Comment Comment by goncharom Every time I see comments like these I think about this research from anthropic: https://www.anthropic.com/research/mapping-mind-language-mod...LLMs activate similar neurons for similar concepts not only across languages, but also across input types. I’d like to know if you’d consider that as a good representation of “understanding” and if not, how would you define it? parent goncharom 11 months ago
LLMs activate similar neurons for similar concepts not only across languages, but also across input types. I’d like to know if you’d consider that as a good representation of “understanding” and if not, how would you define it?