arXiv:2602. 10352v2 Announce Type: replace-cross Abstract: Self-interpretation methods prompt language models to describe their own internal states, but remain unreliable due to hyperparameter sensitivity.
Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.
arXiv:2602. 11852v2 Announce Type: replace Abstract: While state-of-the-art language models (LMs) surpass most humans in certain domains, their reasoning remains largely opaque, reducing trust and increasing the risk of deception and hallucination.
arXiv:2606. 26620v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) have emerged as a powerful tool for decomposing superposed language model representations into sparse and interpretable features.
arXiv:2608. 07594v1 Announce Type: cross Abstract: Interpretability is often treated as a tax on capability: language models are trained as opaque systems, then explained after the fact, with methods whose reliability is difficult to establish.
Sparse autoencoders (SAEs) are proposed to extract numerous features from large language model (LLM) representations, yet explaining these features still relies primarily on external observation. This reliance leads to superficial explanations inferred from observed model behavior and computational inefficiency from collecting such behavioral evidence at scale.
arXiv:2601. 22108v2 Announce Type: replace-cross Abstract: Continued pretraining is optimized with fixed self-supervised tasks but selected by downstream performance, creating a coarse feedback loop in which practitioners evaluate checkpoints, change data mixtures or objectives, and restart runs, while individual updates remain blind to target capabilities.
arXiv:2607. 20652v1 Announce Type: cross Abstract: Language models are thought to exhibit the phenomenon of superposition, representing many more features than dimensions in their residual streams.