weepcat/compute_rewards_hh_partial_reward_model_token_by_token
收藏Hugging Face2025-01-02 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/weepcat/compute_rewards_hh_partial_reward_model_token_by_token
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了选中的(chosen)和拒绝的(rejected)对话内容及其角色信息，并提供了多种奖励模型的评分数据。数据集仅包含训练集，共有291371个示例。
The dataset includes selected (chosen) and rejected (rejected) conversation content and their role information, and provides reward scores from various reward models. The dataset contains only a training set with a total of 291371 examples.
提供机构：
weepcat
