selfcorrexp2/HanningZhang_Llama3-sft-more-corr-rr60k-3ep_moredatatmp10_vllmexp3
收藏Hugging Face2025-01-23 更新2025-04-26 收录
下载链接：
https://hf-mirror.com/datasets/selfcorrexp2/HanningZhang_Llama3-sft-more-corr-rr60k-3ep_moredatatmp10_vllmexp3
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个包含索引、提示、奖励序列、答案序列、真实标签、代理标签和二次奖励序列的文本数据集，主要用于训练模型。数据集分为训练集，共有15000个示例。
This dataset includes index, prompt, reward sequence, answer sequence, ground truth label, proxy label, and second reward sequence. It is a text dataset primarily used for training models, with a training set consisting of 15,000 examples.
提供机构：
selfcorrexp2
