MoeReward/combined_preference_dataset_ood_alpaca_heavy
收藏Hugging Face2025-03-31 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/MoeReward/combined_preference_dataset_ood_alpaca_heavy
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含两个部分：选中的(chosen)和拒绝的(rejected)。每个部分都由内容(content)和角色(role)组成，共有3000个训练样本。数据集未提供详细描述，但根据特征和分割，这可能是一个文本分类或角色标注的数据集。
The dataset consists of two parts: chosen and rejected. Each part includes content and role, with a total of 3000 training samples. No detailed description is provided for the dataset, but based on the features and splits, this might be a text classification or role labeling dataset.
提供机构：
MoeReward
