imminsik/summarize_from_feedback_tldr_3_filtered_oai_preprocessing_gpt2_1757478697
收藏数据链接：
官方服务：
资源简介：
这是一个用于OpenAI的Summarize from Feedback任务的Reddit TL;DR数据集。数据集包含了帖子的唯一标识符、所属subreddit、标题、帖子内容、摘要以及参考回复。另外，还包括了预处理脚本添加的查询相关字段，如查询文本、查询的标记化版本、参考回复的标记化版本等。数据集分为训练集、验证集和测试集。
This is the Reddit TL;DR dataset for OpenAIs Summarize from Feedback task. The dataset includes unique identifiers for posts, the subreddit each post belongs to, the title of the post, the body of the post, the summary, and the reference response. Additionally, it includes fields added by the preprocessing script related to the query, such as the query text, tokenized version of the query, tokenized version of the reference response, etc. The dataset is split into training, validation, and test sets.
提供机构：
imminsik
