Autoregressive Language Model on the 6502 Processor
A language model on a 1975 MOS 6502 processor, using a tiny Mamba-based language model and custom 8-bit ternary inference engine, used to generate text on a BBC Micro from the 1980s.
Summary
A language model has been successfully implemented on a vintage 6502 processor, showcasing how a small Mamba-based model and a unique 8-bit inference engine can generate text on an old BBC Micro. This project highlights the intersection of retro computing and modern AI techniques.
Comments (12 of ) ⌛