| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.3824091778202677, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0076481835564053535, | |
| "eval_loss": 3.030160427093506, | |
| "eval_runtime": 2.2559, | |
| "eval_samples_per_second": 24.38, | |
| "eval_steps_per_second": 12.412, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07648183556405354, | |
| "grad_norm": 21.069334030151367, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 9.16, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0994263862332696, | |
| "eval_loss": 0.5972093343734741, | |
| "eval_runtime": 1.4991, | |
| "eval_samples_per_second": 36.69, | |
| "eval_steps_per_second": 18.678, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.15296367112810708, | |
| "grad_norm": 9.656890869140625, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 1.8166, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1988527724665392, | |
| "eval_loss": 0.2717660069465637, | |
| "eval_runtime": 1.5028, | |
| "eval_samples_per_second": 36.598, | |
| "eval_steps_per_second": 18.632, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2294455066921606, | |
| "grad_norm": 9.939780235290527, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.979, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2982791586998088, | |
| "eval_loss": 0.18662667274475098, | |
| "eval_runtime": 1.5323, | |
| "eval_samples_per_second": 35.893, | |
| "eval_steps_per_second": 18.273, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.30592734225621415, | |
| "grad_norm": 6.2070231437683105, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.91, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3824091778202677, | |
| "grad_norm": 4.962741374969482, | |
| "learning_rate": 0.0, | |
| "loss": 0.6838, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8162390153625600.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |