Sparse Autoencoders

Also known as SAE

Follow this topic Ask about this topic

Momentum

16 papers in the last four weeks, up 100% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 149

May 24, 2026cs.CV

Universal Boosts, Specific Suppressors: Sparse Autoencoder Steering of Medical Vision-Language Models

Medical vision-language models (VLMs) often hallucinate findings when generating chest X-ray reports: they fabricate findings that are not present in the image, miss important ones, or locate them incorrectly. We mitigate this without weight updates by decoding-time residual steering on a per-token sparse autoencoder (SAE) basis: Top-

K

SAEs on late layers, causal steering against clinical errors, then combined suppress/boost intervention at inference time. On the MIMIC-CXR test split, our inference-only method improves the quality of generated reports for three radiology VLMs (RadVLM, LLaVA-Rad, and CheXOne), with relative improvements of +5.4%, +7.2%, and +17.0% in the clinical composite metric, and statistically significant GREEN gains on all backbones. A cross-model feature alignment shows that the quality-promoting (boost) directions overlap strongly across architectures, whereas hallucination-linked (suppress) directions are model-specific. Therefore, transferable steering must treat suppression per-backbone, rather than sharing a universal suppress list. The same recipe transfers zero-shot to IU-Xray (Green

+7.7\%

rel.) without retraining, confirming that the identified features are properties of the model, not of the training corpus. We release causal feature sets and an interactive feature dashboard: https://cxr-sparse-feature-dashboard.netlify.app/.

Farhad Nooralahzadeh, Benjamin Gundersen, Nicolas Deperrois +7

University of Zurich and University Hospital of Zurich, Switzerland · Zurich University of Applied Sciences, Switzerland · Kobe University, Japan +3

Radiology Report Generation Healthcare Object Hallucination in VLMs Sparse Autoencoders Medical VLMs Activation Steering+1

May 24, 2026cs.CV

Interpretability Transfer from Language to Vision via Sparse Autoencoders

Recent advances in language model interpretability using sparse autoencoders (SAEs) have yet to effectively translate to the visual domain, mainly due to the difficulty and ambiguity of labeling visual concepts. In this paper, we introduce Visual Interpretability via SAE Transfer Alignment (VISTA), a framework that transfers interpretability from language to vision in a LLaVA-style vision-language model by constraining a visual projector to map visual tokens into an LLM's pre-existing, labeled textual SAE space. This approach enables visual interpretability without training dedicated vision SAEs. By regularizing the projector using the LLM's SAE reconstruction loss, VISTA achieves a threefold increase in the matching rate, which measures how accurately the most activating textual concepts in the SAE space correspond to semantic elements in the image. Using this framework, we further analyze spatial localization properties of different vision encoders and show that DINOv2 features have stronger localization abilities than other encoders. Leveraging this precision, we validate VISTA's cross-modal alignment through fine-grained, localized concept interventions, where specific objects are removed or replaced in the model's perception while preserving the surrounding scene. This results in improvements of 35% in object removal and 47% in object replacement tasks over vision-only baselines, providing causal evidence that visual tokens inhabit the text SAE manifold. These contributions are validated across multiple LLM architectures.

Alexey Kravets, Da Li, Chuan Li +2

VLM Interpretability Sparse Autoencoders Vision-Language Alignment

May 23, 2026cs.LG

Polymorphism Is Rotation: Operational Mechanistic Interpretability from a Two-Layer Transformer to Pythia-70m

Independently trained transformers compute the same function in residual-stream bases that differ by a uniform random rotation on

\mathrm{SO}(d_{\mathrm{model}})

. We call this phenomenon polymorphism: same function, mutually unintelligible interior coordinates. One matrix multiplication per model pair removes it: an orthogonal Procrustes fit on a single batch of activations transfers sparse-autoencoder feature dictionaries and steering vectors between independently trained models, with no retraining. The phenomenon is invisible to the standard SAE universality metric. Decoder-column cosine similarity matches across seeds at 98%, the SAE-universality headline number, while an SAE trained on one seed reconstructs another seed's activations at negative explained variance, worse than predicting the constant mean. The decoder columns align; the encoder reads from a rotated frame. A single Procrustes rotation

R

restores reconstruction to within 0.025 EV of the within-seed ceiling at every internal site.

R

is Haar-distributed:

\|R - I\|_F

matches the random-orthogonal prediction

\sqrt{2 d_{\mathrm{model}}}

to 0.1% at

d_{\mathrm{model}} = 512

, and a Kolmogorov-Smirnov test of

R

's eigenvalue spectrum against Haar

\mathrm{SO}(d_{\mathrm{model}})

returns

p \approx 1.000

pooled and per-pair. Diff-of-means steering vectors transfer in three regimes by alignment with

R

's invariant subspace: clean when pinned by shared output weights, partial when overlapping the rotated subspace, inverted otherwise. With no shared I/O (Pythia), all three collapse to universally inverted. The same rotation account holds across training checkpoints within a single run. Validated on a 104k-parameter Dyck-3 transformer and nine independently-trained Pythia-70m seeds on The Pile, via a pre-registered four-bar operational framework. Frontier-scale (10B+) replication remains open.

Jordan F. McCann

Independent Researcher

Transformer Interpretability Transformer Sparse Autoencoders Mechanistic Interpretability Representation Alignment Latent Space Alignment

May 21, 2026cs.LG

Steered Generation via Gradient-Based Optimization on Sparse Query Features

Latent steering exploits internal representations of Large Language Models (LLMs) to guide generation, yet interventions on dense states can entangle distinct semantic features. In this paper, we investigate attention query activations as a high-fidelity site for precise control, hypothesizing that manipulating the attention mechanism itself offers sharper steerability than general state interventions. We introduce Prototype-Based Sparse Steering, a framework that applies Sparse Autoencoders (SAEs) specifically to query activations, to decompose them into interpretable features, then apply gradient-based optimization during inference to align the sparse representation with class prototypes of target behaviors. To validate this architectural insight, we first analyze the mechanism in Textualized Gridworld, a controlled environment for verifiable planning constraints. We demonstrate that optimizing sparse query features enables effective navigation of rigid planning requirements (i.e., safe vs. short paths), confirming the method's ability to satisfy objective rules. We then demonstrate the framework's versatility by training SAEs on a high-dimensional educational domain, where the framework steers the cognitive complexity of feedback (i.e., Bloom's Taxonomy). Our experiments establish that sparse query representations provide the necessary disentanglement for unified, interpretable control over both logical planning and stylistic nuance.

Sumanta Bhattacharyya, Pedram Rooshenas

Department of Computer Science University of Illinois Chicago

Constrained Motion Planning Language Model Steering Large Language Model-Guided Optimization Sparse Autoencoders Activation Steering

May 21, 2026cs.CL

Multilingual Steering by Design: Multilingual Sparse Autoencoders and Principled Layer Selection

Sparse autoencoders (SAEs) enable feature-level mechanistic interpretability and activation steering in large language models (LLMs), but SAE-based language control remains unreliable in multilingual settings: most SAEs are trained on English-only data, and steering layers are chosen heuristically. We address these limitations by advancing a principled, mechanistic account of multilingual language steering with SAEs. First, we show that training SAEs on multilingual data consistently strengthens cross-lingual representations and yields more reliable, quality-preserving language control across layers and model families. Second, we introduce an \emph{a priori} steering layer-selection rule based on the intersection of multilingual alignment and language separability, which predicts effective intervention depths without exhaustive layerwise search. We evaluate our approach on LLaMA-3.1-8B and Gemma-2-9B across machine translation and cross-lingual summarization (CrossSumm), using SpBLEU, ROUGE-L, COMET, and LaSE. Our results show that multilingual SAEs combined with intersection-selected layers stabilize the trade-off between language identification accuracy and generation quality, providing a principled, predictive, representation-level account of multilingual SAE steering.

Yusser Al Ghussin, Daniil Gurgurov, Tanja Baeumel +3

Saarland University · German Research Center for Artificial Intelligence (DFKI) · Centre for European Research in Trusted AI (CERTAIN) +1

Multilingual Language Models Language Model Steering Sparse Autoencoders Cross-Lingual Transfer

May 21, 2026cs.CL

Sparse Autoencoders Map Brain-LLM Alignment onto Cortical Semantic Topography

Intermediate layers of large language models (LLMs) best predict human brain responses to language, one of the most robust findings in computational neurolinguistics, yet why remains mechanistically unexplained. We address this gap by bridging sparse autoencoders (SAEs) from mechanistic interpretability with neural encoding models, decomposing GPT-2 XL and Llama-3.1-8B into 16K-32K interpretable features per layer. A human-validated taxonomy (

κ\geq 0.74

) reveals that semantic features alone recover 94% of peak encoding performance (

r=0.285

), substantially exceeding variance-matched baselines (

p<0.001

,

d=1.31

). Beyond this aggregate dominance, we test a novel cortical topography prediction: five semantic subcategories derived a priori from three independent neuroscience programs should map onto distinct brain regions. A formal convergence test confirms this alignment (Spearman

ρ=0.72

,

p<0.001

; hypergeometric

p=0.007

), demonstrating that SAE-discovered features recapitulate known cortical semantic organization at a granularity inaccessible to prior methods. SAE features further predict human reading times beyond lexical controls (

Δ\mathrm{logLik}=38.4

,

p<0.001

), and an exploratory prediction-error analysis provides preliminary evidence that the brain additionally encodes unexpected semantic content. Results generalize across English, Chinese, and French.

Dongxin Guo, Jikun Wu, Siu Ming Yiu

The University of Hong Kong · Hong Kong, China · Stellaris AI Limited

Neural Processes LLM Interpretability Sparse Autoencoders Neural Encoding

May 21, 2026cs.LG

Reading Task Failure Off the Activations: A Sparse-Feature Audit of GPT-2 Small on Indirect Object Identification

We report a small, reproducible audit of which sparse-autoencoder (SAE) features of GPT-2 small fire differently on failed versus successful trials of the Indirect Object Identification (IOI) task. On 300 prompts, GPT-2 small reaches 79.7% accuracy; 146 of the 24,576 features in the layer-8 residual-stream SAE release of Bloom (2024) clear a Holm-corrected significance threshold and 105 reach a large effect size (|Cohen's d| > 0.8). The strongest single correlate of failure -- feature 17,491, d=+2.93, Neuronpedia label 'cryptographic keys' -- is essentially silent except when the prompt's transferred object is 'the keys,' on which GPT-2 small fails 93.3% of the time vs. 7.5% on the other seven objects (Fisher exact p = 8.79 x 10^-33). We put this correlate through three controls that a mechanistic claim should pass. (i) A causal ablation: zeroing feature 17,491 in the residual stream across all token positions of the 45 keys prompts does not restore accuracy (6.7% -> 4.4%); the feature is a correlate, not a sufficient cause at this layer. (ii) A representation baseline: a logistic regression on the raw 768-dimensional residual stream reaches 5-fold ROC AUC = 0.929, matching the top-100 SAE features (0.927); the SAE basis adds interpretability, not predictive power. (iii) A seed-robustness check: across five random seeds the keys-subset failure rate stays in 75.0--93.3% (the behavioural effect is real), but feature 17,491 is the top-|d| feature in only 1 of 5 runs. The methodological contribution is therefore the audit pipeline (cheap, model-agnostic, surfaces named correlates) rather than any single feature found through it. We release the code, the 300-prompt corpus, the 300x24,576 activation matrix, the ablation and baseline scripts, and the figures. The full pipeline runs on a laptop (Apple M3 Max, no discrete GPU).

Mahdi Nasermoghadasi

Research Division, BrightMind AI, Seattle, WA · Texas Tech University, Lubbock, TX

Transformer Interpretability Sparse Autoencoders Language Model Auditing

May 21, 2026cs.CV

Conceptualizing Embeddings: Sparse Disentanglement for Vision-Language Models

Vision-language models learn powerful multimodal embeddings, yet their internal semantics remain opaque. While sparse autoencoders (SAEs) can extract interpretable features, they rely on expanding the representation dimension, which compromises the original geometry and introduces redundancy. We introduce CEDAR (Conceptual Embedding Disentanglement via Adaptive Rotation), a post-hoc method that reveals the compositional structure of pretrained embeddings without increasing dimensionality. By learning an invertible transformation with a top-

k

sparsity bottleneck, CEDAR concentrates semantic information into axis-aligned disentangled coordinates. In CLIP-like architecture, individual coordinates can be interpreted with textual concepts, while for generative models such as BLIP, they can be decoded into natural language descriptions. Experiments demonstrate that CEDAR achieves a competitive reconstruction-sparsity trade-off while producing explanations that are more interpretable and better aligned with human perception. Our results suggest that the apparent entanglement in vision-language representations can be resolved through a suitable change of basis, eliminating the need for overcomplete expansions.

Piotr Kubaty, Patryk Marszałek, Łukasz Struski +3

Faculty of Mathematics and Computer Science, Jagiellonian University, Kraków, Poland · Doctoral School of Exact and Natural Sciences, Jagiellonian University, Kraków, Poland

Disentangled Representation Learning Multimodal Disentangled Representation Learning VLM Interpretability Sparse Autoencoders Multimodal Embedding

May 18, 2026cs.LG

Aligned Training: A Parameter-Free Method to Improve Feature Quality and Stability of Sparse Autoencoders (SAE)

Sparse autoencoders (SAEs) are one of the main methods to interpret the inner workings of deep neural networks (DNNs), decomposing activations into higher-dimensional features. However, they exhibit critical shortcomings where a large fraction of features are never activated and are unstable. Despite variants of SAEs that attempt to mitigate these issues, they require additional data, resampling, or training. We propose the \textbf{aligned training}, a parameter-free reparameterization of SAEs that simultaneously improves reconstruction quality, eliminates dead features, and significantly enhances stability across training seeds. Our approach is motivated by an overlooked observation that SAE feature quality, measured by the inner product between encoder and decoder directions (which we call the \textbf{alignment score}), follows a bimodal distribution across all modern architectures. The proposed aligned training enforces a geometric constraint between the encoder and decoder such that their inner product equals one for every feature, which removes a source of degeneracy in the SAE training without adding any hyperparameters. Across multiple models, dictionary sizes, and sparsity levels, the aligned training shows Pareto improvements on the SAEBench benchmarks. Beyond improving dead features, stability and reconstruction, our method readily integrates with techniques in mechanical interpretability such as Top/BatchTop-K architectures and p-Annealing. Overall, the aligned training substantially improves feature quality and stability of SAE without computational complexity or cost.

Michał Brzozowski, Neo Christopher Chung

Samsung AI Center, Warsaw, Poland · University of Warsaw, Poland

Sparse Autoencoders Mechanistic Interpretability Representation Alignment

May 18, 2026cs.LG

Are Sparse Autoencoder Benchmarks Reliable?

Sparse autoencoders (SAEs) are a core interpretability tool for large language models, and progress on SAE architectures depends on benchmarks that reliably distinguish better SAEs from worse ones. We audit the SAE quality metrics in SAEBench, the de-facto standard SAE evaluation suite, through three complementary lenses: reseed noise on a fixed SAE, ground-truth correlation on synthetic SAEs, and discriminability across training trajectories. We find that two of these metrics, Targeted Probe Perturbation (TPP) and Spurious Correlation Removal (SCR), fail multiple lenses at their canonical settings and should not be used to evaluate SAEs. The other metrics show higher reseed noise and lower discriminability than the field assumes. The sae-probes variant of

k

-sparse probing is the most reliable metric we tested, but even sae-probes struggles to separate variants of the same SAE architecture. Our results show the field needs better SAE benchmarks.

David Chanin

Decode Research, MATS, UCL

Benchmark Auditing LLM Interpretability Sparse Autoencoders Benchmark Validity

May 17, 2026cs.LG

Beyond Linear Superposition: Discovering Climate Features in AI Weather Models with KAN-SAE

Deep learning weather prediction models achieve remarkable predictive skill yet remain largely opaque: we know little about how they represent physical climate phenomena internally. Mechanistic interpretability through Sparse Autoencoders (SAEs) offers a principled route to decomposing these representations, but existing SAEs assume strictly linear feature superposition - a constraint ill-suited for the highly nonlinear atmospheric dynamics encoded in modern transformers. We introduce KAN-SAE, a sparse autoencoder whose encoder replaces the standard ReLU with learnable per-feature B-spline activations drawn from Kolmogorov-Arnold Networks (KANs), allowing each latent dimension to develop its own nonlinear gating profile. Applied to Sonny, KAN-SAE discovers 975 alive features (vs. 566 for a linear baseline, a 72% improvement) with 20% lower inter-feature redundancy and comparable reconstruction fidelity. Without any climate supervision, KAN-SAE identifies an interpretable European heatwave feature spatially concentrated over western Europe, and a western Pacific typhoon tracker confirmed by causal steering experiments. Our results demonstrate that nonlinear activations are essential for mechanistic interpretability of deep learning weather prediction models, recovering climate features that remain invisible to linear baselines.

Minjong Cheon

Department of Computer Science and Engineering Sejong University

Transformer Interpretability Climate Science Sparse Autoencoders Mechanistic Interpretability Climate Modeling Kolmogorov-Arnold Networks+1

May 17, 2026cs.RO

Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies

Vision-Language-Action (VLA) policies translate language and visual inputs into robot actions, where their hidden representations directly shape closed-loop behavior. However, mechanistic interpretability tools from language and vision-language models do not transfer cleanly to VLAs: outputs are robot actions rather than human-readable tokens, and interventions can only be tested via expensive closed-loop rollouts. We propose an event-grounded interpretability pipeline that anchors SAE feature analysis to behavioral events rather than text contexts. End-effector keyframes are clustered within each task using visual, state, and temporal cues, linking SAE features to behaviorally salient events and, via optional VLM annotations, to semantic context. To our knowledge, our pipeline is among the first to ground SAE-based VLA analysis in closed-loop behavioral events. Across two simulation architectures and a real-robot study, event-grounded ranking yields the strongest causal effects on OpenVLA and transfers to the continuous action chunks of

π_{0.5}

. SAE is a sparse but imperfect intervention basis: usability varies with architecture and intervention site, and aggressive intervention reveals safety and interpretability limits. Overall, event-grounded SAE analysis emerges as a practical starting point for behavior-anchored VLA interpretability, motivating future work on SAE features beyond action-aligned coordinates, finer-grained closed-loop evaluation, and safe interventions for high-stakes VLA deployments. Code is available at https://github.com/xc-j/Event-SAE.

Xinchen Jin, Aditya Chatterjee, Pranav Kumar +1

Department of Computer Science, Purdue University West Lafayette, IN 47907

VLM Interpretability Sparse Autoencoders Mechanistic Interpretability Vision-Language-Action Models

May 15, 2026cs.CV

Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models

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 shifts. Recent approaches have attempted to mitigate this trade-off, but often rely on computationally expensive text-guidance. We propose a novel method for robust fine-tuning, SAE-FT, which operates only on the model's visual representations. SAE-FT regularizes changes to these representations by penalizing the addition and removal of semantically meaningful features identified by a Sparse Autoencoder trained on the pre-trained model. This constraint prevents catastrophic forgetting and makes the fine-tuning process interpretable, enabling direct analysis of semantic changes. SAE-FT is both mechanistically transparent and computationally efficient, matching or exceeding state-of-the-art performance on ImageNet and its associated distribution shift benchmarks. Code is publicly available at: https://github.com/Fabian-Mor/sae-ft.

Fabian Morelli, Arnas Uselis, Ankit Sonthalia +1

University of Tübingen ∗ · University of Tübingen

Distribution Shift Robustness VLM Robustness VLM Adaptation VLM Interpretability Sparse Autoencoders

May 14, 2026cs.LG

The Rate-Distortion-Polysemanticity Tradeoff in SAEs

Sparse Autoencoders (SAEs) that can accurately reconstruct their input (minimizing distortion) by making efficient use of few features (minimizing the rate) often fail to learn monosemantic representations (highly interpretable), limiting their usefulness for mechanistic interpretability. In this paper, we characterise this tension in learning faithful, efficient, and interpretable explanations, introducing the Rate-Distortion-Polysemanticity tradeoff in SAEs. Under toy-modeling assumptions, we theoretically and empirically show that restricting the SAE to be monosemantic necessarily comes with an increase in rate and distortion. Assuming a generative model behind the input observations, we further demonstrate that the degree of polysemanticity of optimal SAEs is determined by the training data distribution, especially by the probability of features to co-occur. Finally, we extend the analysis to real-world settings by deriving necessary conditions that a polysemanticity measure should satisfy when the data-generating process is unknown, and we benchmark existing proxy metrics on SAEs trained on Large Language Models. Taken together, our findings show that polysemanticity is a data problem that should be accounted for when addressing it at the architectural and optimization level.

Tommaso Mencattini, Francesco Montagna, Francesco Locatello

EPFL · Institute of Science and Technology Austria

Rate-Distortion Theory Representation Disentanglement Sparse Autoencoders Mechanistic Interpretability Sparse Coding

May 13, 2026cs.LG

Mechanistic Interpretability of EEG Foundation Models via Sparse Autoencoders

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 distinct EEG transformers: SleepFM, REVE, and LaBraM to extract sparse feature dictionaries from their embeddings. By grounding these features in a clinical taxonomy (abnormality, age, sex, and medication), we benchmark monosemanticity and entanglement across architectures. A single hyperparameter procedure, driven by an intrinsic dictionary health audit, transfers robustly across all three architectures. Via concept steering, we introduce a "target vs. off-target" probe area metric to quantify steering selectivity and reveal three operational regimes: selectively steerable, encoded but entangled, and non-encoded. This framework exposes critical representational failures: "wrecking-ball" interventions that collapse global model performance, and clinical entanglements, such as age-pathology confounding, where it is impossible to suppress one concept without corrupting the other. Finally, a spectral decoder maps these interventions back to the amplitude spectrum, translating latent manipulations into physiologically interpretable frequency signatures, such as pathological slow-wave suppression and

α

-band restoration.

William Lehn-Schiøler, Magnus Ruud Kjær, Rahul Thapa +10

BrainCapture, Kongens Lyngby, Denmark · DTU Health Tech, Technical University of Denmark, Kongens Lyngby, Denmark · DTU Compute, Technical University of Denmark, Kongens Lyngby, Denmark +5

Sparse Autoencoders Mechanistic Interpretability EEG Foundation Models

May 13, 2026cs.LG

Descriptive Collision in Sparse Autoencoder Auto-Interpretability: When One Explanation Describes Many Features

Sparse autoencoders (SAEs) are now standard tools for decomposing language model activations into interpretable features, and automated interpretability pipelines routinely assign each feature a short natural-language explanation. Existing critiques of this practice focus on polysemanticity -- one feature with many meanings -- or on whether explanations predict activations. We identify a complementary, structurally distinct problem we call descriptive collision: many distinct SAE features admit the same explanation. Reanalyzing the largest publicly-available dataset of human-annotated SAE features (Marks et al., 2025), comprising 722 annotated features across Gemma 2 2B and Pythia 70M, we find that the mean annotation string is reused across 3.07 features; 82.1% of features share their annotation with at least one other feature; and the single most common annotation string ("plural nouns") labels 101 distinct features spanning 18 layers and four model components. Information-theoretically, the average annotation resolves only 70% of feature identity. We formalize a property called discrimination, prove that current detection-style auto-interpretability scoring is invariant to collision, and propose two complementary corrective metrics -- collision-adjusted detection and discrimination scoring -- that explicitly penalize explanations that fail to distinguish a feature from its neighbors. The collision problem is independent of, and additive with, previously identified failure modes of auto-interpretability; ignoring it inflates reported feature interpretability by a quantity equal to roughly one-third of the bits required to identify a feature.

Jordan F. McCann

Independent Researcher

LLM Interpretability Sparse Autoencoders Mechanistic Interpretability Explainability Evaluation

May 12, 2026cs.LG

WriteSAE: Sparse Autoencoders for Recurrent State

We introduce WriteSAE, a sparse autoencoder for the matrix updates written into recurrent language-model state. In Gated DeltaNet, Mamba-2, and RWKV-7, each token writes a matrix-shaped update to a recurrent cache; a residual-stream SAE has vector-shaped atoms and cannot replace that update directly. WriteSAE learns rank-1 matrix atoms with the same shape as the model's own write. This lets us test a direct replacement: at positions where the SAE activates an atom, we remove the model's write, insert the atom scaled by its SAE activation, and continue the forward pass. The atom gives a closer final token distribution than deleting the write on 92.4% of evaluated positions; averaged per atom, the rate is 89.8%. For Gated DeltaNet, a formula using the forget gate, read query, and output embedding predicts the resulting logit change with

R^2 = 0.98

. The same replacement test transfers to Mamba-2-370M at 88.1%. In generation, the formula chooses a write direction; writing it into three consecutive cache positions at

3\times

the norm of the model's write makes tokens initially ranked 100--1000 by the unmodified model appear in 100% of continuations, up from 33.3%. To our knowledge this is the first cache-level steering intervention reported in a state-space or hybrid recurrent layer.

Jack Young

Indiana University

Language Model Steering Sparse Autoencoders Language Modeling State Representation Learning Causal Interventions in Language Models

May 12, 2026cs.LG

Disentangled Sparse Representations for Concept-Separated Diffusion Unlearning

Unlearning specific concepts in text-to-image diffusion models has become increasingly important for preventing undesirable content generation. Among prior approaches, sparse autoencoder (SAE)-based methods have attracted attention due to their ability to suppress target concepts through lightweight manipulation of latent features, without modifying model parameters. However, SAEs trained with sparse reconstruction objectives do not explicitly enforce concept-wise separation, resulting in shared latent features across concepts. To address this, we propose SAEParate, which organizes latent representations into concept-specific clusters via a concept-aware contrastive objective, enabling more precise concept suppression while reducing unintended interference during unlearning. In addition, we enhance the encoder with a GeLU-based nonlinear transformation to increase its expressive capacity under this separation objective, enabling a more discriminative and disentangled latent space. Experiments on UnlearnCanvas demonstrate state-of-the-art performance, with particularly strong gains in joint style-object unlearning, a challenging setting where existing methods suffer from severe interference between target and non-target concepts.

Hyeonjin Kim, Hangyeol Jung, Heechan Yun +2

Yonsei University · Kookmin University

Disentangled Representation Learning Sparse Autoencoders Diffusion Model Unlearning Machine Unlearning T2I Diffusion Models Diffusion Models

May 12, 2026cs.CL

Qwen-Scope: Turning Sparse Features into Development Tools for Large Language Models

Large language models have achieved remarkable capabilities across diverse tasks, yet their internal decision-making processes remain largely opaque, limiting our ability to inspect, control, and systematically improve them. This opacity motivates a growing body of research in mechanistic interpretability, with sparse autoencoders (SAEs) emerging as one of the most promising tools for decomposing model activations into sparse, interpretable feature representations. We introduce Qwen-Scope, an open-source suite of SAEs built on the Qwen model family, comprising 14 groups of SAEs across 7 model variants from the Qwen3 and Qwen3.5 series, covering both dense and mixture-of-expert architectures. Built on top of these SAEs, we show that SAEs can go beyond post-hoc analysis to serve as practical interfaces for model development along four directions: (i) inference-time steering, where SAE feature directions control language, concepts, and preferences without modifying model weights; (ii) evaluation analysis, where activated SAE features provide a representation-level proxy for benchmark redundancy and capability coverage; (iii) data-centric workflows, where SAE features support multilingual toxicity classification and safety-oriented data synthesis; and (iv) post-training optimization, where SAE-derived signals are incorporated into supervised fine-tuning and reinforcement learning objectives to mitigate undesirable behaviors such as code-switching and repetition. Together, these results demonstrate that SAEs can serve not only as post-hoc analysis tools, but also as reusable representation-level interfaces for diagnosing, controlling, evaluating, and improving large language models. By open-sourcing Qwen-Scope, we aim to support mechanistic research and accelerate practical workflows that connect model internals to downstream behavior.

Boyi Deng, Xu Wang, Yaoning Wang +15

Qwen Team

LLM Interpretability Sparse Autoencoders Activation Steering

May 11, 2026cs.LG

HH-SAE: Discovering and Steering Hierarchical Knowledge of Complex Manifolds

Rare semantic innovations in high-dimensional, mission-critical domains are often obscured by dense background contexts, a challenge we define as \textit{feature density conflict}. We introduce the \textbf{Hybrid Hierarchical SAE (HH-SAE)} to resolve this by factorizing manifolds into a nested hierarchy of \textbf{Contextual} (

L_0

), \textbf{Atomic} (

f_1

), and \textbf{Compository} (

f_2

) tiers. Evaluating across disparate manifolds, HH-SAE demonstrates superior resolution by \textbf{``fracturing'' administrative clinical labels into physiological modes} and achieving a peak \textbf{cross-domain zero-shot AUC of 0.9156 in fraud detection}. Path ablation confirms the architecture's structural necessity, revealing a 13.46% utility collapse when contextual subtraction is removed. Finally, knowledge-steered synthesis achieves a +9.9% AUPRC lift over state-of-the-art generators, proving that HH-SAE effectively prioritizes high-order mechanistic innovation over environmental proxies to enable high-precision discovery in high-stakes environments.

Honghan Wu, Tianyan Wang, Jiacong Mi +2

University of Glasgow, Glasgow, UK · University of International Relations, Beijing, China · University College London, London, UK

Hierarchical Representation Learning Sparse Autoencoders

May 11, 2026cs.LG

The Geometric Wall: Manifold Structure Predicts Layerwise Sparse Autoencoder Scaling Laws

Sparse autoencoders (SAEs) operationalise the linear representation hypothesis: they reconstruct model activations as sparse linear combinations of interpretable dictionary atoms, on the implicit assumption that activation space is well approximated by a globally linear structure. Their reconstruction error varies sharply across layers in ways that existing scaling laws, fitted at single layers, do not explain. We argue that this variation is the empirical trace of a geometric mismatch: where the activation manifold is curved and its intrinsic dimension varies across layers, no sparse linear dictionary can match it uniformly, and the SAE's width-sparsity scaling becomes a layer-dependent function of manifold structure rather than a single universal law. We conduct the first cross-layer SAE scaling study, fitting and regressing on 844 residual-stream Gemma Scope SAE checkpoints across 68 layers of Gemma 2 2B and 9B. Stage 1 fits a per-layer scaling-law surface; Stage 2 regresses the fitted parameters and the derived per-layer width exponents on four layerwise geometric summaries. We find that manifold geometry predicts the per-layer width exponent in both models, and that the same regression coefficients learnt on one model predict the other model's per-layer exponents under cross-model transfer, indicating a transferable geometric law. At the showcase layers where richer width grids permit identification of the asymptotic floor, we find that the fitted floor tracks the layerwise geometric ordering: higher curvature and intrinsic dimension correspond to higher floor, consistent with the irreducible second-order residual that any sparse linear approximation of a curved manifold must leave behind. SAEs thus encounter not a finite-resource ceiling but a geometry-dependent wall, set by the manifold they are trying to reconstruct.

Eslam Zaher, Maciej Trzaskowski, Quan Nguyen +1

ARC Training Centre for Information Resilience (CIRES) · School of Mathematics and Physics, University of Queensland · Institute for Molecular Bioscience, University of Queensland +2

Neural Representation Geometry Sparse Autoencoders

May 10, 2026cs.LG

fmxcoders: Factorized Masked Crosscoders for Cross-Layer Feature Discovery

Many features in pretrained Transformers span multiple layers: they emerge through stages of inference, persist in the residual stream, or are built jointly by parallel MLPs. Crosscoders (namely, sparse dictionaries trained jointly across layers) aim to recover these cross-layer features in a single shared latent space. We show that standard crosscoders largely fail at this purpose. Although their decoder weight norms spread evenly across layers, a functional coherence metric we introduce reveals that each latent's activation is effectively driven by only one or two layers on average. While functionally coherent latents act as human-interpretable concept detectors (e.g., US states and cities), the layer-localized latents that crosscoders predominantly learn collapse onto surface-level patterns such as digit detectors. We trace this failure to two structural limitations: unconstrained cross-layer parameterization and unregularized cross-layer dependence. We address both by introducing fmxcoders, which (i) replace the encoder and decoder with low-rank tensor factorizations that draw every latent's per-layer weights from a shared cross-layer basis, and (ii) apply stochastic layer masking, a denoising regularizer along the layer axis that penalizes latents whose contribution collapses when a single layer is masked. Across GPT2-Small, Pythia-410M, Pythia-1.4B, and Gemma2-2B, fmxcoders lift mean probing F1 by 10-30 points, surpassing per-layer SAE baselines that standard crosscoders fail to reach, reduce reconstruction MSE by 25-50%, and roughly double mean functional coherence. An LLM-as-a-judge evaluation further shows that fmxcoders recover 3-13

\times

more semantically coherent latents than standard crosscoders across all four base LLMs.

Andreas D. Demou, Panagiotis Koromilas, James Oldfield +2

1The Cyprus Institute · University of Athens · 4Archimedes AI/Athena Research Center +2

Transformer Interpretability Sparse Autoencoders Mechanistic Interpretability

May 9, 2026cs.LG

SMIXAE: Towards Unsupervised Manifold Discovery in Language Models

Sparse autoencoders (SAEs) have been used widely to decompose and interpret neural network activations, especially those of transformer language models. One key issue with SAEs is their inability to directly model multidimensional features. Instead, SAEs may tile such features by a set of independent directions that must be grouped together after the SAE training phase, impeding discoverability and interpretation of learned feature representations. We begin to address this issue by introducing the Sparse MIXture of Autoencoders (SMIXAE) architecture. Empirically, we provide evidence that SMIXAE models have success both in directly learning previously identified manifold structures, as well as finding novel structures, within the open source Gemma 2 2B and 9B models. Finally, we discuss several limitations and point towards areas for future work.

Collin Francel

Department of Arts and Sciences, University of Alabama, Tuscaloosa, AL, United States.

Autoencoders LLM Interpretability Sparse Autoencoders Manifold Learning

May 9, 2026cs.LG

Causal Dimensionality of Transformer Representations: Measurement, Scaling, and Layer Structure

Sparse autoencoders (SAEs) decompose transformer residual streams into interpretable feature dictionaries, yet the relationship between SAE width and causal influence on model output has not been systematically characterised. We introduce causal dimensionality kappa(L, M, T), defined as the effective rank of the expected Jacobian outer product at layer L, and show it can be estimated via the SAE width sweep paired with attribution patching. Across seven SAE widths from 16,384 to 1,048,576 features on Gemma-2-2B layer 12, representational capacity grows 15.6x while causal capacity grows only 4.35x: a robust separation we term the representational-causal wedge. A saturating fit yields kappa-hat approximately 1,990 with kappa-hat / d_model = 0.86 and participation-ratio lower bound kappa_PR approximately 280. Crucially, kappa is invariant to model scaling: Gemma-2-9B and Gemma-2-2B yield identical N_causal = 328 at the same SAE width despite a 3.46x parameter increase (the count is forced to 2% of SAE width by calibration; the substantive empirical claim is shape invariance of the AtP score distribution under matched seq=512 conditions). Across eight network depths kappa is constant while the absolute attribution threshold drops 20x from layer 1 to layer 23. Five controls (architecture invariance, threshold robustness, geometric privilege, synthetic ground-truth recovery, and a four-cell encoder/decoder ablation) pin down what kappa measures and what it does not. Our findings establish kappa as a measurable, model-intrinsic property of transformer layers: sub-linearly recoverable by SAE width, invariant to model scaling, and structured across network depth.

Nilesh Sarkar, Dawar Jyoti Deka

Erdős AI Lab

Transformer Interpretability Sparse Autoencoders Causal Interpretability

May 8, 2026cs.LG

Tree SAE: Learning Hierarchical Feature Structures in Sparse Autoencoders

Learning hierarchical features in Sparse Autoencoders (SAEs) is essential for capturing the structured nature of real-world data and mitigating issues like feature absorption or splitting. Existing works attempt to identify hierarchical relationships within independent feature sets by relying on activation coverage, the assumption that child feature should only activate when its parent feature activates. However, we demonstrate that this condition alone is insufficient; that is, it often produces false positives where parent and child concepts are semantically unrelated. To address this, we introduce a novel reconstruction condition that enforces a deeper functional link between hierarchical levels. By combining both activation and reconstruction constraints, we propose the Tree SAE, a model designed to learn hierarchical structures directly from within the feature set. Our results demonstrate that Tree SAEs significantly surpass the existing SAEs at learning hierarchical pairs while maintaining competitive performance to the state-of-the-art on several key benchmarks. Finally, we demonstrate the practical utility of our Tree SAE in mapping the geometry of child feature subspaces and uncovering the complex hierarchical concept structures encoded within large language models.

Tue M. Cao, Hoang X. Nhat, Raed Alharbi +2

Hanoi University of Science and Technology, Hanoi, Vietnam · University of Florida, Florida, USA · Computer Science Department, Saudi Electronic University

Hierarchical Representation Learning Sparse Autoencoders Mechanistic Interpretability

May 8, 2026cs.CV

Sparse Autoencoders as Plug-and-Play Firewalls for Adversarial Attack Detection in VLMs

Vision-language models (VLMs) have advanced rapidly and are increasingly deployed in real-world applications, especially with the rise of agent-based systems. However, their safety has received relatively limited attention. Even the latest proprietary and open-weight VLMs remain highly vulnerable to adversarial attacks, leaving downstream applications exposed to significant risks. In this work, we propose a novel and lightweight adversarial attack detection framework based on sparse autoencoders (SAEs), termed SAEgis. By inserting an SAE module into a pretrained VLM and training it with standard reconstruction objectives, we find that the learned sparse latent features naturally capture attack-relevant signals. These features enable reliable classification of whether an input image has been adversarially perturbed, even for previously unseen samples. Extensive experiments show that SAEgis achieves strong performance across in-domain, cross-domain, and cross-attack settings, with particularly large improvements in cross-domain generalization compared to existing baselines. In addition, combining signals from multiple layers further improves robustness and stability. To the best of our knowledge, this is the first work to explore SAE as a plug-and-play mechanism for adversarial attack detection in VLMs. Our method requires no additional adversarial training, introduces minimal overhead, and provides a practical approach for improving the safety of real-world VLM systems.

Hao Wang, Yiqun Sun, Pengfei Wei +2

2Waseda University · 1Magellan Technology Research Institute (MTRI)

VLM Robustness Adversarial Attacks on VLMs Sparse Autoencoders Adversarial Example Detection

May 8, 2026cs.CL

Activation Differences Reveal Backdoors: A Comparison of SAE Architectures

Backdoor attacks on language models pose a significant threat to AI safety, where models behave normally on most inputs but exhibit harmful behavior when triggered by specific patterns. Detecting such backdoors through mechanistic interpretability remains an open challenge. We investigate two sparse autoencoder architectures -- Crosscoders and Differential SAEs (Diff-SAE) -- for isolating backdoor-related features in fine-tuned models. Using a controlled SQL injection backdoor triggered by year-based context ("2024" triggers vulnerable code, "2023" triggers safe code), we evaluate both approaches across LoRA and full-rank fine-tuning regimes on SmolLM2-360M. We find that Diff-SAE consistently and substantially outperforms Crosscoders for backdoor isolation. Diff-SAE achieves a Backdoor Isolation Score (BIS) of 0.40 with perfect precision (1.0) and zero false positive rate across most experimental conditions, while Crosscoders fail almost entirely with BIS below 0.02 in most cases. This performance gap holds across multiple transformer layers (14, 18, 22, 26) and both fine-tuning regimes, with full-rank fine-tuning producing particularly clean backdoor signals. Our results suggest that backdoors manifest as directional activation shifts rather than sparse feature activations, making difference-based representations fundamentally more effective for detection. These findings have important implications for AI safety monitoring and the development of interpretability tools for detecting model manipulation.

Sachin Kumar

LexisNexis* · Raleigh, USA

LLM Backdoor Attacks LLM Interpretability Backdoor Detection Sparse Autoencoders

May 7, 2026cs.LG

SoftSAE: Dynamic Top-K Selection for Adaptive Sparse Autoencoders

Sparse Autoencoders (SAEs) have become an important tool in mechanistic interpretability, helping to analyze internal representations in both Large Language Models (LLMs) and Vision Transformers (ViTs). By decomposing polysemantic activations into sparse sets of monosemantic features, SAEs aim to translate neural network computations into human-understandable concepts. However, common architectures such as TopK SAEs rely on a fixed sparsity level. They enforce the same number of active features (K) across all inputs, ignoring the varying complexity of real-world data. Natural data often lies on manifolds with varying local intrinsic dimensionality, meaning the number of relevant factors can change significantly across samples. This suggests that a fixed sparsity level is not optimal. Simple inputs may require only a few features, while more complex ones need more expressive representations. Using a constant K can therefore introduce noise in simple cases or miss important structure in more complex ones. To address this issue, we propose SoftSAE, a sparse autoencoder with a Dynamic Top-K selection mechanism. Our method uses a differentiable Soft Top-K operator to learn an input-dependent sparsity level k. This allows the model to adjust the number of active features based on the complexity of each input. As a result, the representation better matches the structure of the data, and the explanation length reflects the amount of information in the input. Experimental results confirm that SoftSAE not only finds meaningful features, but also selects the right number of features for each concept. The source code is available at: https://github.com/St0pien/SoftSAE.

Jakub Stępień, Marcin Mazur, Jacek Tabor +1

Jagiellonian University · IDEAS Research Institute

Sparse Autoencoders Mechanistic Interpretability

May 7, 2026cs.LG

Preference Instability in Reward Models: Detection and Mitigation via Sparse Autoencoders

Preference learning in large language models relies on reward models as proxies for human judgment. However, these models frequently exhibit preference instability, producing contradictory preference assignments in response to subtle, meaning-preserving input variations. We analyze this instability at the representation level under three semantic-preserving perturbation types: paraphrasing, pattern injection, and backdoor triggers. We attribute this instability to over-reliance on predictive yet brittle features, which we term unstable features, and isolate them via Sparse Autoencoders (SAEs) in a sparse latent space where benign and perturbed inputs activate distinctly separable patterns. Building on this separability, we propose two SAE-based instability mitigation strategies: SAE Feature Steering, which identifies and suppresses anomalously activated features at inference, and SAE Residual Correction, which learns adaptive adjustments over SAE features to restore correct preferences. Our methods substantially reduce incorrect preference assignments on harmlessness and hallucination benchmarks while preserving benign performance and general utility on other tasks, without retraining the reward model.

Shunchang Liu, Xin Chen, Belen Martin Urcelay +1

ETH Zürich · Georgia Institute of Technology · ELLIS Institute Finland +1

Reward Modeling Sparse Autoencoders Language Model Robustness

May 7, 2026cs.AI

From Token Lists to Graph Motifs: Weisfeiler-Lehman Analysis of Sparse Autoencoder Features

Sparse autoencoders (SAEs) have become central to mechanistic interpretability, decomposing transformer activations into monosemantic features. Yet existing analyses characterise features almost exclusively through top-activating token lists or decoder weight vectors, leaving the higher-order co-occurrence structure shared across features largely unexamined. We introduce a graph-structured representation in which each SAE feature is modelled as a token co-occurrence graph: nodes are the tokens most frequent near strong activations, and edges connect pairs that co-occur within local context windows. A custom WL-style, frequency-binned graph kernel then provides a similarity measure over this structural space. Applied as a proof of concept to features from a large SAE trained on GPT-2 Small and probed with a synthetic mixed-domain corpus, our clustering recovers heuristic motif families (punctuation-heavy patterns, language and script clusters, and code-like templates) that are not recovered by clustering on decoder cosine similarity. A token-histogram baseline achieves higher overall purity, so the contribution of the graph view is complementary rather than dominant: it surfaces structural relationships that token-frequency and decoder-weight views alone do not capture. Cluster assignments are stable across graph-construction hyperparameters and random seeds.

Ruben Fernandez-Boullon, Pablo Magariños-Docampo, Javier Perez-Robles

University of Vigo

Transformer Interpretability Sparse Autoencoders Kernel Methods