jaeyong2/Ko-emb-PreView
收藏Hugging Face2024-12-04 更新2024-12-14 收录
下载链接:
https://hf-mirror.com/datasets/jaeyong2/Ko-emb-PreView
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含三个特征:context、Title和Fake Title,分别表示上下文、标题和假标题。数据集分为训练集和测试集,训练集包含223,849个样本,测试集包含24,873个样本。数据集的下载大小为496,958,568字节,数据集大小为1,149,522,701字节。数据集的许可证为Apache-2.0,语言为韩语。开发过程中,数据集来源于daje/ko_wiki和maywell/korean_textbooks,并使用Qwen/Qwen2-72B-Instruct模型生成答案。
This is a Korean text dataset containing context, title, and fake title. The dataset is divided into training and test sets, containing 223849 and 24873 samples respectively. The dataset is built based on the daje/ko_wiki and maywell/korean_textbooks source datasets and uses the Qwen/Qwen2-72B-Instruct model to generate answers.
提供机构:
jaeyong2



