| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.39215686274509803, | |
| "eval_steps": 500, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 4.9985361990992455e-05, | |
| "loss": 4.5464, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 4.9941465105674435e-05, | |
| "loss": 3.5691, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 4.986836074908616e-05, | |
| "loss": 2.6628, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 4.976613452940604e-05, | |
| "loss": 1.6234, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.9634906157700036e-05, | |
| "loss": 0.4965, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 4.9474829307735115e-05, | |
| "loss": 0.2314, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 4.9286091436021015e-05, | |
| "loss": 0.2109, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 4.906891356229103e-05, | |
| "loss": 0.243, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 4.882355001067892e-05, | |
| "loss": 0.1753, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 4.855028811189496e-05, | |
| "loss": 0.2237, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 4.8249447866750025e-05, | |
| "loss": 0.1995, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 4.792138157142158e-05, | |
| "loss": 0.259, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "step": 60, | |
| "total_flos": 5.967075326400922e+16, | |
| "train_loss": 1.2066423391302428, | |
| "train_runtime": 1350.3046, | |
| "train_samples_per_second": 5.439, | |
| "train_steps_per_second": 0.34 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 459, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 5.967075326400922e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |