{ "best_metric": null, "best_model_checkpoint": null, "epoch": 1.0001094690749863, "global_step": 2284, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.04, "learning_rate": 2e-05, "loss": 1.5875, "step": 100 }, { "epoch": 0.09, "learning_rate": 4e-05, "loss": 0.9455, "step": 200 }, { "epoch": 0.13, "learning_rate": 3.8080614203454896e-05, "loss": 0.9132, "step": 300 }, { "epoch": 0.18, "learning_rate": 3.6161228406909795e-05, "loss": 0.8896, "step": 400 }, { "epoch": 0.22, "learning_rate": 3.424184261036469e-05, "loss": 0.8479, "step": 500 }, { "epoch": 0.26, "learning_rate": 3.232245681381958e-05, "loss": 0.8214, "step": 600 }, { "epoch": 0.31, "learning_rate": 3.0403071017274476e-05, "loss": 0.7844, "step": 700 }, { "epoch": 0.35, "learning_rate": 2.8483685220729368e-05, "loss": 0.7575, "step": 800 }, { "epoch": 0.39, "learning_rate": 2.6564299424184264e-05, "loss": 0.7334, "step": 900 }, { "epoch": 0.44, "learning_rate": 2.4644913627639156e-05, "loss": 0.6934, "step": 1000 }, { "epoch": 0.48, "learning_rate": 2.2725527831094052e-05, "loss": 0.6665, "step": 1100 }, { "epoch": 0.53, "learning_rate": 2.0806142034548945e-05, "loss": 0.6466, "step": 1200 }, { "epoch": 0.57, "learning_rate": 1.888675623800384e-05, "loss": 0.6095, "step": 1300 }, { "epoch": 0.61, "learning_rate": 1.6967370441458736e-05, "loss": 0.5723, "step": 1400 }, { "epoch": 0.66, "learning_rate": 1.5047984644913629e-05, "loss": 0.5313, "step": 1500 }, { "epoch": 0.7, "learning_rate": 1.3128598848368523e-05, "loss": 0.514, "step": 1600 }, { "epoch": 0.74, "learning_rate": 1.1209213051823417e-05, "loss": 0.4911, "step": 1700 }, { "epoch": 0.79, "learning_rate": 9.289827255278311e-06, "loss": 0.4456, "step": 1800 }, { "epoch": 0.83, "learning_rate": 7.370441458733206e-06, "loss": 0.423, "step": 1900 }, { "epoch": 0.88, "learning_rate": 5.4510556621881e-06, "loss": 0.3877, "step": 2000 }, { "epoch": 0.92, "learning_rate": 3.5508637236084453e-06, "loss": 0.3594, "step": 2100 }, { "epoch": 0.96, "learning_rate": 1.63147792706334e-06, "loss": 0.346, "step": 2200 }, { "epoch": 1.0, "step": 2284, "total_flos": 1.7946134696558592e+16, "train_loss": 0.6671193393433366, "train_runtime": 6110.3906, "train_samples_per_second": 5.981, "train_steps_per_second": 0.374 } ], "max_steps": 2284, "num_train_epochs": 2, "total_flos": 1.7946134696558592e+16, "trial_name": null, "trial_params": null }