unembedding

Here are 2 public repositories matching this topic...

hematteo / learning-to-read-out

Code for "Learning to Read Out: Unembedding Dynamics in Language Model Pretraining" (NeurIPS 2026, arXiv:2610.21025). Parameter-trajectory crosscoders on the unembedding across pretraining checkpoints.

language-models pythia sparse-autoencoders interpretability mechanistic-interpretability olmo crosscoders pretraining-dynamics unembedding

Updated Oct 17, 2026
Python

hematteo / sparse-readout-prism

Star

Companion code for the paper "Sparse Readout Prism: Explaining Logit-Lens Scores in Features Instead of Tokens" (arXiv:2609.01936). Pretrained dictionaries: hf.co/hematteo/sparse-readout-prism

language-models sparse-autoencoders interpretability mechanistic-interpretability logit-lens unembedding

Updated Sep 13, 2026
Python

Add this topic to your repo

To associate your repository with the unembedding topic, visit your repo's landing page and select "manage topics."

Learn more