mlfoundations-dev/base_scale_up_science

该数据集包含了文本、类别、URL、标题、问题、推理、解决方案、推理轨迹、对话、学科、专家、主题数量、子主题数量、问题数量、主题、子主题、得分和年份等字段的信息。数据集分为训练集等不同分割，并提供了详细的大小信息。具体的应用场景和详细描述在README文件中未提供。

The dataset includes information fields such as text, category, URL, title, problem, reasoning, solution, reasoning trace, conversation, discipline, expert, number of topics, number of subtopics, number of questions, topic, subtopic, score, and year. The dataset is split into different parts such as the training set, and detailed size information is provided. The specific application scenario and detailed description are not provided in the README file.