| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 4860, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 4.48559670781893e-05, | |
| "loss": 0.5257, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 3.971193415637861e-05, | |
| "loss": 0.3776, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3.45679012345679e-05, | |
| "loss": 0.3303, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.9423868312757202e-05, | |
| "loss": 0.1916, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 2.4279835390946505e-05, | |
| "loss": 0.154, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.91358024691358e-05, | |
| "loss": 0.1453, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.3991769547325103e-05, | |
| "loss": 0.1009, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 8.847736625514404e-06, | |
| "loss": 0.0669, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0613, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 4860, | |
| "total_flos": 1.0749309854975224e+18, | |
| "train_loss": 0.2053911095293461, | |
| "train_runtime": 288593.9524, | |
| "train_samples_per_second": 8.622, | |
| "train_steps_per_second": 0.017 | |
| } | |
| ], | |
| "max_steps": 4860, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.0749309854975224e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |