clembench-playpen/DPO_turn
收藏Hugging Face2025-08-28 更新2025-08-09 收录
下载链接：
https://hf-mirror.com/datasets/clembench-playpen/DPO_turn
下载链接
链接失效反馈官方服务：
资源简介：
这是一个用于DPO Turn策略的训练数据集，来源于PLAYPEN环境的对话游戏反馈。数据集包含了选择的和拒绝的对话延续，这些延续来自于成功的和不成功的对话轮次。数据集以英语为语言。
This is a training dataset for the DPO Turn strategy, derived from dialogue game feedback in the PLAYPEN environment. The dataset includes chosen and rejected dialogue continuations, coming from successful and unsuccessful dialogue turns. The language of the dataset is English.
提供机构：
clembench-playpen
