| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09649122807017543, | |
| "eval_steps": 14, | |
| "global_step": 55, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0017543859649122807, | |
| "eval_loss": 2.72324275970459, | |
| "eval_runtime": 3.8603, | |
| "eval_samples_per_second": 62.172, | |
| "eval_steps_per_second": 15.543, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017543859649122806, | |
| "grad_norm": 3.5063674449920654, | |
| "learning_rate": 0.0002, | |
| "loss": 4.512, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.02456140350877193, | |
| "eval_loss": 1.663864016532898, | |
| "eval_runtime": 3.3198, | |
| "eval_samples_per_second": 72.293, | |
| "eval_steps_per_second": 18.073, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03508771929824561, | |
| "grad_norm": 3.184525489807129, | |
| "learning_rate": 0.0002, | |
| "loss": 3.373, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04912280701754386, | |
| "eval_loss": 1.5565054416656494, | |
| "eval_runtime": 3.2952, | |
| "eval_samples_per_second": 72.834, | |
| "eval_steps_per_second": 18.209, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.05263157894736842, | |
| "grad_norm": 2.7136929035186768, | |
| "learning_rate": 0.0002, | |
| "loss": 3.0622, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07017543859649122, | |
| "grad_norm": 2.9015309810638428, | |
| "learning_rate": 0.0002, | |
| "loss": 3.1672, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.07368421052631578, | |
| "eval_loss": 1.512506127357483, | |
| "eval_runtime": 3.2972, | |
| "eval_samples_per_second": 72.789, | |
| "eval_steps_per_second": 18.197, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.08771929824561403, | |
| "grad_norm": 2.2857017517089844, | |
| "learning_rate": 0.0002, | |
| "loss": 2.9834, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 55, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 14, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3633741456998400.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |