| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.0, | |
| "global_step": 9370, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 4.7331910352187837e-05, | |
| "loss": 0.0788, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 4.466382070437567e-05, | |
| "loss": 0.0304, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.1995731056563505e-05, | |
| "loss": 0.0206, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3.932764140875134e-05, | |
| "loss": 0.0184, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 3.665955176093917e-05, | |
| "loss": 0.0175, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3.3991462113127e-05, | |
| "loss": 0.0116, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3.1323372465314835e-05, | |
| "loss": 0.0086, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 2.8655282817502672e-05, | |
| "loss": 0.0139, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 2.5987193169690503e-05, | |
| "loss": 0.0065, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 2.3319103521878334e-05, | |
| "loss": 0.0053, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 2.0651013874066168e-05, | |
| "loss": 0.0061, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.7982924226254002e-05, | |
| "loss": 0.0074, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.47, | |
| "learning_rate": 1.5314834578441836e-05, | |
| "loss": 0.0036, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "learning_rate": 1.264674493062967e-05, | |
| "loss": 0.0042, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 9.978655282817503e-06, | |
| "loss": 0.0043, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 7.310565635005337e-06, | |
| "loss": 0.0019, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 4.6424759871931695e-06, | |
| "loss": 0.0038, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.9743863393810032e-06, | |
| "loss": 0.002, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 9370, | |
| "total_flos": 4794764486084310.0, | |
| "train_loss": 0.013111605267005643, | |
| "train_runtime": 1302.6632, | |
| "train_samples_per_second": 57.517, | |
| "train_steps_per_second": 7.193 | |
| } | |
| ], | |
| "max_steps": 9370, | |
| "num_train_epochs": 5, | |
| "total_flos": 4794764486084310.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |