| { | |
| "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": 0.5729032258064516, | |
| "eval_loss": 0.1942509263753891, | |
| "eval_runtime": 6.3251, | |
| "eval_samples_per_second": 490.111, | |
| "eval_steps_per_second": 10.277, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 1.550763701707098e-05, | |
| "loss": 0.3101, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8106451612903226, | |
| "eval_loss": 0.09739375114440918, | |
| "eval_runtime": 5.9524, | |
| "eval_samples_per_second": 520.8, | |
| "eval_steps_per_second": 10.92, | |
| "step": 636 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.06751346588134766, | |
| "eval_runtime": 5.9559, | |
| "eval_samples_per_second": 520.491, | |
| "eval_steps_per_second": 10.914, | |
| "step": 954 | |
| }, | |
| { | |
| "epoch": 3.14, | |
| "learning_rate": 1.101527403414196e-05, | |
| "loss": 0.1122, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8925806451612903, | |
| "eval_loss": 0.053681887686252594, | |
| "eval_runtime": 6.0061, | |
| "eval_samples_per_second": 516.14, | |
| "eval_steps_per_second": 10.822, | |
| "step": 1272 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 6.522911051212939e-06, | |
| "loss": 0.0749, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9006451612903226, | |
| "eval_loss": 0.045948468148708344, | |
| "eval_runtime": 6.1307, | |
| "eval_samples_per_second": 505.651, | |
| "eval_steps_per_second": 10.602, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9048387096774193, | |
| "eval_loss": 0.0419141948223114, | |
| "eval_runtime": 5.9363, | |
| "eval_samples_per_second": 522.212, | |
| "eval_steps_per_second": 10.95, | |
| "step": 1908 | |
| }, | |
| { | |
| "epoch": 6.29, | |
| "learning_rate": 2.0305480682839176e-06, | |
| "loss": 0.0624, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2226, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 7, | |
| "save_steps": 500, | |
| "total_flos": 520587680240208.0, | |
| "train_batch_size": 48, | |
| "trial_name": null, | |
| "trial_params": { | |
| "alpha": 0.20075631208140787, | |
| "num_train_epochs": 7, | |
| "temperature": 18 | |
| } | |
| } | |