| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 683, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 3e-05, | |
| "loss": 1.574, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 3e-05, | |
| "loss": 1.474, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4104, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4536923076923077, | |
| "eval_loss": 3.3574578762054443, | |
| "eval_runtime": 4.4299, | |
| "eval_samples_per_second": 112.869, | |
| "eval_steps_per_second": 14.221, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_bleu": 0.08840557310176213, | |
| "eval_exact_match": 0.0, | |
| "eval_prefix_exact_match": 0.058, | |
| "step": 341 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3889, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3e-05, | |
| "loss": 1.387, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 3e-05, | |
| "loss": 1.389, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4543589743589744, | |
| "eval_loss": 3.4179794788360596, | |
| "eval_runtime": 4.1997, | |
| "eval_samples_per_second": 119.055, | |
| "eval_steps_per_second": 15.001, | |
| "step": 683 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_bleu": 0.09820569565497561, | |
| "eval_exact_match": 0.0, | |
| "eval_prefix_exact_match": 0.056, | |
| "step": 683 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 17050, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "total_flos": 1.3526659229967974e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |