| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.012, | |
| "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": 4.4e-06, | |
| "loss": 1.6269, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 9.4e-06, | |
| "loss": 0.5866, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3799, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4584, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2436, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.146, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.174, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.098, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0585, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0626, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0425, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 3.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0259, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0295, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0185, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0127, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0125, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.011, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0074, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0097, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.007, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0068, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0089, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0063, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0046, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0059, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.004, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0041, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0041, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0044, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 9.07, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0033, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 10.02, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0047, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 10.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0053, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.06, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0036, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0035, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0022, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 11.06, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0029, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.01, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0042, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0049, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0035, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0027, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "eval_loss": 0.46246904134750366, | |
| "eval_runtime": 298.1769, | |
| "eval_samples_per_second": 4.048, | |
| "eval_steps_per_second": 0.255, | |
| "eval_wer": 26.588921282798832, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.01, | |
| "step": 1000, | |
| "total_flos": 9.1597006061568e+18, | |
| "train_loss": 0.10252290242910385, | |
| "train_runtime": 3638.1847, | |
| "train_samples_per_second": 8.796, | |
| "train_steps_per_second": 0.275 | |
| } | |
| ], | |
| "max_steps": 1000, | |
| "num_train_epochs": 9223372036854775807, | |
| "total_flos": 9.1597006061568e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |