kanishka/babylm2-rewritten-clean_hierarchical-adj_size-origin_both-ablation
收藏Hugging Face2025-10-25 更新2025-11-15 收录
下载链接：
https://hf-mirror.com/datasets/kanishka/babylm2-rewritten-clean_hierarchical-adj_size-origin_both-ablation
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含一个文本特征，由训练集和验证集组成。训练集包含11,998,454个示例，大小为522,128,993字节；验证集包含1,223,928个示例，大小为55,417,030字节。整个数据集的大小为577,546,023字节。
The dataset includes a text feature and consists of a training set and a validation set. The training set has 11,998,454 examples with a size of 522,128,993 bytes, and the validation set has 1,223,928 examples with a size of 55,417,030 bytes. The total size of the dataset is 577,546,023 bytes.
提供机构：
kanishka
