nate-rahn/wildchat-reattributed-query-unique

该数据集包含了用户提示（original_prompt）、自定义ID（original_custom_id）、处理流程类型（pipeline_type）、所有属性（all_attributes）、抽取属性（sampled_attributes）、可用属性数量（num_attributes_available）、抽取属性数量（num_attributes_sampled）、类别（category）、重构查询（reconstructed_query）以及消息内容（messages）和角色（role）。数据集分为训练集（train），训练集包含约1398556个示例，大小为4650941973字节。

The dataset includes user prompts (original_prompt), custom IDs (original_custom_id), pipeline types (pipeline_type), all attributes (all_attributes), sampled attributes (sampled_attributes), number of available attributes (num_attributes_available), number of sampled attributes (num_attributes_sampled), categories (category), reconstructed queries (reconstructed_query), and message content (messages) with roles (role). The dataset is split into a training set (train) which contains approximately 1,398,556 examples and is 4,650,941,973 bytes in size.