ChaiML/DPO_experiment_triplets_alignment_v1
收藏Hugging Face2024-07-04 更新2024-07-06 收录
下载链接:
https://hf-mirror.com/datasets/ChaiML/DPO_experiment_triplets_alignment_v1
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含对话数据,每个对话包括对话内容、选择的完成内容、拒绝的完成内容以及对齐分数。数据集分为训练集,包含4122个示例,总大小为28430284字节。
This dataset contains conversation data, each including conversation content, chosen completion content, rejected completion content, and alignment scores. The dataset is divided into a training set, containing 4122 examples, with a total size of 28430284 bytes.
提供机构:
ChaiML



