News for hackers. Stuff that matters.
Re:Show HN: Speeding up LLM inference 2x times (possibly) by Macuyiko on Thursday April 18, 2024 @08:25AM (#40075468)
Have a look at https://arxiv.org/pdf/2306.11695.pdf which also uses the norm of inputs based on calibration

There are no comments on this comment yet.