 2.0133333333333337e-06, | |
| "loss": 0.0008, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 45.3, | |
| "learning_rate": 1.9022222222222222e-06, | |
| "loss": 0.0007, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 45.54, | |
| "learning_rate": 1.7911111111111113e-06, | |
| "loss": 0.0007, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 45.79, | |
| "learning_rate": 1.6800000000000002e-06, | |
| "loss": 0.0007, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 46.04, | |
| "learning_rate": 1.568888888888889e-06, | |
| "loss": 0.0008, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 46.29, | |
| "learning_rate": 1.457777777777778e-06, | |
| "loss": 0.0007, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 46.53, | |
| "learning_rate": 1.3466666666666668e-06, | |
| "loss": 0.0007, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 46.78, | |
| "learning_rate": 1.2355555555555557e-06, | |
| "loss": 0.0007, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 47.03, | |
| "learning_rate": 1.1244444444444446e-06, | |
| "loss": 0.0007, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 47.28, | |
| "learning_rate": 1.0133333333333333e-06, | |
| "loss": 0.0007, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 47.52, | |
| "learning_rate": 9.022222222222222e-07, | |
| "loss": 0.0007, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 47.77, | |
| "learning_rate": 7.911111111111111e-07, | |
| "loss": 0.0007, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 48.02, | |
| "learning_rate": 6.800000000000001e-07, | |
| "loss": 0.0007, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 48.27, | |
| "learning_rate": 5.68888888888889e-07, | |
| "loss": 0.0007, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 48.51, | |
| "learning_rate": 4.5777777777777784e-07, | |
| "loss": 0.0007, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 48.76, | |
| "learning_rate": 3.466666666666667e-07, | |
| "loss": 0.0007, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 49.01, | |
| "learning_rate": 2.3555555555555556e-07, | |
| "loss": 0.0007, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 49.26, | |
| "learning_rate": 1.2444444444444446e-07, | |
| "loss": 0.0007, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "learning_rate": 1.3333333333333334e-08, | |
| "loss": 0.0007, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "eval_loss": 0.7428141832351685, | |
| "eval_runtime": 254.5036, | |
| "eval_samples_per_second": 14.216, | |
| "eval_steps_per_second": 0.448, | |
| "eval_wer": 33.344092963202066, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 49.5, | |
| "step": 5000, | |
| "total_flos": 7.86356733394944e+18, | |
| "train_loss": 0.09864863240011036, | |
| "train_runtime": 10305.0998, | |
| "train_samples_per_second": 31.053, | |
| "train_steps_per_second": 0.485 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 50, | |
| "total_flos": 7.86356733394944e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |