Opening the AI black box: program synthesis via mechanistic interpretability

1 code implementation • 7 Feb 2024 • Eric J. Michaud, Isaac Liao, Vedang Lad, Ziming Liu, Anish Mudide, Chloe Loughridge, Zifan Carl Guo, Tara Rezaei Kheirkhah, Mateja Vukelić, Max Tegmark

We present MIPS, a novel method for program synthesis based on automated mechanistic interpretability of neural networks trained to perform the desired task, auto-distilling the learned algorithm into Python code.

Program Synthesis Symbolic Regression

Universal Neurons in GPT2 Language Models

1 code implementation • 22 Jan 2024 • Wes Gurnee, Theo Horsley, Zifan Carl Guo, Tara Rezaei Kheirkhah, Qinyi Sun, Will Hathaway, Neel Nanda, Dimitris Bertsimas

In other words, are neural mechanisms universal across different models?