{ "best_metric": null, "best_model_checkpoint": null, "epoch": 25.0, "global_step": 9450, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 1.0, "learning_rate": 1.665198237885463e-05, "loss": 9.873, "step": 378 }, { "epoch": 2.0, "learning_rate": 1.9979082546575355e-05, "loss": 2.5441, "step": 756 }, { "epoch": 3.0, "learning_rate": 1.989409994368087e-05, "loss": 1.9226, "step": 1134 }, { "epoch": 4.0, "learning_rate": 1.9744283913135673e-05, "loss": 1.6855, "step": 1512 }, { "epoch": 5.0, "learning_rate": 1.953061615987832e-05, "loss": 1.5525, "step": 1890 }, { "epoch": 6.0, "learning_rate": 1.9254496792347266e-05, "loss": 1.4545, "step": 2268 }, { "epoch": 7.0, "learning_rate": 1.8917735147939518e-05, "loss": 1.3937, "step": 2646 }, { "epoch": 8.0, "learning_rate": 1.852253793689982e-05, "loss": 1.3153, "step": 3024 }, { "epoch": 9.0, "learning_rate": 1.8071494782330375e-05, "loss": 1.2177, "step": 3402 }, { "epoch": 10.0, "learning_rate": 1.7567561251073607e-05, "loss": 1.1404, "step": 3780 }, { "epoch": 11.0, "learning_rate": 1.7014039486662075e-05, "loss": 1.0829, "step": 4158 }, { "epoch": 12.0, "learning_rate": 1.6414556571242747e-05, "loss": 0.9988, "step": 4536 }, { "epoch": 13.0, "learning_rate": 1.5773040758264232e-05, "loss": 0.9398, "step": 4914 }, { "epoch": 14.0, "learning_rate": 1.50936957316679e-05, "loss": 0.8877, "step": 5292 }, { "epoch": 15.0, "learning_rate": 1.4380973060255608e-05, "loss": 0.8544, "step": 5670 }, { "epoch": 16.0, "learning_rate": 1.36395430277333e-05, "loss": 0.7904, "step": 6048 }, { "epoch": 17.0, "learning_rate": 1.2874264029573555e-05, "loss": 0.7554, "step": 6426 }, { "epoch": 18.0, "learning_rate": 1.2090150737231307e-05, "loss": 0.7142, "step": 6804 }, { "epoch": 19.0, "learning_rate": 1.1292341238324266e-05, "loss": 0.6749, "step": 7182 }, { "epoch": 20.0, "learning_rate": 1.0486063368099783e-05, "loss": 0.6421, "step": 7560 }, { "epoch": 21.0, "learning_rate": 9.67660045280918e-06, "loss": 0.6075, "step": 7938 }, { "epoch": 22.0, "learning_rate": 8.86925668946412e-06, "loss": 0.5757, "step": 8316 }, { "epoch": 23.0, "learning_rate": 8.069322388832444e-06, "loss": 0.5459, "step": 8694 }, { "epoch": 24.0, "learning_rate": 7.282039309426844e-06, "loss": 0.5311, "step": 9072 }, { "epoch": 25.0, "learning_rate": 6.51256630964373e-06, "loss": 0.516, "step": 9450 } ], "max_steps": 15120, "num_train_epochs": 40, "total_flos": 1794560967573504.0, "trial_name": null, "trial_params": null }