| { | |
| "best_metric": 23.5020721174329, | |
| "best_model_checkpoint": "./checkpoint-800", | |
| "epoch": 9.073, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 6.712113404111031e-06, | |
| "loss": 0.5024, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.360489289678585e-06, | |
| "loss": 0.3849, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 9.28666248215634e-06, | |
| "loss": 0.3349, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.933858671331224e-06, | |
| "loss": 0.3442, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 9.766666666666667e-06, | |
| "loss": 0.2877, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 9.48888888888889e-06, | |
| "loss": 0.2288, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 9.211111111111111e-06, | |
| "loss": 0.1896, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 8.933333333333333e-06, | |
| "loss": 0.1935, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_loss": 0.424072265625, | |
| "eval_runtime": 1295.5788, | |
| "eval_samples_per_second": 2.549, | |
| "eval_steps_per_second": 0.16, | |
| "eval_wer": 29.652581432278552, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 8.644444444444445e-06, | |
| "loss": 0.1725, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 8.366666666666667e-06, | |
| "loss": 0.1351, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 8.08888888888889e-06, | |
| "loss": 0.116, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 7.822222222222224e-06, | |
| "loss": 0.1143, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 7.544444444444445e-06, | |
| "loss": 0.1026, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "learning_rate": 7.266666666666668e-06, | |
| "loss": 0.0801, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 6.9888888888888895e-06, | |
| "loss": 0.0655, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "learning_rate": 6.711111111111111e-06, | |
| "loss": 0.0649, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "eval_loss": 0.46826171875, | |
| "eval_runtime": 1242.5517, | |
| "eval_samples_per_second": 2.657, | |
| "eval_steps_per_second": 0.167, | |
| "eval_wer": 26.062202254153917, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 6.422222222222223e-06, | |
| "loss": 0.0575, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 6.155555555555556e-06, | |
| "loss": 0.0485, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "learning_rate": 5.877777777777778e-06, | |
| "loss": 0.0347, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "learning_rate": 5.600000000000001e-06, | |
| "loss": 0.0331, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "learning_rate": 5.322222222222223e-06, | |
| "loss": 0.0307, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 5.044444444444445e-06, | |
| "loss": 0.0264, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 4.766666666666667e-06, | |
| "loss": 0.0192, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 4.488888888888889e-06, | |
| "loss": 0.0156, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "eval_loss": 0.54443359375, | |
| "eval_runtime": 1180.1513, | |
| "eval_samples_per_second": 2.798, | |
| "eval_steps_per_second": 0.175, | |
| "eval_wer": 25.810449668848523, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 4.2000000000000004e-06, | |
| "loss": 0.0136, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 6.03, | |
| "learning_rate": 3.922222222222223e-06, | |
| "loss": 0.0119, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 3.644444444444445e-06, | |
| "loss": 0.0098, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "learning_rate": 3.366666666666667e-06, | |
| "loss": 0.0071, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.088888888888889e-06, | |
| "loss": 0.0069, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 7.03, | |
| "learning_rate": 2.811111111111111e-06, | |
| "loss": 0.0062, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 7.05, | |
| "learning_rate": 2.5333333333333338e-06, | |
| "loss": 0.0046, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 2.2555555555555557e-06, | |
| "loss": 0.0039, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_loss": 0.5947265625, | |
| "eval_runtime": 1100.8609, | |
| "eval_samples_per_second": 2.999, | |
| "eval_steps_per_second": 0.188, | |
| "eval_wer": 23.5020721174329, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.9666666666666668e-06, | |
| "loss": 0.0039, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 1.688888888888889e-06, | |
| "loss": 0.0031, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 1.4111111111111111e-06, | |
| "loss": 0.0026, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "learning_rate": 1.1333333333333334e-06, | |
| "loss": 0.0023, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "learning_rate": 8.555555555555556e-07, | |
| "loss": 0.0023, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 5.777777777777778e-07, | |
| "loss": 0.0021, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "learning_rate": 3.0000000000000004e-07, | |
| "loss": 0.002, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 2.2222222222222224e-08, | |
| "loss": 0.0019, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "eval_loss": 0.6123046875, | |
| "eval_runtime": 1135.4108, | |
| "eval_samples_per_second": 2.908, | |
| "eval_steps_per_second": 0.182, | |
| "eval_wer": 23.893256903830512, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "step": 1000, | |
| "total_flos": 1.3615046685661279e+20, | |
| "train_loss": 0.09166860008239747, | |
| "train_runtime": 21210.666, | |
| "train_samples_per_second": 3.017, | |
| "train_steps_per_second": 0.047 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.3615046685661279e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |