| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 8346, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.910136592379583e-05, | |
| "loss": 2.1513, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.8202731847591663e-05, | |
| "loss": 1.4112, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.7304097771387492e-05, | |
| "loss": 1.3035, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.6405463695183324e-05, | |
| "loss": 1.1842, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 1.5506829618979153e-05, | |
| "loss": 1.1289, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_exact_match": 77.59697256385998, | |
| "eval_f1": 86.11156390697873, | |
| "step": 2782 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 1.4608195542774984e-05, | |
| "loss": 0.9913, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.3709561466570813e-05, | |
| "loss": 0.884, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.2810927390366643e-05, | |
| "loss": 0.88, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 1.1912293314162472e-05, | |
| "loss": 0.8624, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.1013659237958305e-05, | |
| "loss": 0.858, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 1.0115025161754135e-05, | |
| "loss": 0.8303, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_exact_match": 79.47019867549669, | |
| "eval_f1": 87.26677361902833, | |
| "step": 5564 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 9.216391085549964e-06, | |
| "loss": 0.6891, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 8.317757009345795e-06, | |
| "loss": 0.6715, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 7.419122933141625e-06, | |
| "loss": 0.6626, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 6.520488856937456e-06, | |
| "loss": 0.6423, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 5.621854780733286e-06, | |
| "loss": 0.6547, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_exact_match": 79.81078524124882, | |
| "eval_f1": 87.66343771063279, | |
| "step": 8346 | |
| } | |
| ], | |
| "max_steps": 11128, | |
| "num_train_epochs": 4, | |
| "total_flos": 5.233878433057997e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |