The global neuronal workspace (GNW) is currently the best documented neuroscience mechanism by which conscious processing arises in the human brain — and now Anthropic researchers have discovered a similar workspace inside their large language model !
New Anthropic research: A global workspace in language models. Of everything happening in your brain right now, only a tiny fraction is consciously accessible—thoughts you can describe, hold in mind, and reason with. We found a strikingly similar divide inside Claude.
retweets 81
likes 350
quoted by 1 archive account
substance 4 · opinion · rank 237 · Anthropic discovered a global neuronal workspace mechanism in LLMs similar to human conscious processing.
view on X ↗
1 top reply
"a similar workspace inside their large language model !" Some, well versed in LLM, beg to differ.