dogtooth/uf_tulu_2_uf_iter2_2

该数据集包含了对话场景中的提示（prompt）、消息内容（messages.content）、消息角色（messages.role）、模型完成的内容（model_completion）以及参考完成的内容（reference_completion）。数据集被划分为训练集，其中包含122,270个示例，总大小为687,519,266字节。

The dataset includes conversation scenarios with prompts (prompt), message content (messages.content), message roles (messages.role), content completed by the model (model_completion), and reference completion content (reference_completion). The dataset is split into a training set, which contains 122,270 examples and has a total size of 687,519,266 bytes.