X · @teortaxesTex
· X / Twitter
Really important The pool of actors who can train a "10T Mythos" is much larger than you might have been led to believe. Total param scale is no bottl…
Really importantThe pool of actors who can train a "10T Mythos" is much larger than you might have been led to believe. Total param scale is no bottleneck at all.Alexander Doria: So K3, with one simple question: how do you even train a 3T parameters model?Even though the paper is not out, there is enough public informa