daniazie/space-data

该数据集包含用于训练专注于空间科学和天文学的语言模型的提示。数据集的目标是确保模型能够提供深入、详细和系统的解释，同时避免从提取的文本中引入噪声，并保持写作风格的清晰和彻底。数据集通过distilabel工具生成，包含instruction、distilabel_metadata、model_name和response等字段，展示了模型如何生成和响应空间科学相关的提示。

This dataset contains prompts for training a language model focused on space science and astronomy. The goal of the dataset is to ensure the model can provide deep, detailed, and systematic explanations while avoiding noise from the extracted text and keeping the writing style clean but thorough. The dataset is generated using the distilabel tool and includes fields such as instruction, distilabel_metadata, model_name, and response, demonstrating how the model generates and responds to space science-related prompts.