snap-stanford/Qwen_Qwen3-8b_persona_communication_post_dist

该数据集包含对话生成和相关指标的数据，主要特征包括数据集索引、响应键、提示内容（包含内容、名称和角色）、生成文本、额外信息（如评论指标、层次名称、索引、是否顶级、媒体来源、名称、角色、帖子ID、帖子指标、原始提示和分割）以及多种指标（如内容类型、情感、敌意、幽默、情绪和立场）。数据集包含3366个训练样本，总大小为91,648,202字节。

This dataset contains data for dialogue generation and related metrics, with main features including dataset index, response key, prompt content (including content, name, and role), generated text, extra information (such as comment metrics, hierarchy name, index, is_top_level, media source, name, persona, post_id, post metrics, raw prompt, and split), and various metrics (such as content type, emotion, hostility, humor, sentiment, and stance). The dataset includes 3,366 training samples with a total size of 91,648,202 bytes.