| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.769230769230769, | |
| "eval_steps": 500, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 10.593662748249553, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.6581, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 6.546361411339603, | |
| "learning_rate": 4.578674030756364e-06, | |
| "loss": 1.6436, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 2.207878589630127, | |
| "eval_runtime": 4.1795, | |
| "eval_samples_per_second": 11.006, | |
| "eval_steps_per_second": 0.718, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 5.902377335162108, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.4254, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.200706720352173, | |
| "eval_runtime": 3.9993, | |
| "eval_samples_per_second": 11.502, | |
| "eval_steps_per_second": 0.75, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 5.611927864077392, | |
| "learning_rate": 4.2132596924363666e-07, | |
| "loss": 1.1406, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_loss": 2.207685708999634, | |
| "eval_runtime": 3.9892, | |
| "eval_samples_per_second": 11.531, | |
| "eval_steps_per_second": 0.752, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "step": 18, | |
| "total_flos": 3716488888320.0, | |
| "train_loss": 1.355054795742035, | |
| "train_runtime": 319.8737, | |
| "train_samples_per_second": 3.789, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 3716488888320.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |