ahmedshark/Tokenized_Ar_en_dataset

该数据集包含三个主要特征：input_ids（int32序列）、attention_mask（int8序列）和labels（int64序列）。数据集分为训练集和验证集，训练集包含1,879,992个样本，验证集包含208,889个样本。数据集的下载大小为2,603,109,602字节，总大小为6,324,798,439字节。配置文件中指定了训练集和验证集的数据文件路径。

The dataset contains three main features: input_ids (int32 sequence), attention_mask (int8 sequence), and labels (int64 sequence). The dataset is divided into a training set and a validation set, with the training set containing 1,879,992 samples and the validation set containing 208,889 samples. The download size of the dataset is 2,603,109,602 bytes, and the total size is 6,324,798,439 bytes. The configuration file specifies the data file paths for the training and validation sets.