[BinRoo]

> GPT-4 is actually a pile of 3.5s

I understand the intension and reference you're making. I bet the implementation of GPT-4 is probably something along those lines. However, spreading speculation in definitive language like that when the truth is unknown is dishonest, wouldn't you agree?

[trash_cat]

Sure, I could it put it less definitively, but realistically, what else can it be? The transformer won't change much and all of the models, at the core use it. It's a closely guarded secret because it's easy to replicate.