raftstudy/dpo_exp_llama3_v2_sftmodel_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_sftmodel_train_20k
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含了选中的(chosen)和被拒绝的(rejected)两个类别的内容和角色信息,同时为每个类别提供了特征向量。训练集包含20000个示例,数据集总大小为1,395,657,975字节。
The dataset includes content and role information for chosen and rejected categories, along with feature vectors for each category. The training set contains 20,000 examples, and the total size of the dataset is 1,395,657,975 bytes.
提供机构:
raftstudy



