kanishka/babylm2-clean-spacy

该数据集包含一个名为text的字符串类型特征，分为训练集和验证集。训练集包含12,037,215个样本，验证集包含1,227,879个样本。数据集的下载大小为360,176,595字节，总大小为607,352,068字节。数据文件路径在配置信息中指定。

The dataset contains a feature named text of string type, divided into training and validation sets. The training set includes 12,037,215 samples, and the validation set includes 1,227,879 samples. The download size of the dataset is 360,176,595 bytes, and the total size is 607,352,068 bytes. The data file paths are specified in the configuration information.