rshwndsz/nectar-cleaned-r5-binarized

该数据集包含了用于训练的文本数据，其中包括提示文本(prompt)、唯一标识符(nectar_id和binary_id)、两个完成答案(completion_a和completion_b，每个答案包括答案文本和模型名称)以及一个排名信息(nectar_rank，包括答案文本、模型名称和排名)。数据集仅包含训练集(train split)，共有1,829,520个示例，大小为7,247,915,524字节。The dataset includes text data for training, which contains prompt text, unique identifiers (nectar_id and binary_id), two completion answers (completion_a and completion_b, each answer includes answer text and model name) and a ranking information (nectar_rank, including answer text, model name and rank). The dataset contains only the training set (train split) with a total of 1,829,520 examples and a size of 7,247,915,524 bytes.