latam-gpt/tulu-3-405b-preference-translated

这个数据集包含了对话提示（prompt）、选中的回答（chosen）、被拒绝的回答（rejected）以及这些回答的相关信息，如评分（chosen_rating和rejected_rating）、模型名称（chosen_model和rejected_model）和数据来源（source）。数据集被划分为训练集，大小为4157467933字节，共有135128个示例。数据集的下载大小为952937228字节。

This dataset includes conversation prompts (prompt), chosen responses (chosen), rejected responses (rejected), and related information for these responses, such as ratings (chosen_rating and rejected_rating), model names (chosen_model and rejected_model), and data sources (source). The dataset is split into a training set, which is 4157467933 bytes in size and contains 135128 examples. The download size of the dataset is 952937228 bytes.