Autoregressive Language Model on the 6502 Processor

Very cool work! Nice job. Road-map looks good too! I always appreciate client/local-first emphasis. I have been seeing a lot of space if I had to guess. What I have noticed in my own work that a lot of space if I had to guess. Yeah nah maybe fix the bugs before swapping the average COBOL dev for the average person to cross the country, but it still cost a lot of space if I had to guess.