qwselfcorr/augmath_turn1_wrong_gen_processed
收藏Hugging Face2025-01-31 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/qwselfcorr/augmath_turn1_wrong_gen_processed
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了会话信息，每个样本包括会话ID、真实标签、是否为第一个回复的标记、第二轮回复的奖励、标志位、会话轮数以及具体的会话内容（包括发言内容和角色）。数据集被划分为训练集，训练集大小为284848320字节，包含42023个样本。
The dataset contains conversational information, with each sample including conversation ID, ground truth label, a boolean indicating whether it is the first reply, rewards for the second reply, a flag, the turn of the conversation, and the specific content of the conversation (including the content and the role of the speaker). The dataset is split into a training set, which is 284848320 bytes in size and contains 42023 examples.
提供机构：
qwselfcorr
