Torsten Hoefler π¨π
@thoefler
Professor @ETH_en, head of @spcl_eth, Chief Architect ML @cscsch researching large-scale #HPC and #AI systems and #Climate computing - youtube: https://t.co/dxrzSPIFVb
ID:62059377
http://htor.inf.ethz.ch 01-08-2009 15:54:19
2,8K Tweets
4,6K Followers
211 Following
[1/7] Happy to release π₯QuaRot, a post-training quantization scheme that enables 4-bit inference of LLMs by removing the outlier features.
With Amirkeivan Mohtashami Max Croci Dan Alistarh Torsten Hoefler π¨π James Hensman and others
Paper: arxiv.org/abs/2404.00456
Code: github.com/spcl/QuaRot
![[1/7] Happy to release π₯QuaRot, a post-training quantization scheme that enables 4-bit inference of LLMs by removing the outlier features. With @akmohtashami_a @max_croci @DAlistarh @thoefler @jameshensman and others Paper: arxiv.org/abs/2404.00456 Code: github.com/spcl/QuaRot](https://pbs.twimg.com/media/GKItBQxW4AAh30A.jpg "Saleh Ashkboos (@AshkboosSaleh) on Twitter photo 2024-04-02 05:42:33 [1/7] Happy to release π₯QuaRot, a post-training quantization scheme that enables 4-bit inference of LLMs by removing the outlier features. With @akmohtashami_a @max_croci @DAlistarh @thoefler @jameshensman and others Paper: arxiv.org/abs/2404.00456 Code: github.com/spcl/QuaRot")
In the area of Earth systems science, Peter Bauer, Torsten Hoefler π¨π and colleagues argue that flexible human-in-the-loop interaction is essential for understanding and making efficient use of the data provided by digital twins of Earth. nature.com/articles/s4358β¦
π§΅9/11
Thinking about it - 'RDMA over Converged Ethernet' (RoCE) is really a misnomer and should be called 'InfiniBand over CE'.
RDMA is much wider and there will be a fundamentally different specification in Ultra Ethernet Consortium!
An old discussion on the terms: htor.inf.ethz.ch/blog/index.phpβ¦
