ProtSAE: Disentangling and Interpreting Protein Language Models via Semantically-Guided Sparse Autoencoders

• URL: http://arxiv.org/abs/2509.05309v1
• Date: Tue, 26 Aug 2025 11:20:31 GMT
• Title: ProtSAE: Disentangling and Interpreting Protein Language Models via Semantically-Guided Sparse Autoencoders
• Authors: Xiangyu Liu, Haodi Lei, Yi Liu, Yang Liu, Wei Hu,
• Abstract summary: Sparse Autoencoder (SAE) has emerged as a powerful tool for mechanistic interpretability of large language models.<n>We propose a semantically-guided SAE, called ProtSAE.<n>We show that ProtSAE learns more biologically relevant and interpretable hidden features compared to previous methods.
• Score: 30.219733023958188
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Sparse Autoencoder (SAE) has emerged as a powerful tool for mechanistic interpretability of large language models. Recent works apply SAE to protein language models (PLMs), aiming to extract and analyze biologically meaningful features from their latent spaces. However, SAE suffers from semantic entanglement, where individual neurons often mix multiple nonlinear concepts, making it difficult to reliably interpret or manipulate model behaviors. In this paper, we propose a semantically-guided SAE, called ProtSAE. Unlike existing SAE which requires annotation datasets to filter and interpret activations, we guide semantic disentanglement during training using both annotation datasets and domain knowledge to mitigate the effects of entangled attributes. We design interpretability experiments showing that ProtSAE learns more biologically relevant and interpretable hidden features compared to previous methods. Performance analyses further demonstrate that ProtSAE maintains high reconstruction fidelity while achieving better results in interpretable probing. We also show the potential of ProtSAE in steering PLMs for downstream generation tasks.

Related papers

• Mechanistic Interpretability of Antibody Language Models Using SAEs [1.7218681244575125]
  We employparse autoencoders (SAEs) to provide insight into learned concepts within large protein language models.<n>TopK SAEs can reveal biologically meaningful latent features, but high feature concept correlation does not guarantee causal control over generation.<n>Ordered SAEs impose an hierarchical structure that reliably identifies steerable features, but at the expense of more complex and less interpretable activation patterns.
  arXiv  Detail & Related papers  (2025-12-05T15:18:50Z)
• Sparse Autoencoders are Topic Models [47.62628339598771]
  We show thatparse autoencoders (SAEs) can be naturally understood as topic models.<n>We introduce SAE-TM, a topic modeling framework that trains an SAE to learn reusable topic atoms.<n>We analyze thematic structure in image datasets and trace topic changes over time in Japanese woodblock prints.
  arXiv  Detail & Related papers  (2025-11-20T12:37:54Z)
• Where MLLMs Attend and What They Rely On: Explaining Autoregressive Token Generation [59.40886078302025]
  Multimodal large language models (MLLMs) have demonstrated remarkable capabilities in aligning visual inputs with natural language outputs.<n>Yet, the extent to which generated tokens depend on visual modalities remains poorly understood.<n>We present a lightweight black-box framework for explaining autoregressive token generation in MLLMs.
  arXiv  Detail & Related papers  (2025-09-26T15:38:42Z)
• Dense SAE Latents Are Features, Not Bugs [75.08462524662072]
  We show that dense latents serve functional roles in language model computation.<n>We identify classes tied to position tracking, context binding, entropy regulation, letter-specific output signals, part-of-speech, and principal component reconstruction.
  arXiv  Detail & Related papers  (2025-06-18T17:59:35Z)
• Taming Polysemanticity in LLMs: Provable Feature Recovery via Sparse Autoencoders [50.52694757593443]
  Existing SAE training algorithms often lack rigorous mathematical guarantees and suffer from practical limitations.<n>We first propose a novel statistical framework for the feature recovery problem, which includes a new notion of feature identifiability.<n>We introduce a new SAE training algorithm based on bias adaptation'', a technique that adaptively adjusts neural network bias parameters to ensure appropriate activation sparsity.
  arXiv  Detail & Related papers  (2025-06-16T20:58:05Z)
• Ensembling Sparse Autoencoders [10.81463830315253]
  Sparse autoencoders (SAEs) are used to decompose neural network activations into human-interpretable features.<n>We propose to ensemble multiple SAEs through naive bagging and boosting.<n>Our empirical results demonstrate that ensembling SAEs can improve the reconstruction of language model activations, diversity of features, and SAE stability.
  arXiv  Detail & Related papers  (2025-05-21T23:31:21Z)
• Sparse Autoencoders Learn Monosemantic Features in Vision-Language Models [50.587868616659826]
  We introduce a comprehensive framework for evaluating monosemanticity at the neuron-level in vision representations.<n>Our experimental results reveal that SAEs trained on Vision-Language Models significantly enhance the monosemanticity of individual neurons.
  arXiv  Detail & Related papers  (2025-04-03T17:58:35Z)
• Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders [115.34050914216665]
  Sparse Autoencoders (SAEs) have emerged as a powerful unsupervised method for extracting sparse representations from language models. We introduce a suite of 256 SAEs, trained on each layer and sublayer of the Llama-3.1-8B-Base model, with 32K and 128K features. We assess the generalizability of SAEs trained on base models to longer contexts and fine-tuned models.
  arXiv  Detail & Related papers  (2024-10-27T17:33:49Z)
• Automatically Interpreting Millions of Features in Large Language Models [1.8035046415192353]
  sparse autoencoders (SAEs) can be used to transform activations into a higher-dimensional latent space.<n>We build an open-source pipeline to generate and evaluate natural language explanations for SAE features.<n>Our large-scale analysis confirms that SAE latents are indeed much more interpretable than neurons.
  arXiv  Detail & Related papers  (2024-10-17T17:56:01Z)
• Can sparse autoencoders make sense of gene expression latent variable models? [0.0]
  This work explores the potential of SAEs for decomposing embeddings in complex and high-dimensional biological data.<n>The application to embeddings from pretrained single-cell models shows that SAEs can find and steer key biological processes.<n> scFeatureLens is an automated interpretability approach for linking SAE features and biological concepts from gene sets.
  arXiv  Detail & Related papers  (2024-10-15T10:16:01Z)
• Disentangling Dense Embeddings with Sparse Autoencoders [0.0]
  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. We show that the resulting sparse representations maintain semantic fidelity while offering interpretability.
  arXiv  Detail & Related papers  (2024-08-01T15:46:22Z)
• Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs [47.14410674505256]
  We present a case study of syntax acquisition in masked language models (MLMs)<n>We study Syntactic Attention Structure (SAS), a naturally emerging property of accessibles wherein specific Transformer heads tend to focus on specific syntactic relations.<n>We examine the causal role of SAS by manipulating SAS during training, and demonstrate that SAS is necessary for the development of grammatical capabilities.
  arXiv  Detail & Related papers  (2023-09-13T20:57:11Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.