I am not garymarcusing here. LLMs are proof that it is possible to distill intelligent reasoning behavior by doing statistics over human language patterns

→ View original post on X — @plinz