uplimit/uplimit-synthetic-data-week-2-filtered

该数据集包含多个字符串类型的特征，如系统信息、输入、选择、拒绝、生成内容、顺序、标注模型、标注提示内容、原始标注响应、评分、评分依据、状态、原始选择、原始拒绝、选择得分、是否在GSM8K训练集中、可读性分数以及领域信息。数据集分为训练集，共有1468个示例，大小为19070396字节。提供默认配置，训练数据文件路径为data/train-*。

The dataset contains multiple string features such as system information, input, chosen, rejected, generated content, order, labeling model, labeling prompt content, original labeling response, rating, rationale, status, original chosen, original rejected, chosen score, whether in the GSM8K training set, readability score, and domain information. The dataset is split into a training set with a total of 1468 examples, with a size of 19070396 bytes. It provides a default configuration, and the path to the training data files is data/train-*.