Scientists suspect that ‘reasoning models don’t always say what they think’
OpenAI says 500 million people use ChatGPT. And, even though it’s exactly as smart as a toaster, everyone seems to think it’s the cat’s meow.
Anthropic bends over backwards calling transformer work ‘thought’ in 2 new papers
In two new papers, Anthropic describes a new methodology by which developers are attempting to peek under the models’ hoods to discern exactly what’s occurring inside.