CohenQu/RLRecipe-guru-0.3-0.6

该数据集是一个包含多种特征的结构化数据集，主要用于自然语言处理任务。数据集包含data_source（数据来源）、prompt（提示内容，包括内容和角色）、ability（能力）、reward_model（奖励模型，包括ground_truth和style）、extra_info（额外信息，包括index、original_question、reward_metric和split）、level（级别）和problem_id（问题ID）等字段。数据集分为训练集（5983个样本）和测试集（3630个样本），总大小为6924434字节，下载大小为14590461字节。

This dataset is a structured dataset containing various features, primarily used for natural language processing tasks. The dataset includes fields such as data_source, prompt (including content and role), ability, reward_model (including ground_truth and style), extra_info (including index, original_question, reward_metric, and split), level, and problem_id. The dataset is divided into a training set (5,983 samples) and a test set (3,630 samples), with a total size of 6,924,434 bytes and a download size of 14,590,461 bytes.