_plap retweeted
Quadratic attention has been indispensable for information-dense modalities such as language... until now.
Announcing Mamba: a new SSM arch. that has linear-time scaling, ultra long context, and most importantly--outperforms Transformers everywhere we've tried.
With @tri_dao 1/
_plap retweeted
If science serves truth, how come scientists don't care about epistemology enough to at least understand the implications of Gödel's proof? Science is as cringe as modern Christianity ("Oh, so God does not interact with physics? Ok, fine, let's pray to the rest of it!").
_plap retweeted
I wish more people talked in terms of distributions of outcomes. So much discourse is around “facts” as sound bites, only occasionally quantified, and even more rarely discussed as probabilities. Lots of people can have an intuitive notion of the full spectrum of possibilities \
_plap retweeted
It’s been a great pleasure to help Tesla towards its goals over the last 5 years and a difficult decision to part ways. In that time, Autopilot graduated from lane keeping to city streets and I look forward to seeing the exceptionally strong Autopilot team continue that momentum.
_plap retweeted
Yeah you forgot the most effective option, @Battlefield. "Buy an SSD." Check. GG