zlyngkhoi/bbq_unlearning_v2

该数据集包含以下字段：example_id（示例ID）、question_index（问题索引）、question_polarity（问题极性）、context_condition（上下文条件）、category（类别）、answer_info（答案信息，包括ans0、ans1、ans2三个答案）、additional_metadata（附加元数据，包括source、stereotyped_groups、subcategory、version）、context（上下文）、question（问题）、ans0、ans1、ans2（三个答案）、label（标签）、text_input（文本输入）。数据集分为train（训练集）、test（测试集）、retain（保留集）和forget（遗忘集）四个部分，分别包含不同的示例数量和大小。总下载大小为35420274字节，数据集总大小为130037247字节。

The dataset includes the following fields: example_id (example ID), question_index (question index), question_polarity (question polarity), context_condition (context condition), category (category), answer_info (answer information including ans0, ans1, ans2), additional_metadata (additional metadata including source, stereotyped_groups, subcategory, version), context (context), question (question), ans0, ans1, ans2 (three answers), label (label), text_input (text input). The dataset is divided into four parts: train (training set), test (test set), retain (retain set), and forget (forget set), each containing different numbers of examples and sizes. The total download size is 35420274 bytes, and the total dataset size is 130037247 bytes.