| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "eval_steps": 100, | |
| "global_step": 725, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6896551724137931, | |
| "eval_loss": 21.946964263916016, | |
| "eval_runtime": 163.3947, | |
| "eval_samples_per_second": 14.786, | |
| "eval_steps_per_second": 1.848, | |
| "eval_wer": 1.0003446578040374, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.3793103448275863, | |
| "eval_loss": 17.29347801208496, | |
| "eval_runtime": 162.2438, | |
| "eval_samples_per_second": 14.891, | |
| "eval_steps_per_second": 1.861, | |
| "eval_wer": 1.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0689655172413794, | |
| "eval_loss": 3.521054267883301, | |
| "eval_runtime": 161.4701, | |
| "eval_samples_per_second": 14.963, | |
| "eval_steps_per_second": 1.87, | |
| "eval_wer": 1.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7586206896551726, | |
| "eval_loss": 3.507359027862549, | |
| "eval_runtime": 160.5709, | |
| "eval_samples_per_second": 15.046, | |
| "eval_steps_per_second": 1.881, | |
| "eval_wer": 1.0, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.4482758620689653, | |
| "grad_norm": 0.5087113976478577, | |
| "learning_rate": 0.00029759999999999997, | |
| "loss": 9.5434, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.4482758620689653, | |
| "eval_loss": 3.4881255626678467, | |
| "eval_runtime": 161.0375, | |
| "eval_samples_per_second": 15.003, | |
| "eval_steps_per_second": 1.875, | |
| "eval_wer": 1.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.137931034482759, | |
| "eval_loss": 3.487558603286743, | |
| "eval_runtime": 159.8397, | |
| "eval_samples_per_second": 15.115, | |
| "eval_steps_per_second": 1.889, | |
| "eval_wer": 1.0, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.827586206896552, | |
| "eval_loss": 3.489640951156616, | |
| "eval_runtime": 160.42, | |
| "eval_samples_per_second": 15.06, | |
| "eval_steps_per_second": 1.883, | |
| "eval_wer": 1.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 725, | |
| "total_flos": 3.0520251232525025e+18, | |
| "train_loss": 7.648091746363146, | |
| "train_runtime": 3314.4653, | |
| "train_samples_per_second": 6.985, | |
| "train_steps_per_second": 0.219 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 725, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 400, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.0520251232525025e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |