Distill

An open-access, peer-reviewed online journal of machine learning research that treated clear explanation as a research contribution in its own right. Its articles used interactive diagrams and 'explorable explanations', and were reviewed for clarity as well as correctness. It published influential interpretabilityinterpretability, n.: The effort to understand how an AI system arrives at its outputs — what information it uses and how it combines it — instead of treating it as a sealed black box. work, including 'Feature Visualization' (2017), which systematized feature visualizationfeature visualization, n.: Making pictures that show what a part of an image-recognition network is looking for, by repeatedly adjusting an image until that part responds as strongly as possible., 'The Building Blocks of Interpretability' (2018), 'Activation Atlas' (2019), the Circuits thread opened by 'Zoom In' (2020), which introduced circuitscircuit, n.: A small, identifiable piece of wiring inside a neural network — a few internal parts and the connections between them — that together carry out one specific job, such as spotting curves or copying a name., the universality hypothesisuniversality hypothesis, n.: The idea that different AI models trained on similar tasks tend to develop the same internal concepts and mechanisms, so what is learned about one model may carry over to others. and other foundations of mechanistic interpretabilitymechanistic interpretability, n.: A branch of AI research that tries to reverse-engineer a neural network — working out, piece by piece, which internal parts represent which concepts and how they combine to produce behavior., and 'Multimodal Neurons in Artificial Neural Networks' (2021); it also ran explanatory pieces such as 'Why Momentum Really Works' (2017). Its editors put it on hiatus in July 2021, judging that self-publication would serve most such work better; its style continues in venues such as the Transformer Circuits ThreadTransformer Circuits Thread, prop. n.: Anthropic's online publication for research on how language models work inside, presented as long interactive web articles rather than conventional papers..