| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 9822, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 1.8981877418041133e-05, | |
| "loss": 0.4423, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.7963754836082265e-05, | |
| "loss": 0.3526, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 1.69456322541234e-05, | |
| "loss": 0.329, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.592750967216453e-05, | |
| "loss": 0.3065, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.4909387090205662e-05, | |
| "loss": 0.3113, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 1.3891264508246794e-05, | |
| "loss": 0.2888, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.2873141926287927e-05, | |
| "loss": 0.2525, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.1855019344329057e-05, | |
| "loss": 0.1912, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.083689676237019e-05, | |
| "loss": 0.1926, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 9.818774180411322e-06, | |
| "loss": 0.198, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.800651598452455e-06, | |
| "loss": 0.1876, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 7.782529016493586e-06, | |
| "loss": 0.1815, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.764406434534719e-06, | |
| "loss": 0.1856, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 5.74628385257585e-06, | |
| "loss": 0.1215, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 4.728161270616982e-06, | |
| "loss": 0.1146, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.7100386886581147e-06, | |
| "loss": 0.1142, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 2.6919161066992467e-06, | |
| "loss": 0.1148, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.6737935247403788e-06, | |
| "loss": 0.1169, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "learning_rate": 6.55670942781511e-07, | |
| "loss": 0.1159, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 9822, | |
| "total_flos": 2.6421487163260416e+16, | |
| "train_runtime": 2637.8461, | |
| "train_samples_per_second": 119.123, | |
| "train_steps_per_second": 3.723 | |
| } | |
| ], | |
| "max_steps": 9822, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.6421487163260416e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |