| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.8311170212765957, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.999, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 5e-06, | |
| "loss": 0.7998, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.6289, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5845, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.948979591836737e-06, | |
| "loss": 0.5564, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 9.89795918367347e-06, | |
| "loss": 0.5448, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.846938775510205e-06, | |
| "loss": 0.5098, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.795918367346939e-06, | |
| "loss": 0.5203, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.744897959183674e-06, | |
| "loss": 0.5095, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 9.693877551020408e-06, | |
| "loss": 0.5024, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_loss": 0.507320761680603, | |
| "eval_runtime": 183.288, | |
| "eval_samples_per_second": 5.412, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.2948083939915912, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.642857142857144e-06, | |
| "loss": 0.4942, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.591836734693878e-06, | |
| "loss": 0.4981, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.540816326530612e-06, | |
| "loss": 0.4824, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.489795918367348e-06, | |
| "loss": 0.48, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.438775510204082e-06, | |
| "loss": 0.4599, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.387755102040818e-06, | |
| "loss": 0.4482, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.336734693877552e-06, | |
| "loss": 0.4674, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.285714285714288e-06, | |
| "loss": 0.4648, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.234693877551022e-06, | |
| "loss": 0.4611, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 9.183673469387756e-06, | |
| "loss": 0.4684, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_loss": 0.46387979388237, | |
| "eval_runtime": 182.1295, | |
| "eval_samples_per_second": 5.447, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.2783982515604453, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.13265306122449e-06, | |
| "loss": 0.4524, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 9.081632653061225e-06, | |
| "loss": 0.4511, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.03061224489796e-06, | |
| "loss": 0.4479, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 8.979591836734695e-06, | |
| "loss": 0.4444, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 8.92857142857143e-06, | |
| "loss": 0.4537, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.877551020408163e-06, | |
| "loss": 0.4349, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 8.826530612244899e-06, | |
| "loss": 0.4395, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.775510204081633e-06, | |
| "loss": 0.4236, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.724489795918369e-06, | |
| "loss": 0.428, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 8.673469387755103e-06, | |
| "loss": 0.4246, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.43956783413887024, | |
| "eval_runtime": 182.7895, | |
| "eval_samples_per_second": 5.427, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.27576818358615324, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.622448979591837e-06, | |
| "loss": 0.4315, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8.571428571428571e-06, | |
| "loss": 0.431, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.520408163265307e-06, | |
| "loss": 0.4164, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 8.469387755102042e-06, | |
| "loss": 0.4197, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.418367346938776e-06, | |
| "loss": 0.4328, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.36734693877551e-06, | |
| "loss": 0.4136, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 8.316326530612246e-06, | |
| "loss": 0.4275, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.26530612244898e-06, | |
| "loss": 0.4084, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 8.214285714285714e-06, | |
| "loss": 0.4173, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.16326530612245e-06, | |
| "loss": 0.4132, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.4221729636192322, | |
| "eval_runtime": 183.8667, | |
| "eval_samples_per_second": 5.395, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.26639625122705635, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.112244897959184e-06, | |
| "loss": 0.4238, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 8.06122448979592e-06, | |
| "loss": 0.4153, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 8.010204081632654e-06, | |
| "loss": 0.4125, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 7.959183673469388e-06, | |
| "loss": 0.4161, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 7.908163265306124e-06, | |
| "loss": 0.4055, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 7.857142857142858e-06, | |
| "loss": 0.4028, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7.806122448979593e-06, | |
| "loss": 0.4051, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7.755102040816327e-06, | |
| "loss": 0.4036, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 7.704081632653061e-06, | |
| "loss": 0.4018, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 7.653061224489796e-06, | |
| "loss": 0.4021, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_loss": 0.4100790321826935, | |
| "eval_runtime": 182.7493, | |
| "eval_samples_per_second": 5.428, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.26334018632735084, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 7.602040816326531e-06, | |
| "loss": 0.4036, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.551020408163265e-06, | |
| "loss": 0.3895, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.4034, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 7.448979591836736e-06, | |
| "loss": 0.4014, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.39795918367347e-06, | |
| "loss": 0.4036, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 7.346938775510205e-06, | |
| "loss": 0.3913, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.295918367346939e-06, | |
| "loss": 0.3865, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 7.244897959183675e-06, | |
| "loss": 0.3969, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.193877551020409e-06, | |
| "loss": 0.3928, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.1428571428571436e-06, | |
| "loss": 0.3871, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 0.39817574620246887, | |
| "eval_runtime": 183.3354, | |
| "eval_samples_per_second": 5.411, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.26191402270748826, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 7.091836734693878e-06, | |
| "loss": 0.4017, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 7.0408163265306125e-06, | |
| "loss": 0.4, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 6.989795918367348e-06, | |
| "loss": 0.3988, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.938775510204082e-06, | |
| "loss": 0.3863, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.887755102040817e-06, | |
| "loss": 0.381, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 6.836734693877551e-06, | |
| "loss": 0.373, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 6.785714285714287e-06, | |
| "loss": 0.3968, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 6.734693877551021e-06, | |
| "loss": 0.3876, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 6.683673469387756e-06, | |
| "loss": 0.3899, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 6.63265306122449e-06, | |
| "loss": 0.3813, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.3895135223865509, | |
| "eval_runtime": 182.7821, | |
| "eval_samples_per_second": 5.427, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.25772813987516435, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 6.581632653061225e-06, | |
| "loss": 0.3816, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 6.530612244897959e-06, | |
| "loss": 0.3805, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 6.4795918367346946e-06, | |
| "loss": 0.3901, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 6.4285714285714295e-06, | |
| "loss": 0.3785, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 6.3775510204081635e-06, | |
| "loss": 0.3803, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.326530612244899e-06, | |
| "loss": 0.3803, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.275510204081633e-06, | |
| "loss": 0.3767, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 6.224489795918368e-06, | |
| "loss": 0.3728, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.173469387755102e-06, | |
| "loss": 0.3747, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 6.122448979591837e-06, | |
| "loss": 0.3878, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_loss": 0.38268980383872986, | |
| "eval_runtime": 182.9118, | |
| "eval_samples_per_second": 5.423, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.2533014761719546, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.071428571428571e-06, | |
| "loss": 0.3725, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 6.020408163265307e-06, | |
| "loss": 0.3761, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 5.969387755102042e-06, | |
| "loss": 0.3851, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 5.918367346938776e-06, | |
| "loss": 0.3764, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 5.867346938775511e-06, | |
| "loss": 0.3737, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5.816326530612246e-06, | |
| "loss": 0.3724, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 5.7653061224489805e-06, | |
| "loss": 0.3837, | |
| "step": 2175 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5.7142857142857145e-06, | |
| "loss": 0.3666, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5.663265306122449e-06, | |
| "loss": 0.3748, | |
| "step": 2225 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 5.6122448979591834e-06, | |
| "loss": 0.3704, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 0.3770383894443512, | |
| "eval_runtime": 183.5258, | |
| "eval_samples_per_second": 5.405, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.2533385193828601, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5.561224489795919e-06, | |
| "loss": 0.3625, | |
| "step": 2275 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 5.510204081632653e-06, | |
| "loss": 0.3609, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5.459183673469388e-06, | |
| "loss": 0.3715, | |
| "step": 2325 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5.408163265306123e-06, | |
| "loss": 0.3676, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 5.357142857142857e-06, | |
| "loss": 0.3703, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5.306122448979593e-06, | |
| "loss": 0.3606, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 5.255102040816327e-06, | |
| "loss": 0.3691, | |
| "step": 2425 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.204081632653062e-06, | |
| "loss": 0.3776, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 5.153061224489796e-06, | |
| "loss": 0.3652, | |
| "step": 2475 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.1020408163265315e-06, | |
| "loss": 0.3516, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_loss": 0.3713986277580261, | |
| "eval_runtime": 182.6295, | |
| "eval_samples_per_second": 5.432, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.25398677557370675, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5.0510204081632655e-06, | |
| "loss": 0.3715, | |
| "step": 2525 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 5e-06, | |
| "loss": 0.3557, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.948979591836735e-06, | |
| "loss": 0.3676, | |
| "step": 2575 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.897959183673469e-06, | |
| "loss": 0.3661, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.846938775510204e-06, | |
| "loss": 0.3649, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.795918367346939e-06, | |
| "loss": 0.3525, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 4.744897959183674e-06, | |
| "loss": 0.3715, | |
| "step": 2675 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.693877551020409e-06, | |
| "loss": 0.3639, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 4.642857142857144e-06, | |
| "loss": 0.3619, | |
| "step": 2725 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.591836734693878e-06, | |
| "loss": 0.3792, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_loss": 0.3675037920475006, | |
| "eval_runtime": 183.0666, | |
| "eval_samples_per_second": 5.419, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24952306865959142, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 4.540816326530613e-06, | |
| "loss": 0.3601, | |
| "step": 2775 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.489795918367348e-06, | |
| "loss": 0.363, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.438775510204082e-06, | |
| "loss": 0.3729, | |
| "step": 2825 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.3877551020408165e-06, | |
| "loss": 0.365, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.336734693877551e-06, | |
| "loss": 0.356, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 4.2857142857142855e-06, | |
| "loss": 0.3539, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.234693877551021e-06, | |
| "loss": 0.3608, | |
| "step": 2925 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.183673469387755e-06, | |
| "loss": 0.361, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 4.13265306122449e-06, | |
| "loss": 0.3486, | |
| "step": 2975 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.081632653061225e-06, | |
| "loss": 0.3476, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 0.36362290382385254, | |
| "eval_runtime": 182.4755, | |
| "eval_samples_per_second": 5.436, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.24563353151451167, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 4.03061224489796e-06, | |
| "loss": 0.3764, | |
| "step": 3025 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.979591836734694e-06, | |
| "loss": 0.3542, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 3.928571428571429e-06, | |
| "loss": 0.3615, | |
| "step": 3075 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.877551020408164e-06, | |
| "loss": 0.3655, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.826530612244898e-06, | |
| "loss": 0.3709, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 3.7755102040816327e-06, | |
| "loss": 0.3613, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.724489795918368e-06, | |
| "loss": 0.3608, | |
| "step": 3175 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.6734693877551024e-06, | |
| "loss": 0.353, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.6224489795918373e-06, | |
| "loss": 0.354, | |
| "step": 3225 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.5714285714285718e-06, | |
| "loss": 0.3522, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 0.36105236411094666, | |
| "eval_runtime": 182.9956, | |
| "eval_samples_per_second": 5.421, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24617065807264174, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 3.5204081632653062e-06, | |
| "loss": 0.3526, | |
| "step": 3275 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3.469387755102041e-06, | |
| "loss": 0.3578, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 3.4183673469387756e-06, | |
| "loss": 0.3519, | |
| "step": 3325 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.3673469387755105e-06, | |
| "loss": 0.3487, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 3.316326530612245e-06, | |
| "loss": 0.354, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.2653061224489794e-06, | |
| "loss": 0.3605, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.2142857142857147e-06, | |
| "loss": 0.3644, | |
| "step": 3425 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 3.1632653061224496e-06, | |
| "loss": 0.3592, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.112244897959184e-06, | |
| "loss": 0.3502, | |
| "step": 3475 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 3.0612244897959185e-06, | |
| "loss": 0.3545, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 0.35600587725639343, | |
| "eval_runtime": 183.8697, | |
| "eval_samples_per_second": 5.395, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.243985108629216, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3.0102040816326534e-06, | |
| "loss": 0.3603, | |
| "step": 3525 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.959183673469388e-06, | |
| "loss": 0.3518, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 2.908163265306123e-06, | |
| "loss": 0.3635, | |
| "step": 3575 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.8571428571428573e-06, | |
| "loss": 0.352, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 2.8061224489795917e-06, | |
| "loss": 0.3506, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.7551020408163266e-06, | |
| "loss": 0.3449, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 2.7040816326530615e-06, | |
| "loss": 0.3591, | |
| "step": 3675 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.6530612244897964e-06, | |
| "loss": 0.3499, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.602040816326531e-06, | |
| "loss": 0.3546, | |
| "step": 3725 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 2.5510204081632657e-06, | |
| "loss": 0.3426, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_loss": 0.3543338179588318, | |
| "eval_runtime": 184.2454, | |
| "eval_samples_per_second": 5.384, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24635587412716933, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.354, | |
| "step": 3775 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 2.4489795918367347e-06, | |
| "loss": 0.3416, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.3979591836734696e-06, | |
| "loss": 0.355, | |
| "step": 3825 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.3469387755102044e-06, | |
| "loss": 0.355, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 2.295918367346939e-06, | |
| "loss": 0.3428, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2.244897959183674e-06, | |
| "loss": 0.35, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 2.1938775510204083e-06, | |
| "loss": 0.3511, | |
| "step": 3925 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 2.1428571428571427e-06, | |
| "loss": 0.3477, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 2.0918367346938776e-06, | |
| "loss": 0.3494, | |
| "step": 3975 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 2.0408163265306125e-06, | |
| "loss": 0.3437, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 0.3523597717285156, | |
| "eval_runtime": 184.0518, | |
| "eval_samples_per_second": 5.39, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24635587412716933, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.989795918367347e-06, | |
| "loss": 0.35, | |
| "step": 4025 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 1.938775510204082e-06, | |
| "loss": 0.339, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.8877551020408163e-06, | |
| "loss": 0.3504, | |
| "step": 4075 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 1.8367346938775512e-06, | |
| "loss": 0.3401, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.7857142857142859e-06, | |
| "loss": 0.3424, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.7346938775510206e-06, | |
| "loss": 0.3567, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.6836734693877552e-06, | |
| "loss": 0.3622, | |
| "step": 4175 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.6326530612244897e-06, | |
| "loss": 0.3425, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 1.5816326530612248e-06, | |
| "loss": 0.3546, | |
| "step": 4225 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.5306122448979593e-06, | |
| "loss": 0.3562, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.3507314622402191, | |
| "eval_runtime": 182.7316, | |
| "eval_samples_per_second": 5.429, | |
| "eval_steps_per_second": 0.088, | |
| "eval_wer": 0.2451519697727399, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.479591836734694e-06, | |
| "loss": 0.3458, | |
| "step": 4275 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1.4285714285714286e-06, | |
| "loss": 0.3528, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.3775510204081633e-06, | |
| "loss": 0.3499, | |
| "step": 4325 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.3265306122448982e-06, | |
| "loss": 0.3419, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.2755102040816329e-06, | |
| "loss": 0.3436, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 1.2244897959183673e-06, | |
| "loss": 0.3436, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.1734693877551022e-06, | |
| "loss": 0.3528, | |
| "step": 4425 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.122448979591837e-06, | |
| "loss": 0.3397, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 1.0714285714285714e-06, | |
| "loss": 0.3449, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 1.0224489795918368e-06, | |
| "loss": 0.3555, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 0.34912917017936707, | |
| "eval_runtime": 183.2099, | |
| "eval_samples_per_second": 5.415, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24263303143116446, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 9.714285714285715e-07, | |
| "loss": 0.3449, | |
| "step": 4525 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 9.204081632653062e-07, | |
| "loss": 0.3393, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.693877551020409e-07, | |
| "loss": 0.3454, | |
| "step": 4575 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 8.183673469387755e-07, | |
| "loss": 0.3411, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.673469387755103e-07, | |
| "loss": 0.3416, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 7.16326530612245e-07, | |
| "loss": 0.3298, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.653061224489797e-07, | |
| "loss": 0.3472, | |
| "step": 4675 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 6.142857142857143e-07, | |
| "loss": 0.3519, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 5.632653061224491e-07, | |
| "loss": 0.3387, | |
| "step": 4725 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 5.122448979591837e-07, | |
| "loss": 0.3397, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_loss": 0.34834620356559753, | |
| "eval_runtime": 183.7297, | |
| "eval_samples_per_second": 5.399, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.24191068881850677, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.612244897959184e-07, | |
| "loss": 0.3428, | |
| "step": 4775 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.102040816326531e-07, | |
| "loss": 0.3493, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3.591836734693878e-07, | |
| "loss": 0.3368, | |
| "step": 4825 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 3.0816326530612243e-07, | |
| "loss": 0.3547, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.5714285714285716e-07, | |
| "loss": 0.3426, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 2.0612244897959186e-07, | |
| "loss": 0.3496, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.5510204081632654e-07, | |
| "loss": 0.3479, | |
| "step": 4925 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.0408163265306123e-07, | |
| "loss": 0.3472, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 5.306122448979593e-08, | |
| "loss": 0.3397, | |
| "step": 4975 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 2.0408163265306126e-09, | |
| "loss": 0.3516, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_loss": 0.34790730476379395, | |
| "eval_runtime": 184.5434, | |
| "eval_samples_per_second": 5.375, | |
| "eval_steps_per_second": 0.087, | |
| "eval_wer": 0.2426515530366172, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "step": 5000, | |
| "total_flos": 9.23473281024e+19, | |
| "train_loss": 0.38958581829071043, | |
| "train_runtime": 37554.5281, | |
| "train_samples_per_second": 8.521, | |
| "train_steps_per_second": 0.133 | |
| } | |
| ], | |
| "max_steps": 5000, | |
| "num_train_epochs": 1, | |
| "total_flos": 9.23473281024e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |