Dr4kl3s/hellaswag_coregrid_seed17
收藏Hugging Face2025-12-11 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/Dr4kl3s/hellaswag_coregrid_seed17
下载链接
链接失效反馈官方服务:
资源简介:
该数据集包含多种语言(阿拉伯语、孟加拉语、德语、英语、西班牙语、法语、印地语、印尼语、意大利语、葡萄牙语、中文)的配置,每个配置包含训练集和测试集。数据集的特征包括上下文部分(ctx_a, ctx_b, ctx)、多个结尾(endings)、ID、索引、活动标签、来源ID、分割类型和标签。数据集的结构表明它可能用于自然语言处理任务,如文本生成或选择,但具体用途未明确说明。
The dataset includes configurations for multiple languages (Arabic, Bengali, German, English, Spanish, French, Hindi, Indonesian, Italian, Portuguese, Chinese), each with train and test splits. Features include context parts (ctx_a, ctx_b, ctx), multiple endings, IDs, indices, activity labels, source IDs, split types, and labels. The structure suggests potential use in NLP tasks like text generation or selection, but the exact purpose is not explicitly stated.
提供机构:
Dr4kl3s



