{ "best_metric": 0.608456552028656, "best_model_checkpoint": "./lora-alpaca/checkpoint-200", "epoch": 0.3341687552213868, "global_step": 200, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.02, "learning_rate": 2.9999999999999997e-05, "loss": 2.7003, "step": 10 }, { "epoch": 0.03, "learning_rate": 5.9999999999999995e-05, "loss": 2.566, "step": 20 }, { "epoch": 0.05, "learning_rate": 8.999999999999999e-05, "loss": 2.2648, "step": 30 }, { "epoch": 0.07, "learning_rate": 0.00011099999999999999, "loss": 1.657, "step": 40 }, { "epoch": 0.08, "learning_rate": 0.00014099999999999998, "loss": 1.1599, "step": 50 }, { "epoch": 0.1, "learning_rate": 0.00017099999999999998, "loss": 0.9037, "step": 60 }, { "epoch": 0.12, "learning_rate": 0.000201, "loss": 0.8137, "step": 70 }, { "epoch": 0.13, "learning_rate": 0.00023099999999999998, "loss": 0.7827, "step": 80 }, { "epoch": 0.15, "learning_rate": 0.000261, "loss": 0.7554, "step": 90 }, { "epoch": 0.17, "learning_rate": 0.00029099999999999997, "loss": 0.7357, "step": 100 }, { "epoch": 0.18, "learning_rate": 0.0002957831325301205, "loss": 0.6893, "step": 110 }, { "epoch": 0.2, "learning_rate": 0.00028975903614457827, "loss": 0.6606, "step": 120 }, { "epoch": 0.22, "learning_rate": 0.0002837349397590361, "loss": 0.6506, "step": 130 }, { "epoch": 0.23, "learning_rate": 0.00027771084337349395, "loss": 0.6462, "step": 140 }, { "epoch": 0.25, "learning_rate": 0.0002716867469879518, "loss": 0.6315, "step": 150 }, { "epoch": 0.27, "learning_rate": 0.0002656626506024096, "loss": 0.6337, "step": 160 }, { "epoch": 0.28, "learning_rate": 0.00025963855421686746, "loss": 0.6223, "step": 170 }, { "epoch": 0.3, "learning_rate": 0.00025361445783132525, "loss": 0.6136, "step": 180 }, { "epoch": 0.32, "learning_rate": 0.00024759036144578314, "loss": 0.6198, "step": 190 }, { "epoch": 0.33, "learning_rate": 0.00024156626506024095, "loss": 0.6084, "step": 200 }, { "epoch": 0.33, "eval_loss": 0.608456552028656, "eval_runtime": 123.856, "eval_samples_per_second": 16.148, "eval_steps_per_second": 1.009, "step": 200 } ], "max_steps": 598, "num_train_epochs": 1, "total_flos": 1.7194991763849216e+17, "trial_name": null, "trial_params": null }