selfcorrexp2/llama31_14w_wrong_18w_corr_20k_test
收藏Hugging Face2024-12-26 更新2025-04-26 收录
下载链接：
https://hf-mirror.com/datasets/selfcorrexp2/llama31_14w_wrong_18w_corr_20k_test
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对话或问答信息的数据集，具体包含字段如索引、是否为第一轮、正确答案、奖励、个人解答、标记、轮数、对话内容（包括内容和角色）、问题、正确答案上下文、难度等级、类型、解决方案、预测答案和代码。数据集被分为训练集，共20000个示例。
This dataset is a collection of dialogue or question-and-answer information, including fields such as index, whether it is the first round, correct answer, reward, personal solution, flag, turn, conversation content (including content and role), question, correct answer context, difficulty level, type, solution, predicted answer, and code. The dataset is split into a training set with a total of 20000 examples.
提供机构：
selfcorrexp2
