JungHun/Efficient_ToolCalling_train
收藏Hugging Face2025-12-12 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/JungHun/Efficient_ToolCalling_train
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含训练数据，主要用于训练模型区分‘chosen’（选中）和‘rejected’（拒绝）的响应。每个样本包括‘chosen’和‘rejected’两个部分，每部分包含‘content’（内容）和‘role’（角色）字段。此外，每个样本还包含一个‘prompt’（提示）字段和一个‘metadata’（元数据）字段。数据集仅包含一个‘train’（训练）分割，共有84,843个样本，总大小为1,601,301,068字节。
This dataset contains training data primarily designed to train models in distinguishing between chosen and rejected responses. Each sample includes chosen and rejected sections, with each containing content and role fields. Additionally, each sample has a prompt field and a metadata field. The dataset consists of a single train split with 84,843 samples, totaling 1,601,301,068 bytes.
提供机构：
JungHun
