selfcorrexp2/llama3_sft_balanced_rr60k_train_on_corr_ep3_full_testtmp07_vllmexp
收藏Hugging Face2025-01-07 更新2025-04-26 收录
下载链接：
https://hf-mirror.com/datasets/selfcorrexp2/llama3_sft_balanced_rr60k_train_on_corr_ep3_full_testtmp07_vllmexp
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了索引、提示、答案序列、正确答案、代理标签以及两轮奖励等信息。数据集被划分为训练集，其中包含15000个示例，总大小为39662017字节。
The dataset includes index, prompt, answer sequence, ground truth, proxy label, and rewards for two rounds. The dataset is split into a training set, which contains 15000 examples and totals 39662017 bytes in size.
提供机构：
selfcorrexp2
