Related papers: AbsTopK: Rethinking Sparse Autoencoders For Bidire…

Modern text classification methods heavily rely on contextual embeddings from large language models (LLMs). Compared to human-engineered features, these embeddings provide automatic and effective representations for classification model…

We investigate whether sparse autoencoders (SAEs) can be used to remove knowledge from language models. We use the biology subset of the Weapons of Mass Destruction Proxy dataset and test on the gemma-2b-it and gemma-2-2b-it language…

Sparse autoencoders (SAEs) have shown promise in extracting interpretable features from complex neural networks. We present one of the first applications of SAEs to dense text embeddings from large language models, demonstrating their…

Large-scale pre-trained vision-language models like CLIP demonstrate remarkable zero-shot performance across diverse tasks. However, fine-tuning these models to improve downstream performance often degrades robustness against distribution…

Large language models (LLMs) store vast amounts of information, making them powerful yet raising privacy and safety concerns when selective knowledge removal is required. Existing unlearning strategies, ranging from gradient-based…

Sparse Autoencoders (SAEs) are powerful tools for interpreting neural representations, yet their use in audio remains underexplored. We train SAEs across all encoder layers of Whisper and HuBERT, provide an extensive evaluation of their…

Sparse autoencoders are usually trained one layer at a time, even though transformer residual stream activations are strongly coupled across depth. This creates a practical problem for multi-layer interventions: different layerwise…

With the integration of image modality, the semantic space of multimodal large language models (MLLMs) is more complex than text-only models, making their interpretability more challenging and their alignment less stable, particularly…

Although Multimodal Large Language Models (MLLMs) have advanced substantially, they remain vulnerable to object hallucination caused by language priors and visual information loss. To address this, we propose SAVE (Sparse Autoencoder-Driven…

EEG foundation models achieve state-of-the-art clinical performance, yet the internal computations driving their predictions remain opaque: a barrier to clinical trust. We apply TopK Sparse Autoencoders (SAEs) across three architecturally…

Large Language Models (LLMs) remain vulnerable to optimization-based jailbreak attacks that exploit internal gradient structure. While Sparse Autoencoders (SAEs) are widely used for interpretability, their robustness implications remain…

The evaluation of large language models (LLMs) relies heavily on standardized benchmarks. These benchmarks provide useful aggregated metrics for a given capability, but those aggregated metrics can obscure (i) particular sub-areas where the…

A common goal of mechanistic interpretability is to decompose the activations of neural networks into features: interpretable properties of the input computed by the model. Sparse autoencoders (SAEs) are a popular method for finding these…

Sparse autoencoders (SAEs) are a popular technique for interpreting language model activations, and there is extensive recent work on improving SAE effectiveness. However, most prior work evaluates progress using unsupervised proxy metrics…

Sparse autoencoders (SAEs) are used to disentangle the dense, polysemantic internal representations of large language models (LLMs) into interpretable, monosemantic concepts. However, standard $\ell_1$-regularized SAEs suffer from feature…

Sparse Autoencoders (SAEs) extract features from LLM internal activations, meant to correspond to interpretable concepts. A core SAE training hyperparameter is L0: how many SAE features should fire per token on average. Existing work…

SAEs have recently been employed as a promising unsupervised approach for understanding the representations of layers of Large Language Models (LLMs). However, with the growth in model size and complexity, training SAEs is computationally…

Large vision-language models (LVLMs) have achieved remarkable performance on multimodal tasks. However, they still suffer from hallucinations, generating text inconsistent with visual input, posing significant risks in real-world…

Sparse autoencoders (SAEs) \citep{bricken2023monosemanticity,gao2024scalingevaluatingsparseautoencoders} rely on dictionary learning to extract interpretable features from neural networks at scale in an unsupervised manner, with…

Sparse Autoencoders (SAEs) have emerged as a useful tool for interpreting the internal representations of neural networks. However, naively optimising SAEs for reconstruction loss and sparsity results in a preference for SAEs that are…