遇见数据集

selfcorrexp2/llama3_sft_first_wrong_prompt3_generation4

收藏
Hugging Face2024-12-28 更新2025-04-26 收录
官方服务:

资源简介:

该数据集包含五个字段:索引(idx),提示(prompt),答案序列(answers),是否为第一轮(first_round)和真实标签(gt)。数据集被划分为训练集,共有12496个示例,大小为165,323,855字节。数据集配置中提供了训练集的数据文件路径。

The dataset includes five fields: index (idx), prompt, answer sequence (answers), whether its the first round (first_round), and ground truth label (gt). The dataset is split into a training set, which contains 12,496 examples and is 165,323,855 bytes in size. The dataset configuration provides the path to the data files for the training set.

提供机构:
selfcorrexp2
二维码
社区交流群
二维码
科研交流群
商业服务