| { | |
| "best_metric": 26.093088857545837, | |
| "best_model_checkpoint": "./checkpoint-250", | |
| "epoch": 49.02, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.25e-07, | |
| "loss": 0.0008, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.0004, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1.8750000000000003e-06, | |
| "loss": 0.0002, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0001, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.0001, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.0001, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 4.84375e-06, | |
| "loss": 0.0, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 4.6875000000000004e-06, | |
| "loss": 0.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "eval_loss": 0.5089225769042969, | |
| "eval_runtime": 291.4341, | |
| "eval_samples_per_second": 0.755, | |
| "eval_steps_per_second": 0.024, | |
| "eval_wer": 26.093088857545837, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 13.02, | |
| "learning_rate": 4.53125e-06, | |
| "loss": 0.0, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 0.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4.21875e-06, | |
| "loss": 0.0, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 17.01, | |
| "learning_rate": 4.0625000000000005e-06, | |
| "loss": 0.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 3.90625e-06, | |
| "loss": 0.0, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 3.59375e-06, | |
| "loss": 0.0, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 22.01, | |
| "learning_rate": 3.4375e-06, | |
| "loss": 0.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 23.02, | |
| "learning_rate": 3.28125e-06, | |
| "loss": 0.0, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "learning_rate": 3.125e-06, | |
| "loss": 0.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 24.02, | |
| "eval_loss": 0.5541692972183228, | |
| "eval_runtime": 294.1855, | |
| "eval_samples_per_second": 0.748, | |
| "eval_steps_per_second": 0.024, | |
| "eval_wer": 26.629055007052187, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2.96875e-06, | |
| "loss": 0.0, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 27.01, | |
| "learning_rate": 2.8125e-06, | |
| "loss": 0.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 28.02, | |
| "learning_rate": 2.65625e-06, | |
| "loss": 0.0, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 29.02, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "learning_rate": 2.3437500000000002e-06, | |
| "loss": 0.0, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 32.01, | |
| "learning_rate": 2.1875000000000002e-06, | |
| "loss": 0.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 33.02, | |
| "learning_rate": 2.0312500000000002e-06, | |
| "loss": 0.0, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 34.02, | |
| "learning_rate": 1.8750000000000003e-06, | |
| "loss": 0.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 36.01, | |
| "learning_rate": 1.71875e-06, | |
| "loss": 0.0, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "learning_rate": 1.5625e-06, | |
| "loss": 0.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 37.01, | |
| "eval_loss": 0.5804433822631836, | |
| "eval_runtime": 298.447, | |
| "eval_samples_per_second": 0.737, | |
| "eval_steps_per_second": 0.023, | |
| "eval_wer": 26.516220028208743, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 38.02, | |
| "learning_rate": 1.40625e-06, | |
| "loss": 0.0, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 39.02, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.0, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 41.01, | |
| "learning_rate": 1.0937500000000001e-06, | |
| "loss": 0.0, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 42.01, | |
| "learning_rate": 9.375000000000001e-07, | |
| "loss": 0.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 43.02, | |
| "learning_rate": 7.8125e-07, | |
| "loss": 0.0, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 44.02, | |
| "learning_rate": 6.25e-07, | |
| "loss": 0.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 46.01, | |
| "learning_rate": 4.6875000000000006e-07, | |
| "loss": 0.0, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 47.01, | |
| "learning_rate": 3.125e-07, | |
| "loss": 0.0, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 48.02, | |
| "learning_rate": 1.5625e-07, | |
| "loss": 0.0, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "eval_loss": 0.590242326259613, | |
| "eval_runtime": 291.7429, | |
| "eval_samples_per_second": 0.754, | |
| "eval_steps_per_second": 0.024, | |
| "eval_wer": 26.459802538787024, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 49.02, | |
| "step": 1000, | |
| "total_flos": 8.917288869888e+18, | |
| "train_loss": 5.474784423131496e-05, | |
| "train_runtime": 4906.4923, | |
| "train_samples_per_second": 6.522, | |
| "train_steps_per_second": 0.204 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 8.917288869888e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |