ai2-adapt-dev/uf_cleaned_25

该数据集包含了对话或提示相关的信息，其中有提示文本(prompt)、提示ID(prompt_id)、选中的内容(chosen)、被拒绝的内容(rejected)、消息(messages)，以及数据来源(source)和是否交换(is_swapped)的标识。数据集分为训练集(train)，提供了训练集的字节数和示例数量。这是一个用于对话或文本生成任务的数据集。

The dataset consists of conversation or prompt-related information, including prompt text (prompt), prompt ID (prompt_id), chosen content (chosen), rejected content (rejected), messages (messages), data source (source), and an indicator of whether it is swapped (is_swapped). The dataset is split into a training set (train), and the byte size and number of examples of the training set are provided. This is a dataset for dialogue or text generation tasks.