@karpathy I suggest a minor alternation on this. Keep your LLM wiki in md, just create thin HTML viewer to navigate it, so you can stay clean-memory, skipping the HTML layer. MD is still more compact if done right. Keeping each md file under 20k tokens each etc
I've always been passionate about games and they've played a big part in @GoogleDeepMind’s history, as the perfect proving ground for AI. Thrilled to announce this research partnership with @FenrisCreations - @EveOnline is one of the most extraordinary games ever built and has an amazing community. Very excited to work with @HilmarVeigar and the team!