semran1/cosmopedia_4B-tokenized
收藏Hugging Face2025-01-14 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/semran1/cosmopedia_4B-tokenized
下载链接
链接失效反馈官方服务：
资源简介：
这个数据集包含了文本内容及其相应的编码ID、注意力掩码和长度信息。它被划分为一个训练集，共有约429万条样本，总大小约为32.7GB。数据集提供了默认配置，其中包含了训练集文件的路径。
The dataset consists of text content and its corresponding encoded IDs, attention masks, and length information. It is split into a training set with approximately 4,290,000 samples, totaling about 32.7GB in size. The dataset provides a default configuration that includes the file paths for the training set.
提供机构：
semran1
