richard-park/polytech-2.8M-exclude-trans

该数据集包含多个字段，包括指令（instruction）、输入（input）、输出（output）、系统（system）和文本（text）。数据集分为训练集（train）和验证集（valid），其中训练集包含2,574,112个示例，验证集包含291,523个示例。数据集的总大小为7,659,116,514字节，下载大小为2,419,145,673字节。这些字段和分割表明数据集可能用于训练和验证自然语言处理模型，特别是那些需要理解和生成文本的模型。

The dataset includes multiple fields such as instruction, input, output, system, and text. It is divided into a training set (train) and a validation set (valid), with the training set containing 2,574,112 examples and the validation set containing 291,523 examples. The total size of the dataset is 7,659,116,514 bytes, with a download size of 2,419,145,673 bytes. These fields and splits suggest that the dataset is likely used for training and validating natural language processing models, particularly those that require understanding and generating text.