Tim Dettmers personal blog

Emergent Features in 6.7B+ Param Transformers

When I attended NAACL, I wanted to do a little test. I had two pitches for my LLM.int8() paper. One pitch is about how I use advanced quantization methods to achieve no performance degradation transformer inference at scale that makes large models more accessible. The other pitch talks about emergent outliers in…
Points
185
HN comments
15
Submissions
1
Cited by
6 commenters

AI & Machine Learning2022

On Hacker News

SubmittedTitle on HNPointsComments
Aug 18, 2022Emergent Features in 6.7B+ Param Transformers by nl18515

Cited in comments

Linked from 6 Hacker News comments by 6 different people, from Dec 2022 to Jan 2024.

Most recent:

Search every comment that links here →

More from Tim Dettmers

All 10 posts →

Related posts from the same era