LangAGI-Lab/qwen-7b-instruct-8k-dpo-preference-set
收藏Hugging Face2025-02-08 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/LangAGI-Lab/qwen-7b-instruct-8k-dpo-preference-set
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了指令、答案以及用户选择的响应信息，其中用户选择的响应包括是否正确、响应内容和响应的token长度。同时，还包括了被拒绝的响应信息。数据集分为训练集，其大小为21,805,800字节，共有8,162个示例。
The dataset includes instructions, answers, and user-selected responses, where the users selected response consists of correctness, response content, and token length. It also contains information on rejected responses. The dataset is split into a training set, which is 21,805,800 bytes in size and contains 8,162 examples.
提供机构：
LangAGI-Lab
