| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00046425669824207415, | |
| "eval_steps": 10, | |
| "global_step": 26, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.7856026855464392e-05, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1050.5981, | |
| "eval_samples_per_second": 22.445, | |
| "eval_steps_per_second": 11.223, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 8.928013427732195e-05, | |
| "grad_norm": NaN, | |
| "learning_rate": 5e-05, | |
| "loss": 6.0812, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0001785602685546439, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0001785602685546439, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1061.0365, | |
| "eval_samples_per_second": 22.224, | |
| "eval_steps_per_second": 11.113, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00026784040283196585, | |
| "grad_norm": NaN, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0003571205371092878, | |
| "grad_norm": NaN, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0003571205371092878, | |
| "eval_loss": NaN, | |
| "eval_runtime": 1063.5522, | |
| "eval_samples_per_second": 22.172, | |
| "eval_steps_per_second": 11.086, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0004464006713866098, | |
| "grad_norm": NaN, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1621491292569600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |