arXiv (All)
open access
Learning to Read Out: Unembedding Dynamics in Language Model Pretraining
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
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- Language Model Activations Inhabit Privileged Error-Correcting Basins#1199457
- Rethinking Memorization Mitigation in Diffusion Models: Reinforcing Text Conditioning#1156994
- From what they hear to what they say: The distribution of word classes in the output of young English-speaking children and their caregivers#1135412
- From what they hear to what they say: The distribution of word classes in the output of young English-speaking children and their caregivers#1135255
- Knowledge Acquisition During Pre-training? Large Language Models Learn Better With Auxiliary Views#639702
Record · ID 1356361
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive