| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 8340, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 4.9980433547419305e-05, | |
| "loss": 2.3482, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 4.929881824859985e-05, | |
| "loss": 1.0043, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 4.7669293205603196e-05, | |
| "loss": 0.6055, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 4.515542705171981e-05, | |
| "loss": 0.3829, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4.1855287034436555e-05, | |
| "loss": 0.2724, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 3.789761336027403e-05, | |
| "loss": 0.1855, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "learning_rate": 3.3436796971278526e-05, | |
| "loss": 0.1394, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.59, | |
| "learning_rate": 2.864685667276201e-05, | |
| "loss": 0.0952, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.79, | |
| "learning_rate": 2.3714650567721016e-05, | |
| "loss": 0.0658, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 1.883258662346596e-05, | |
| "loss": 0.0426, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 13.19, | |
| "learning_rate": 1.4191116735110007e-05, | |
| "loss": 0.0287, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 9.971307096468203e-06, | |
| "loss": 0.0209, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.59, | |
| "learning_rate": 6.337774712125597e-06, | |
| "loss": 0.0166, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 3.4322656010722542e-06, | |
| "loss": 0.0146, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3681252095757852e-06, | |
| "loss": 0.0129, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 19.18, | |
| "learning_rate": 2.258767454083155e-07, | |
| "loss": 0.0116, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 8340, | |
| "total_flos": 7.012455568878336e+16, | |
| "train_loss": 0.31504503674358486, | |
| "train_runtime": 11228.5117, | |
| "train_samples_per_second": 8.906, | |
| "train_steps_per_second": 0.743 | |
| } | |
| ], | |
| "max_steps": 8340, | |
| "num_train_epochs": 20, | |
| "total_flos": 7.012455568878336e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |