| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12195121951219512, | |
| "eval_steps": 10, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.024390243902439025, | |
| "grad_norm": 8.752623558044434, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 0.5615, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04878048780487805, | |
| "grad_norm": 11.641535758972168, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.8518, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04878048780487805, | |
| "eval_loss": 1.3595986366271973, | |
| "eval_runtime": 233.8268, | |
| "eval_samples_per_second": 2.793, | |
| "eval_steps_per_second": 1.398, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07317073170731707, | |
| "grad_norm": 10.912981033325195, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8933, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0975609756097561, | |
| "grad_norm": 10.059850692749023, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.8969, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0975609756097561, | |
| "eval_loss": 1.281290888786316, | |
| "eval_runtime": 234.7195, | |
| "eval_samples_per_second": 2.782, | |
| "eval_steps_per_second": 1.393, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12195121951219512, | |
| "grad_norm": 9.400066375732422, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.8338, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2183717354136384.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |