Work done with @paulsoulos.bsky.social, @tallinzen.bsky.social, and Paul Smolensky For much more, see our paper! arxiv.org/abs/2608.29530 12/12
AI Weekly's analysis
→
- A new preprint reports that neural networks' vector representations can be closely approximated by closed-form symbolic equations with behavior 'largely unchanged.'
- The substitution was tested on small networks trained to manipulate lists and on LLMs across arithmetic, logic, computer code, and language.
- The authors say the symbolic approximation also lets them steer an LLM's behavior via targeted interventions on its internal representations.
Read full analysis →
