← Back to all news
arXiv AI By Devesh Tiwari, Camille Davis, Shivank Sinha, Talia Weaver, Aditya Shah, Maheep Chaudhary

Decodability is Not Causality: Dissociating Probe Readouts from Behavioral Drivers via SAE Decomposition

Read the original on arXiv AI →

Linear probes can decode safety‑relevant concepts such as truthfulness from language‑model activations, but probe accuracy may reflect only decodability, not causal influence on model behavior. The authors show that probe weight geometry alone cannot identify the features the model actually uses, because geometrically aligned features need not be causally relevant. They introduce a sparse‑autoencoder (SAE) decomposition that ranks features by probe alignment and gradient sensitivity, and demonstrate that ablating shared, probe‑only, and random feature sets reveals a sharp dissociation: shared features drive model output changes far more than probe‑only or random features, confirming that causal relevance requires intervention beyond weight geometry.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Machine Learning
Jul 15

From Geometric Recovery to Causal Validation: A Reproducible Audit of Sparse Autoencoder Features, from Superposition Geometry to Causal Inertness

arXiv:2607. 12166v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) are the standard for decomposing superposed neural representations into interpretable features, and evaluation relies predominantly on correlational recovery metrics -- cosine similarity between ground-truth directions and decoder atoms.

By Mohamed Abdessalem Bal
efficiencybenchmarkssafety
More like this →
arXiv Computation and Language
2d ago

Geometry of Confabulation and Refusal in Small Language Models

The paper introduces ConfabQA, a 784‑item benchmark designed to distinguish between confabulation and refusal in small language models when faced with unanswerable questions. Using linear probes, a sparse autoencoder, and direct optimization of intervention directions, the study maps the hidden‑state geometry on models such as Qwen3‑1.7B, Gemma 2‑2B, and Llama 3.2‑3B, showing that certain hidden‑state directions can flip confabulations into refusals. The work also demonstrates that the refusal signal is model‑dependent and that the hidden state’s contribution can be quantified beyond prompt features.

By Vladimir Kazei
llmsbenchmarks
More like this →
arXiv Machine Learning
Aug 4

Sign-Aware Gated Sparse Autoencoders: Modeling Anticorrelated Features with Bi-Jump-ReLU Activations

arXiv:2605. 28149v2 Announce Type: replace Abstract: Sparse Autoencoders (SAEs) extract interpretable features from Large Language Model activations, but standard variants enforce non-negative latents, so a bidirectional semantic axis (e.

By Bartosz Wieciech, Zmnako Awrahman, Marcin Czelej, Victor Hugo Jaramillo Velasquez, Wioletta Stobieniecka
llmsefficiencysafety
More like this →