violetxi/NuminaMath-TIR-Filtered

该数据集包含多个字段，包括问题（problem）、解决方案（solution）、消息（messages）、答案（answer）、步骤令牌计数（step_token_counts）和代码步骤（code_steps）。消息字段是一个列表，包含内容（content）和角色（role）。数据集分为训练集和测试集，训练集包含59,575个示例，测试集包含80个示例。数据集的下载大小为112,828,839字节，总大小为272,025,393.20337975字节。

This dataset includes multiple fields such as problem, solution, messages, answer, step_token_counts, and code_steps. The messages field is a list containing content and role. The dataset is divided into a training set and a test set, with the training set containing 59,575 examples and the test set containing 80 examples. The download size of the dataset is 112,828,839 bytes, and the total size is 272,025,393.20337975 bytes.