Recovering meaningful concepts from language model activations is a central aim of interpretability. While existing feature extraction methods aim to identify concepts that are independent directions,...

Interpretability provides a toolset for understanding how and why neural networks behave in certain ways. However, there is little unity in the field: most studies employ ad-hoc evaluations and do not...

How do language models track mental states of each character in a story, often referred to as Theory of Mind? We reverse-engineered how LLaMA-3-70B-Instruct handles a belief-tracking task and found something surprising: it uses mechanisms strikingly similar to pointer variables in C programming!

How can we know whether new mechanistic interpretability methods achieve real improvements? In pursuit of meaningful and lasting evaluation standards, we propose MIB, a benchmark with two tracks spann...