I don't think LLMs are that chaotic, you can replace words in an input at get a similar answer, and they are very good at dealing with typos.
They are definitely not interpretable, I was reading some stuff from mechanistic interpretability researchers saying they've given up trying to build a bottom up model of how they work.
They are definitely not interpretable, I was reading some stuff from mechanistic interpretability researchers saying they've given up trying to build a bottom up model of how they work.