arXiv AI By Matteo He, William F. Shen, Xinchi Qiu, Nicholas D. Lane

Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens

Read the original on arXiv AI →

The paper introduces Sparse Readout Prism (SRP), a method that decomposes a language model’s readout matrix into sparse features, allowing logit‑lens scores to be expressed as sums of feature contributions. SRP reveals that lens readings depend on the corpus used to fit the readout, a phenomenon called corpus conditionality, and that the dominant readout feature remains stable across different corpora. By replacing the original readout with SRP’s sparse approximation, the authors recover 8.9–17.3 percentage points more of the tested logit differences than six geometric‑relation baselines, and ablating features shifts logit differences proportionally to their SRP contributions.

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.

Hugging Face Trending Papers
Aug 10

Interpreting Language Model Hidden States at Scale

Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network. Trained lenses remain expensive: affine-translator parameters grow quadratically with model width, while exact, full-vocabulary Kullback--Leibler (KL) training dominates memory.

arXiv AI
Aug 12

Interpreting Language Model Hidden States at Scale

arXiv:2608. 10260v1 Announce Type: new Abstract: Lens methods interpret large language models (LLMs) by mapping intermediate activations to the output vocabulary, revealing how next-token predictions develop through the network.

By Jordan Pettyjohn, Mansi Sakarvadia, Nathaniel Hudson, Daniel McKenzie, Kyle Chard, Ian Foster
arXiv AI
Jul 10

LEXIC: Lightweight Eye-tracking eXtension via Injected Complexity

arXiv:2607. 08152v1 Announce Type: cross Abstract: On the recent EyeBench benchmark, predicting reading comprehension from eye movements exposes a stark gap: text-aware models using pretrained language models reach 56--63% AUROC, while gaze-only models operate at chance.

By Sumin Lee, Kyeonghun Kim, Subeen Lee, Jiwon Yang, Tien Nguyen, Ken Ying-Kai Liao, Nam-Joon Kim