What do LLMs Really Learn About Model Weights? A Stratification Approach to Examine Weight Space Programming
TLDRI attempted to replicate “Sparks of Weight-Space Programming” by Takehi et al (2026) with stratification on two reasoning models, GPT 6 Astra and Deepseek-V4.1-Flash. The task for the LLMs is to identify the game a ML model is trained for giving the weight of the model (using exact same prompt and weight defined by Takehi). Although both of these models do well on the surface (4/4 correct for Astra and 3/4 correct for Flash), the stratification approach exposes the difference in mechanism be...
Read full article →