teamcore/DPO_Pm3B_RMAB_TG_clean_beta0.25dpo_proEurus_RM_7bbt_noise_flip_paper0.3_nu0.008
收藏数据链接：
官方服务：
资源简介：
这是一个名为CR的数据集，包含任务类型、选择、拒绝等信息，以及与之相关的分布和决策概率。数据集还包含任务类别、是否为原始任务、提示和响应等字段。数据集分为默认分片，共有2000个示例。
This is a dataset named CR, which includes information such as task type, selection, rejection, and related distributions and decision probabilities. The dataset also contains fields like task category, whether it is an original task, prompt, and response. The dataset is divided into a default split with a total of 2000 examples.
提供机构：
teamcore
