raftstudy/dpo_exp_llama3_v2_dpo_iter1_train_20k
收藏Hugging Face2025-04-03 更新2025-04-12 收录
下载链接：
https://hf-mirror.com/datasets/raftstudy/dpo_exp_llama3_v2_dpo_iter1_train_20k
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含选中的(chosen)和被拒绝的(rejected)两个类别，每个类别下有内容(content)和角色(role)两个字段。此外，还有对应的特征向量(chosen_feature和rejected_feature)。数据集划分为训练集，共有20000个示例。
The dataset includes two categories: chosen and rejected, each with content and role fields. In addition, there are corresponding feature vectors (chosen_feature and rejected_feature). The dataset is split into a training set with a total of 20,000 examples.
提供机构：
raftstudy
