| { | |
| "best_metric": 0.8155183792114258, | |
| "best_model_checkpoint": "./output/training_results/C020_Meta-Llama-3-8B_instruct_20240729_032429/checkpoint-10", | |
| "epoch": 0.20833333333333334, | |
| "eval_steps": 5, | |
| "global_step": 10, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.020833333333333332, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.0187, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0625, | |
| "grad_norm": 0.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.946, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.10416666666666667, | |
| "eval_loss": 1.0251731872558594, | |
| "eval_runtime": 0.8943, | |
| "eval_samples_per_second": 380.189, | |
| "eval_steps_per_second": 3.355, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.125, | |
| "grad_norm": 13.805769787987279, | |
| "learning_rate": 2.25e-06, | |
| "loss": 1.0778, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1875, | |
| "grad_norm": 5.41697677358006, | |
| "learning_rate": 2.865909090909091e-06, | |
| "loss": 0.8781, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.20833333333333334, | |
| "eval_loss": 0.8155183792114258, | |
| "eval_runtime": 0.8756, | |
| "eval_samples_per_second": 388.315, | |
| "eval_steps_per_second": 3.426, | |
| "step": 10 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "total_flos": 207641640960.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |