pile EleutherAI
The Pile is a 825 GiB diverse, open source language modelling data set that consists of 22 smaller, high-quality datasets combined together.
- 種別
- dataset
- ライセンス
- other
- 言語
- en
- ダウンロード
- 2,877
- いいね
- 502
- アクセス
- public
- ファイル
- 0
The Pile is a 825 GiB diverse, open source language modelling data set that consists of 22 smaller, high-quality datasets combined together.