| { | |
| "best_metric": 32.04202832343535, | |
| "best_model_checkpoint": "./checkpoint-2000", | |
| "epoch": 13.038, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 2.0833333333333333e-07, | |
| "loss": 0.011, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 0.0111, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 6.25e-07, | |
| "loss": 0.0111, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.333333333333333e-07, | |
| "loss": 0.0132, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.0416666666666667e-06, | |
| "loss": 0.0139, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.013, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 1.4583333333333335e-06, | |
| "loss": 0.0092, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.0098, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 1.8750000000000003e-06, | |
| "loss": 0.0091, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 2.0833333333333334e-06, | |
| "loss": 0.0101, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.2916666666666666e-06, | |
| "loss": 0.0092, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0084, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.7083333333333334e-06, | |
| "loss": 0.0067, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 2.916666666666667e-06, | |
| "loss": 0.0066, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.0066, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.0066, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_loss": 0.374257892370224, | |
| "eval_runtime": 1097.1851, | |
| "eval_samples_per_second": 4.292, | |
| "eval_steps_per_second": 0.135, | |
| "eval_wer": 32.910004568296024, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.5416666666666673e-06, | |
| "loss": 0.0059, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.0053, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 3.958333333333333e-06, | |
| "loss": 0.0046, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.0052, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.0049, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 4.583333333333333e-06, | |
| "loss": 0.0051, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 4.791666666666668e-06, | |
| "loss": 0.0047, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0052, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 5.208333333333334e-06, | |
| "loss": 0.0049, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 5.416666666666667e-06, | |
| "loss": 0.0052, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 5.625e-06, | |
| "loss": 0.0059, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 5.833333333333334e-06, | |
| "loss": 0.0084, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 6.041666666666667e-06, | |
| "loss": 0.0101, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.0078, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 6.458333333333334e-06, | |
| "loss": 0.0077, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.0084, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "eval_loss": 0.3786599338054657, | |
| "eval_runtime": 1048.12, | |
| "eval_samples_per_second": 4.493, | |
| "eval_steps_per_second": 0.141, | |
| "eval_wer": 33.41708542713568, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 6.875e-06, | |
| "loss": 0.0098, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 7.083333333333335e-06, | |
| "loss": 0.0097, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 7.291666666666667e-06, | |
| "loss": 0.0093, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0092, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 7.708333333333334e-06, | |
| "loss": 0.0086, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 7.916666666666667e-06, | |
| "loss": 0.0087, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 8.125000000000001e-06, | |
| "loss": 0.0091, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.0119, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 8.541666666666666e-06, | |
| "loss": 0.0115, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 0.0093, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 8.958333333333334e-06, | |
| "loss": 0.01, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 9.166666666666666e-06, | |
| "loss": 0.0101, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.0095, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "learning_rate": 9.583333333333335e-06, | |
| "loss": 0.0103, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 9.791666666666666e-06, | |
| "loss": 0.0111, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0098, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.01, | |
| "eval_loss": 0.397890567779541, | |
| "eval_runtime": 1133.829, | |
| "eval_samples_per_second": 4.153, | |
| "eval_steps_per_second": 0.131, | |
| "eval_wer": 33.24805847418913, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 9.6875e-06, | |
| "loss": 0.0106, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 0.009, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 9.0625e-06, | |
| "loss": 0.0093, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 8.06, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 0.0087, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 8.4375e-06, | |
| "loss": 0.0098, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 9.01, | |
| "learning_rate": 8.125000000000001e-06, | |
| "loss": 0.0057, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 7.8125e-06, | |
| "loss": 0.0053, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 0.0061, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 7.1875e-06, | |
| "loss": 0.007, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "learning_rate": 6.875e-06, | |
| "loss": 0.0049, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 6.5625e-06, | |
| "loss": 0.0042, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 6.25e-06, | |
| "loss": 0.003, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 5.9375e-06, | |
| "loss": 0.0048, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 5.625e-06, | |
| "loss": 0.0025, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 10.05, | |
| "learning_rate": 5.3125e-06, | |
| "loss": 0.0028, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0019, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "eval_loss": 0.40835869312286377, | |
| "eval_runtime": 1079.5315, | |
| "eval_samples_per_second": 4.362, | |
| "eval_steps_per_second": 0.137, | |
| "eval_wer": 32.311557788944725, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 10.07, | |
| "learning_rate": 4.6875000000000004e-06, | |
| "loss": 0.0024, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.0013, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 11.02, | |
| "learning_rate": 4.0625000000000005e-06, | |
| "loss": 0.0017, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.0012, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 3.4375e-06, | |
| "loss": 0.0012, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.001, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 11.07, | |
| "learning_rate": 2.8125e-06, | |
| "loss": 0.0011, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0009, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 2.1875000000000002e-06, | |
| "loss": 0.001, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1.8750000000000003e-06, | |
| "loss": 0.0008, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 12.05, | |
| "learning_rate": 1.5625e-06, | |
| "loss": 0.0008, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.0009, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 9.375000000000001e-07, | |
| "loss": 0.0009, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 6.25e-07, | |
| "loss": 0.0008, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 3.125e-07, | |
| "loss": 0.0009, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "learning_rate": 0.0, | |
| "loss": 0.0008, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "eval_loss": 0.4333903193473816, | |
| "eval_runtime": 1067.3464, | |
| "eval_samples_per_second": 4.412, | |
| "eval_steps_per_second": 0.139, | |
| "eval_wer": 32.04202832343535, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 13.04, | |
| "step": 2000, | |
| "total_flos": 3.673634428993536e+19, | |
| "train_loss": 0.006580068808048963, | |
| "train_runtime": 18902.2125, | |
| "train_samples_per_second": 6.772, | |
| "train_steps_per_second": 0.106 | |
| } | |
| ], | |
| "max_steps": 2000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 3.673634428993536e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |