hyungjikim/wikitext-tags-deberta-v3

该数据集包含输入ID、标记类型ID、注意力掩码、终结符ID、非终结符ID、依赖关系ID、头依赖ID和分类ID等多个字段。数据集分为训练集和验证集两部分，其中训练集包含约373万个示例，大小约为88.07GB；验证集包含约3.77万个示例，大小约为0.89GB。总下载大小约为714MB，总数据集大小约为88.96GB。

The dataset includes fields such as input IDs, token type IDs, attention masks, terminal IDs, nonterminal IDs, dependency relation IDs, head dependency IDs, and classification IDs. The dataset is split into a training set and a validation set, with the training set containing approximately 3.73 million examples and being about 88.07GB in size; the validation set contains approximately 37,700 examples and is about 0.89GB in size. The total download size is about 714MB, and the total dataset size is about 88.96GB.