{ "best_metric": null, "best_model_checkpoint": null, "epoch": 0.40361911809222695, "eval_steps": 500, "global_step": 3000, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.01, "learning_rate": 4.9997567688496474e-05, "loss": 1.6283, "step": 100 }, { "epoch": 0.03, "learning_rate": 4.9990172715142793e-05, "loss": 0.068, "step": 200 }, { "epoch": 0.04, "learning_rate": 4.997781629993153e-05, "loss": 0.0404, "step": 300 }, { "epoch": 0.05, "learning_rate": 4.9960500896052476e-05, "loss": 0.0626, "step": 400 }, { "epoch": 0.07, "learning_rate": 4.993822994123172e-05, "loss": 0.0428, "step": 500 }, { "epoch": 0.08, "learning_rate": 4.9911007857049264e-05, "loss": 0.0265, "step": 600 }, { "epoch": 0.09, "learning_rate": 4.987884004806111e-05, "loss": 0.0263, "step": 700 }, { "epoch": 0.11, "learning_rate": 4.984173290072626e-05, "loss": 0.03, "step": 800 }, { "epoch": 0.12, "learning_rate": 4.979969378213884e-05, "loss": 0.0253, "step": 900 }, { "epoch": 0.13, "learning_rate": 4.975273103856537e-05, "loss": 0.038, "step": 1000 }, { "epoch": 0.15, "learning_rate": 4.970085399378785e-05, "loss": 0.0322, "step": 1100 }, { "epoch": 0.16, "learning_rate": 4.964407294725254e-05, "loss": 0.0296, "step": 1200 }, { "epoch": 0.17, "learning_rate": 4.958239917202523e-05, "loss": 0.0295, "step": 1300 }, { "epoch": 0.19, "learning_rate": 4.9515844912553106e-05, "loss": 0.0171, "step": 1400 }, { "epoch": 0.2, "learning_rate": 4.944442338223378e-05, "loss": 0.0205, "step": 1500 }, { "epoch": 0.22, "learning_rate": 4.9368148760792e-05, "loss": 0.029, "step": 1600 }, { "epoch": 0.23, "learning_rate": 4.9287036191464414e-05, "loss": 0.0172, "step": 1700 }, { "epoch": 0.24, "learning_rate": 4.9201101777993116e-05, "loss": 0.0235, "step": 1800 }, { "epoch": 0.26, "learning_rate": 4.911036258142853e-05, "loss": 0.0168, "step": 1900 }, { "epoch": 0.27, "learning_rate": 4.9014836616742065e-05, "loss": 0.0227, "step": 2000 }, { "epoch": 0.28, "learning_rate": 4.891454284924965e-05, "loss": 0.0227, "step": 2100 }, { "epoch": 0.3, "learning_rate": 4.8810575041794e-05, "loss": 0.0173, "step": 2200 }, { "epoch": 0.31, "learning_rate": 4.8700853511049656e-05, "loss": 0.0227, "step": 2300 }, { "epoch": 0.32, "learning_rate": 4.858642651436035e-05, "loss": 0.0144, "step": 2400 }, { "epoch": 0.34, "learning_rate": 4.846731676957191e-05, "loss": 0.03, "step": 2500 }, { "epoch": 0.35, "learning_rate": 4.834354792422293e-05, "loss": 0.0189, "step": 2600 }, { "epoch": 0.36, "learning_rate": 4.821514455084985e-05, "loss": 0.0111, "step": 2700 }, { "epoch": 0.38, "learning_rate": 4.8082132142108465e-05, "loss": 0.0152, "step": 2800 }, { "epoch": 0.39, "learning_rate": 4.794453710571272e-05, "loss": 0.0182, "step": 2900 }, { "epoch": 0.4, "learning_rate": 4.780238675919182e-05, "loss": 0.0111, "step": 3000 } ], "logging_steps": 100, "max_steps": 22296, "num_input_tokens_seen": 0, "num_train_epochs": 3, "save_steps": 3000, "total_flos": 8.395428848876913e+17, "train_batch_size": 4, "trial_name": null, "trial_params": null }