Skip to content

Learning to Read Out: Unembedding Dynamics in Language Model Pretraining

Oct 2026 · 0 citations
Computer Science

TL;DR

Early in training, where many developmental transitions are dated, a behavioral date can reflect when the readout learns to express task information and not when the model learns it, which implies that early in training, a behavioral date can reflect when the readout learns to express task information and not when the model learns it.

Abstract

When does a language model acquire a capability? Developmental studies usually answer this by tracking when models start producing the correct token. But that prediction passes through the unembedding matrix, or readout, which is itself still being learned during pretraining. We show that a model's hidden states can encode task information before its own readout expresses it, so timelines read from outputs conflate what the model encodes with what its readout expresses. To track the readout directly, we introduce parameter-trajectory crosscoding, which fits one sparse dictionary to every token's readout row at every checkpoint. Unlike crosscoders on activations, it needs only checkpoint weights. We use it to locate a narrow early window, 0.4-1.1% of the way through Pythia's pretraining, in which the readout reorganizes and WordNet word categories become decodable from its rows. By then, linear probes on the hidden states already decode task information that the model's own outputs do not yet show. To test whether such gaps depend on how fast the readout learns, we pretrain small models from scratch that differ only in the readout's learning rate. We find that raising it makes the readout develop earlier and the gap close sooner, with little effect on final loss. Together, these results imply that early in training, where many developmental transitions are dated, a behavioral date can reflect when the readout learns to express task information and not when the model learns it.

View source

Similar papers

#machine learning Preprint Nov 2026

A Circuit for In-Context Learning

In-context learning (ICL) lets a pretrained language model perform a task specified by demonstrations in its prompt, with no parameter updates. With arbitrary labels the model must acquire a mapping it cannot have memorized, so the learning happens within a single forward pass. Prior work found candidate parts but not...

Andrew Gao, William Gao · 0 citations
Preprint Nov 2026

A Closer Look at Language Model Introspection: Probing Information and Reporting

One way to study language-model introspection is to test whether models can identify where their activations have been perturbed. We train a simple linear classifier to identify the perturbed sentence from the model's activations, then compare its predictions with the model's own answers across pretraining and injectio...

D. Rubin, Vincenzo Timmel, Xu-Feng He et al. · 0 citations
#artificial intelligence Preprint Oct 2026

Learned Readout Tokens for Extracting Large Language Model Confidence

Large language models (LLMs) are increasingly used within everyday applications where their responses have significant downstream impacts. For responsible deployment of such systems, we desire insight into how much we can trust their individual outputs. A popular approach for estimating LLM confidence is to train probe...

Nima Seifi, M. Ester · 0 citations
#natural language process... Preprint Oct 2026

HeadEdit: Calibrating Language Model Behavior Through the Frozen Unembedding Matrix

Alignment does not eliminate behavioral errors in language models. Models may still refuse benign requests, call unnecessary tools, or yield to false user claims. Current methods mitigate such errors as a computation problem, and rarely explore if the desired behavior is already encoded in the model's representation. M...

Zirui He, Hai-Yan Zhao, Jing-Yu Hu et al. · 0 citations

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.