Interpretability as a Science: Toward Rigorous Foundations for Understanding LLMs

The workshop asks what constitutes a scientific explanation of a language model. It connects interpretability with hypothesis testing, measurement, causal reasoning and experimental design, seeking ways to distinguish evidence about internal mechanisms from artefacts of an analysis method. Session: 08:00–17:00 (Australia/Sydney). A paid NeurIPS virtual pass includes workshop livestreams; on-site access follows the host-location registration rules.