| { | |
| "best_metric": 17.77988614800759, | |
| "best_model_checkpoint": "./checkpoint-400", | |
| "epoch": 16.056, | |
| "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": 2.0999999999999997e-07, | |
| "loss": 1.3438, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 4.5e-07, | |
| "loss": 0.8978, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 7e-07, | |
| "loss": 0.4779, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 9.499999999999999e-07, | |
| "loss": 0.3079, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 9.766666666666667e-07, | |
| "loss": 0.2482, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 9.488888888888888e-07, | |
| "loss": 0.2535, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.211111111111111e-07, | |
| "loss": 0.1647, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 8.933333333333333e-07, | |
| "loss": 0.2124, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "eval_loss": 0.24853515625, | |
| "eval_runtime": 632.8323, | |
| "eval_samples_per_second": 3.505, | |
| "eval_steps_per_second": 0.22, | |
| "eval_wer": 18.40607210626186, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 8.655555555555555e-07, | |
| "loss": 0.1383, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "learning_rate": 8.366666666666667e-07, | |
| "loss": 0.153, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 8.088888888888888e-07, | |
| "loss": 0.1306, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 7.81111111111111e-07, | |
| "loss": 0.1085, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.03, | |
| "learning_rate": 7.533333333333332e-07, | |
| "loss": 0.1213, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 5.05, | |
| "learning_rate": 7.255555555555555e-07, | |
| "loss": 0.0798, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 6.966666666666666e-07, | |
| "loss": 0.1017, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 6.688888888888889e-07, | |
| "loss": 0.0704, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "eval_loss": 0.263427734375, | |
| "eval_runtime": 620.9514, | |
| "eval_samples_per_second": 3.572, | |
| "eval_steps_per_second": 0.224, | |
| "eval_wer": 17.77988614800759, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.01, | |
| "learning_rate": 6.411111111111111e-07, | |
| "loss": 0.0748, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 6.133333333333332e-07, | |
| "loss": 0.0669, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 5.844444444444444e-07, | |
| "loss": 0.0546, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 8.03, | |
| "learning_rate": 5.566666666666666e-07, | |
| "loss": 0.0626, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "learning_rate": 5.288888888888888e-07, | |
| "loss": 0.0431, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 5.011111111111111e-07, | |
| "loss": 0.0516, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 4.733333333333333e-07, | |
| "loss": 0.0386, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "learning_rate": 4.444444444444444e-07, | |
| "loss": 0.0379, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.01, | |
| "eval_loss": 0.310302734375, | |
| "eval_runtime": 621.7077, | |
| "eval_samples_per_second": 3.568, | |
| "eval_steps_per_second": 0.224, | |
| "eval_wer": 17.817836812144215, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 4.1666666666666667e-07, | |
| "loss": 0.0367, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.888888888888889e-07, | |
| "loss": 0.0271, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 11.03, | |
| "learning_rate": 3.6111111111111107e-07, | |
| "loss": 0.0334, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 3.333333333333333e-07, | |
| "loss": 0.0247, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 12.02, | |
| "learning_rate": 3.0444444444444445e-07, | |
| "loss": 0.0277, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 2.766666666666667e-07, | |
| "loss": 0.0229, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 2.4888888888888886e-07, | |
| "loss": 0.0222, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "learning_rate": 2.2111111111111111e-07, | |
| "loss": 0.0228, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "eval_loss": 0.35546875, | |
| "eval_runtime": 628.2239, | |
| "eval_samples_per_second": 3.531, | |
| "eval_steps_per_second": 0.221, | |
| "eval_wer": 18.40607210626186, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 1.9333333333333332e-07, | |
| "loss": 0.0168, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 14.02, | |
| "learning_rate": 1.6444444444444444e-07, | |
| "loss": 0.0222, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 14.05, | |
| "learning_rate": 1.3666666666666665e-07, | |
| "loss": 0.0166, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 15.02, | |
| "learning_rate": 1.0888888888888888e-07, | |
| "loss": 0.0189, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 15.04, | |
| "learning_rate": 8.11111111111111e-08, | |
| "loss": 0.0168, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 16.01, | |
| "learning_rate": 5.2222222222222224e-08, | |
| "loss": 0.0158, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 16.03, | |
| "learning_rate": 2.4444444444444444e-08, | |
| "loss": 0.0185, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "learning_rate": 0.0, | |
| "loss": 0.0139, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "eval_loss": 0.373291015625, | |
| "eval_runtime": 625.4198, | |
| "eval_samples_per_second": 3.546, | |
| "eval_steps_per_second": 0.222, | |
| "eval_wer": 18.453510436432637, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 16.06, | |
| "step": 1000, | |
| "total_flos": 1.3683412531969996e+20, | |
| "train_loss": 0.13993059158325194, | |
| "train_runtime": 17951.1904, | |
| "train_samples_per_second": 3.565, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 1.3683412531969996e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |