mlfoundations-dev/scale_up_science_50K

该数据集包含了文本内容、类别、URL、标题、问题、推理、解决方案、推理轨迹、对话、学科、专家、话题数量、子话题数量、问题数量、话题、子话题、得分、年份和ID等字段。数据集被分为训练集，提供了字节数和示例数的统计信息，以及下载大小和总大小。具体的数据集内容和用途没有在README中描述。

The dataset includes fields such as text content, category, URL, title, problem, reasoning, solution, reasoning trace, conversation, discipline, expert, number of topics, number of subtopics, number of questions, topic, subtopic, score, year, and ID. The dataset is split into a training set and provides statistics on the number of bytes and examples, as well as the download size and total size. The specific content and purpose of the dataset are not described in the README.