Goodfire researcher Nick Cammarata on why LLMs are easier to understand, as peers push back

MIT Technology Review interviewed Goodfire researcher Nick Cammarata, whose viral blog post argued that large language models are easier to understand than previously thought. Other researchers are disputing his conclusions about neural network interpretability.

• Cammarata claimed he has "good news about the nature of neural networks"
• The Goodfire researcher's blog post went viral
• Some researchers disagree with his conclusions on LLM interpretability