X · @teortaxesTex
· X / Twitter
Very good overview of Huawei's Tau scaling paper update, addresses some basic retorts. Should be a bigger news than it is; it's not just "3D stacking"…
Very good overview of Huawei's Tau scaling paper update, addresses some basic retorts. Should be a bigger news than it is; it's not just "3D stacking". It's analogous to pretrain-more-for-inference-efficiency route – vastly higher intensity of design per mm^2, but worth it.Macro_Lin ｜ 市场观察员: I've been discussing Huawei