starlife/m1k_diverse

这个数据集包含了问题(question)、解决方案(solution)、问题类型(cot_type)、来源类型(source_type)、元数据(metadata，包括类别(category)和标识符(id))、思考轨迹(thinking_trajectories)和尝试(attempt)等字段。数据集被划分为训练集(train)，共有1000个示例，总大小为17,525,695字节。

The dataset includes fields for question, solution, problem type (cot_type), source type (source_type), metadata (including category and identifier), thinking trajectories, and attempts. The dataset is split into a training set (train) with a total of 1,000 examples and a size of 17,525,695 bytes.