yuancarrieyjy/tulu3_infinite-chats-taxonomy_mix_10k_no_thinking

该数据集包含多个特征字段，包括数据来源（data_source）、提示信息（prompt，包含内容和角色）、奖励模型（reward_model，包含真实值和风格）、类型（type）以及额外信息（extra_info，包含问题、参考生成的响应列表、参考最大分数、参考平均分数和参考最小分数）。数据集分为训练集（10347个样本）和验证集（2587个样本），总下载大小约为107.58MB，数据集总大小约为108.10MB。

The dataset includes multiple feature fields such as data source (data_source), prompts (prompt, containing content and role), reward model (reward_model, containing ground truth and style), type (type), and extra information (extra_info, containing question, reference generated responses list, reference max score, reference mean score, and reference min score). The dataset is divided into a training set (10,347 samples) and a validation set (2,587 samples), with a total download size of approximately 107.58MB and a total dataset size of approximately 108.10MB.