IParraMartin/OpenNeuro
收藏Hugging Face2025-03-22 更新2025-04-12 收录
下载链接:
https://hf-mirror.com/datasets/IParraMartin/OpenNeuro
下载链接
链接失效反馈官方服务:
资源简介:
OpenNeuro数据集是用于计算脑评分缩放规律的,包含了用于训练20个语言模型的文本分割。这些分割提供了不同规模 progressively larger 的文本语料库,以便在不同的尺度上进行系统性实验。数据集包括NanoText、MiniText、MidiText、CoreText和MegaText五个子集,分别对应不同大小的文本语料库。
The OpenNeuro dataset is used to compute brain score scaling laws, containing text splits for training 20 language models. These splits provide a progressively larger corpus of text for systematic experimentation at different scales. The dataset includes five subsets: NanoText, MiniText, MidiText, CoreText, and MegaText, each corresponding to a different size of text corpus.
提供机构:
IParraMartin



