HN
Hacker News
Top
New
Best
Ask
Show
Jobs
Comment by olq_plo | Hacker News Reader
Full thread
olq_plo
·
Very cool idea. Can't wait for converted models on HF.
View on HN
mm321
·
deepseek-v2,v3,r1 are all using multi-headed attention.
Reply on news.ycombinator.com