Uniform manifold approximation and projection is a nonlinear dimension reduction method
often used for visualizing data and as pre-processing for further machine-learning tasks …

Understanding AI systems' inner workings is critical for ensuring value alignment and safety.
This review explores mechanistic interpretability: reverse engineering the computational …

Conversational large language models are fine-tuned for both instruction-following and
safety, resulting in models that obey benign requests but refuse harmful ones. While this …

Sparse autoencoders provide a promising unsupervised approach for extracting
interpretable features from a language model by reconstructing activations from a sparse …

Sparse autoencoders (SAEs) are an unsupervised method for learning a sparse
decomposition of a neural network's latent representations into seemingly interpretable …

ABSTRACT Large Language Models (LLMs) are rapidly surpassing human knowledge in
many domains. While improving these models traditionally relies on costly human data …

The rapid development of large language models (LLMs) has opened new avenues across
various fields, including cybersecurity, which faces an evolving threat landscape and …

Disentangling model activations into meaningful features is a central problem in
interpretability. However, the absence of ground-truth for these features in realistic scenarios …

A key goal in mechanistic interpretability is circuit analysis: finding sparse subgraphs of
models corresponding to specific behaviors or capabilities. However, MLP sublayers make …

Existing approaches to estimating politicians' latent positions along specific dimensions
often fail when relevant data is limited. We leverage the embedded knowledge in generative …