| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.289308176100629, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6122580645161291, | |
| "eval_loss": 0.20235490798950195, | |
| "eval_runtime": 5.9992, | |
| "eval_samples_per_second": 516.736, | |
| "eval_steps_per_second": 10.835, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.3237, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8187096774193549, | |
| "eval_loss": 0.10080192238092422, | |
| "eval_runtime": 5.8671, | |
| "eval_samples_per_second": 528.366, | |
| "eval_steps_per_second": 11.079, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8764516129032258, | |
| "eval_loss": 0.06733943521976471, | |
| "eval_runtime": 5.745, | |
| "eval_samples_per_second": 539.6, | |
| "eval_steps_per_second": 11.314, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1156, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8977419354838709, | |
| "eval_loss": 0.052980903536081314, | |
| "eval_runtime": 5.7366, | |
| "eval_samples_per_second": 540.391, | |
| "eval_steps_per_second": 11.331, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0761, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9029032258064517, | |
| "eval_loss": 0.04506753385066986, | |
| "eval_runtime": 5.8089, | |
| "eval_samples_per_second": 533.661, | |
| "eval_steps_per_second": 11.19, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9067741935483871, | |
| "eval_loss": 0.04121686518192291, | |
| "eval_runtime": 5.7614, | |
| "eval_samples_per_second": 538.061, | |
| "eval_steps_per_second": 11.282, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0627, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 519271419317532.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.7143131281575181, | |
| "num_train_epochs": 7, | |
| "temperature": 11 | |
| } | |
| } | |