Vincent Chow / South China Morning Post:

DeepSeek researchers detail mHC, a new architecture they used to train 3B, 9B, and 27B models, finding it scaled without adding significant computational burden  —  DeepSeek has published a technical paper co-authored by founder Liang Wenfeng proposing a rethink of its core deep learning architecture

Vincent Chow / South China Morning Post:

DeepSeek researchers detail mHC, a new architecture they used to train 3B, 9B, and 27B models, finding it scaled without adding significant computational burden  —  DeepSeek has published a technical paper co-authored by founder Liang Wenfeng proposing a rethink of its core deep learning architecture

Source: TechMeme
Source Link: http://www.techmeme.com/260101/p11#a260101p11