| { | |
| "best_metric": 23.81094676029683, | |
| "best_model_checkpoint": "./checkpoint-5000", | |
| "epoch": 2.1788, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 4.6000000000000004e-07, | |
| "loss": 1.7155, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 9.600000000000001e-07, | |
| "loss": 1.2833, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.46e-06, | |
| "loss": 0.6549, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.9600000000000003e-06, | |
| "loss": 0.4167, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.46e-06, | |
| "loss": 0.3397, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 2.96e-06, | |
| "loss": 0.3208, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.46e-06, | |
| "loss": 0.2839, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 3.96e-06, | |
| "loss": 0.2953, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 4.4600000000000005e-06, | |
| "loss": 0.3841, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.960000000000001e-06, | |
| "loss": 0.416, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.460000000000001e-06, | |
| "loss": 0.3926, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.9600000000000005e-06, | |
| "loss": 0.3496, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.460000000000001e-06, | |
| "loss": 0.4149, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 6.96e-06, | |
| "loss": 0.482, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 7.4600000000000006e-06, | |
| "loss": 0.4613, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 7.960000000000002e-06, | |
| "loss": 0.4522, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.46e-06, | |
| "loss": 0.4463, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.96e-06, | |
| "loss": 0.3219, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.460000000000001e-06, | |
| "loss": 0.3347, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.960000000000001e-06, | |
| "loss": 0.2778, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_loss": 0.49452078342437744, | |
| "eval_runtime": 2170.0908, | |
| "eval_samples_per_second": 3.463, | |
| "eval_steps_per_second": 0.217, | |
| "eval_wer": 36.98391354151103, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.94888888888889e-06, | |
| "loss": 0.2984, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 9.893333333333334e-06, | |
| "loss": 0.3115, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.837777777777778e-06, | |
| "loss": 0.2502, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.782222222222222e-06, | |
| "loss": 0.2514, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 9.726666666666668e-06, | |
| "loss": 0.3575, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9.671111111111112e-06, | |
| "loss": 0.4231, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.615555555555558e-06, | |
| "loss": 0.4043, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.56e-06, | |
| "loss": 0.4514, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 9.504444444444446e-06, | |
| "loss": 0.4511, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.44888888888889e-06, | |
| "loss": 0.4803, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 9.393333333333334e-06, | |
| "loss": 0.4437, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 9.33777777777778e-06, | |
| "loss": 0.4146, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.282222222222222e-06, | |
| "loss": 0.4221, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.226666666666668e-06, | |
| "loss": 0.424, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 9.171111111111112e-06, | |
| "loss": 0.4471, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.115555555555556e-06, | |
| "loss": 0.4358, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 9.060000000000001e-06, | |
| "loss": 0.4, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 9.004444444444445e-06, | |
| "loss": 0.4205, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.94888888888889e-06, | |
| "loss": 0.4146, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.893333333333333e-06, | |
| "loss": 0.4343, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_loss": 0.42902088165283203, | |
| "eval_runtime": 2158.1828, | |
| "eval_samples_per_second": 3.483, | |
| "eval_steps_per_second": 0.218, | |
| "eval_wer": 35.183525075949994, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 8.83777777777778e-06, | |
| "loss": 0.418, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.782222222222223e-06, | |
| "loss": 0.4076, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 8.726666666666667e-06, | |
| "loss": 0.4059, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 8.671111111111113e-06, | |
| "loss": 0.3988, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.615555555555555e-06, | |
| "loss": 0.3707, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.560000000000001e-06, | |
| "loss": 0.3797, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 8.504444444444445e-06, | |
| "loss": 0.3647, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.448888888888889e-06, | |
| "loss": 0.3669, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 8.393333333333335e-06, | |
| "loss": 0.3469, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8.337777777777777e-06, | |
| "loss": 0.336, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.282222222222223e-06, | |
| "loss": 0.34, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 8.226666666666667e-06, | |
| "loss": 0.3328, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.171111111111113e-06, | |
| "loss": 0.3497, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 8.115555555555557e-06, | |
| "loss": 0.3747, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.06e-06, | |
| "loss": 0.3966, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 8.004444444444445e-06, | |
| "loss": 0.3475, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 7.948888888888889e-06, | |
| "loss": 0.3441, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.893333333333335e-06, | |
| "loss": 0.3357, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 7.837777777777779e-06, | |
| "loss": 0.378, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.782222222222223e-06, | |
| "loss": 0.3523, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_loss": 0.3780156672000885, | |
| "eval_runtime": 2149.1405, | |
| "eval_samples_per_second": 3.497, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 30.651426863887643, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 7.726666666666667e-06, | |
| "loss": 0.3923, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.67111111111111e-06, | |
| "loss": 0.3449, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.6155555555555564e-06, | |
| "loss": 0.3714, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 7.5600000000000005e-06, | |
| "loss": 0.4006, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.504444444444445e-06, | |
| "loss": 0.3612, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 7.44888888888889e-06, | |
| "loss": 0.3639, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.393333333333333e-06, | |
| "loss": 0.4153, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.337777777777778e-06, | |
| "loss": 0.3818, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 7.282222222222222e-06, | |
| "loss": 0.3564, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.226666666666667e-06, | |
| "loss": 0.3629, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 7.171111111111112e-06, | |
| "loss": 0.2898, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.115555555555557e-06, | |
| "loss": 0.3372, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 7.06e-06, | |
| "loss": 0.3539, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 7.004444444444445e-06, | |
| "loss": 0.3791, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.948888888888889e-06, | |
| "loss": 0.3461, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 6.893333333333334e-06, | |
| "loss": 0.3498, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.837777777777779e-06, | |
| "loss": 0.3502, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 6.782222222222222e-06, | |
| "loss": 0.3696, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.726666666666667e-06, | |
| "loss": 0.3524, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 6.671111111111112e-06, | |
| "loss": 0.3833, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_loss": 0.343627005815506, | |
| "eval_runtime": 2165.5496, | |
| "eval_samples_per_second": 3.471, | |
| "eval_steps_per_second": 0.217, | |
| "eval_wer": 28.031774490761492, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.615555555555556e-06, | |
| "loss": 0.3653, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 6.560000000000001e-06, | |
| "loss": 0.3649, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.504444444444446e-06, | |
| "loss": 0.3044, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.448888888888889e-06, | |
| "loss": 0.2725, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 6.393333333333334e-06, | |
| "loss": 0.2116, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.3377777777777786e-06, | |
| "loss": 0.1644, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 6.282222222222223e-06, | |
| "loss": 0.1579, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.2266666666666675e-06, | |
| "loss": 0.1648, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 6.171111111111112e-06, | |
| "loss": 0.1634, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.1155555555555555e-06, | |
| "loss": 0.1412, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 6.0600000000000004e-06, | |
| "loss": 0.2006, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 6.004444444444445e-06, | |
| "loss": 0.2172, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 5.948888888888889e-06, | |
| "loss": 0.1989, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.893333333333334e-06, | |
| "loss": 0.2036, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 5.837777777777777e-06, | |
| "loss": 0.1612, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.782222222222222e-06, | |
| "loss": 0.192, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 5.726666666666667e-06, | |
| "loss": 0.1787, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5.671111111111112e-06, | |
| "loss": 0.1886, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 5.615555555555556e-06, | |
| "loss": 0.1996, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.560000000000001e-06, | |
| "loss": 0.2029, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 0.33604511618614197, | |
| "eval_runtime": 2142.8094, | |
| "eval_samples_per_second": 3.508, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 26.51277454056477, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 5.504444444444444e-06, | |
| "loss": 0.2064, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.448888888888889e-06, | |
| "loss": 0.1988, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 5.393333333333334e-06, | |
| "loss": 0.1774, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.337777777777779e-06, | |
| "loss": 0.2277, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 5.282222222222223e-06, | |
| "loss": 0.1804, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.226666666666667e-06, | |
| "loss": 0.2029, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 5.171111111111111e-06, | |
| "loss": 0.169, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.115555555555556e-06, | |
| "loss": 0.1985, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 5.060000000000001e-06, | |
| "loss": 0.2063, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 5.004444444444445e-06, | |
| "loss": 0.2117, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 4.94888888888889e-06, | |
| "loss": 0.2109, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.893333333333334e-06, | |
| "loss": 0.163, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 4.837777777777778e-06, | |
| "loss": 0.1834, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.7822222222222226e-06, | |
| "loss": 0.196, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.7266666666666674e-06, | |
| "loss": 0.1746, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.6711111111111115e-06, | |
| "loss": 0.1844, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 4.6155555555555555e-06, | |
| "loss": 0.1548, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.56e-06, | |
| "loss": 0.1546, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 4.504444444444444e-06, | |
| "loss": 0.1539, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.448888888888889e-06, | |
| "loss": 0.1797, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 0.3350614607334137, | |
| "eval_runtime": 2157.9989, | |
| "eval_samples_per_second": 3.483, | |
| "eval_steps_per_second": 0.218, | |
| "eval_wer": 25.892723741222174, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.393333333333334e-06, | |
| "loss": 0.1737, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.337777777777778e-06, | |
| "loss": 0.1736, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.282222222222222e-06, | |
| "loss": 0.2304, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.226666666666667e-06, | |
| "loss": 0.2147, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 4.171111111111111e-06, | |
| "loss": 0.2228, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.115555555555556e-06, | |
| "loss": 0.238, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 4.060000000000001e-06, | |
| "loss": 0.1769, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 4.004444444444445e-06, | |
| "loss": 0.1293, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 3.948888888888889e-06, | |
| "loss": 0.1206, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.893333333333333e-06, | |
| "loss": 0.1162, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 3.837777777777778e-06, | |
| "loss": 0.1242, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.782222222222223e-06, | |
| "loss": 0.1193, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 3.726666666666667e-06, | |
| "loss": 0.0961, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.6711111111111113e-06, | |
| "loss": 0.1158, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 3.615555555555556e-06, | |
| "loss": 0.1642, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.5600000000000002e-06, | |
| "loss": 0.1731, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.5044444444444447e-06, | |
| "loss": 0.1629, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.4488888888888896e-06, | |
| "loss": 0.1607, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 3.3933333333333336e-06, | |
| "loss": 0.1553, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.337777777777778e-06, | |
| "loss": 0.1687, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.32494643330574036, | |
| "eval_runtime": 2166.1529, | |
| "eval_samples_per_second": 3.47, | |
| "eval_steps_per_second": 0.217, | |
| "eval_wer": 25.862841774988794, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 3.282222222222223e-06, | |
| "loss": 0.1774, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.226666666666667e-06, | |
| "loss": 0.1959, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 3.1711111111111114e-06, | |
| "loss": 0.163, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.1155555555555555e-06, | |
| "loss": 0.1777, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 3.0600000000000003e-06, | |
| "loss": 0.1728, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 3.004444444444445e-06, | |
| "loss": 0.1662, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 2.948888888888889e-06, | |
| "loss": 0.1561, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.8933333333333337e-06, | |
| "loss": 0.1787, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 2.837777777777778e-06, | |
| "loss": 0.1754, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.7822222222222222e-06, | |
| "loss": 0.1773, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 2.726666666666667e-06, | |
| "loss": 0.1892, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.6711111111111116e-06, | |
| "loss": 0.1488, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 2.6155555555555556e-06, | |
| "loss": 0.1608, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.56e-06, | |
| "loss": 0.1294, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 2.504444444444445e-06, | |
| "loss": 0.1756, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.448888888888889e-06, | |
| "loss": 0.1413, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 2.3933333333333334e-06, | |
| "loss": 0.1489, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.337777777777778e-06, | |
| "loss": 0.1581, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 2.2822222222222223e-06, | |
| "loss": 0.1448, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.226666666666667e-06, | |
| "loss": 0.1603, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 0.3088971972465515, | |
| "eval_runtime": 2143.1525, | |
| "eval_samples_per_second": 3.507, | |
| "eval_steps_per_second": 0.219, | |
| "eval_wer": 23.987748393844317, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 2.1711111111111113e-06, | |
| "loss": 0.1934, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.1155555555555557e-06, | |
| "loss": 0.1749, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.06e-06, | |
| "loss": 0.1772, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 2.0044444444444446e-06, | |
| "loss": 0.2196, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.948888888888889e-06, | |
| "loss": 0.1208, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.8933333333333333e-06, | |
| "loss": 0.0793, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.837777777777778e-06, | |
| "loss": 0.0843, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.7822222222222225e-06, | |
| "loss": 0.0769, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1.7266666666666667e-06, | |
| "loss": 0.0777, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6711111111111112e-06, | |
| "loss": 0.086, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 1.6155555555555559e-06, | |
| "loss": 0.0766, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.56e-06, | |
| "loss": 0.0885, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.5044444444444446e-06, | |
| "loss": 0.0778, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.4488888888888892e-06, | |
| "loss": 0.088, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.3933333333333335e-06, | |
| "loss": 0.0715, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.337777777777778e-06, | |
| "loss": 0.0683, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.2822222222222222e-06, | |
| "loss": 0.0714, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.2266666666666666e-06, | |
| "loss": 0.0788, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1.171111111111111e-06, | |
| "loss": 0.0792, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.1155555555555558e-06, | |
| "loss": 0.0762, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "eval_loss": 0.3345198631286621, | |
| "eval_runtime": 2127.0754, | |
| "eval_samples_per_second": 3.533, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 24.580407390806318, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 1.06e-06, | |
| "loss": 0.0786, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 1.0044444444444445e-06, | |
| "loss": 0.0932, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 9.488888888888889e-07, | |
| "loss": 0.102, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 8.933333333333334e-07, | |
| "loss": 0.0689, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 8.37777777777778e-07, | |
| "loss": 0.0841, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.822222222222223e-07, | |
| "loss": 0.077, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 7.266666666666668e-07, | |
| "loss": 0.0685, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 6.711111111111111e-07, | |
| "loss": 0.0774, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 6.155555555555556e-07, | |
| "loss": 0.0784, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.6e-07, | |
| "loss": 0.0741, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.044444444444445e-07, | |
| "loss": 0.0779, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 4.488888888888889e-07, | |
| "loss": 0.1124, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.9333333333333336e-07, | |
| "loss": 0.1224, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 3.3777777777777777e-07, | |
| "loss": 0.1299, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 2.822222222222222e-07, | |
| "loss": 0.0879, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 2.266666666666667e-07, | |
| "loss": 0.0581, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.7111111111111114e-07, | |
| "loss": 0.0656, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 1.1555555555555556e-07, | |
| "loss": 0.0622, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 6.000000000000001e-08, | |
| "loss": 0.0549, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 4.444444444444445e-09, | |
| "loss": 0.0676, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_loss": 0.31885361671447754, | |
| "eval_runtime": 2126.4706, | |
| "eval_samples_per_second": 3.534, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 23.81094676029683, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "step": 5000, | |
| "total_flos": 2.717998645248e+20, | |
| "train_loss": 0.2540892813682556, | |
| "train_runtime": 48626.5038, | |
| "train_samples_per_second": 1.645, | |
| "train_steps_per_second": 0.103 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 1000, | |
| "total_flos": 2.717998645248e+20, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |