vinhpx/nvidia-math-dataset
收藏Hugging Face2025-05-29 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/vinhpx/nvidia-math-dataset
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了经过tokenization处理的文本数据，具体特征包括输入ID、注意力掩码和标签。数据集的总大小为744,190,848,660字节，共有13,972,791个示例。提供了一个默认配置，包含了tokenized数据集的文件路径。
The dataset consists of tokenized text data, with features including input IDs, attention masks, and labels. The total size of the dataset is 744,190,848,660 bytes, containing 13,972,791 examples. A default configuration is provided, which includes the file path for the tokenized dataset.
提供机构：
vinhpx
