imminsik/summarize_from_feedback_tldr_3_filtered_oai_preprocessing_gpt2_1757513032

该数据集是基于OpenAI的反馈摘要任务构建的，包含了从Reddit获取的帖子的标题、内容、摘要和参考回复。此外，数据集还包含了预处理脚本添加的查询信息、参考回复的标记化版本和长度信息。数据集分为训练集、验证集和测试集，可用于训练和评估自动摘要模型。This dataset is constructed for OpenAIs Summarize from Feedback task, including the title, content, summary, and reference responses of posts taken from Reddit. In addition, the dataset contains query information added by the preprocessing script, tokenized versions of the reference responses, and length information. The dataset is split into training, validation, and test sets for training and evaluating automatic summarization models.