koajoel/DAPO-Math-17k-dedup

该数据集包含了一系列的特征字段，用于描述某种交互或对话场景。具体包括数据来源、交互提示（包括内容和角色）、能力、奖励模型（包括真实值和风格）以及额外信息（如索引）。数据集被划分为训练集，共有17917个样本，文件大小为11132384字节。同时，提供了默认配置，用于指定训练集的数据文件。

The dataset consists of a set of feature fields that describe some kind of interactive or conversational scenario. It includes data source, interaction prompts (including content and role), ability, reward model (including ground truth and style), and extra information (such as index). The dataset is split into a training set with a total of 17917 samples, with a file size of 11132384 bytes. Additionally, a default configuration is provided to specify the data files for the training set.