Mechanistic interpretability: belief-state geometry in a transformer's residual stream. From-scratch replication of Shai et al. 2024 (arXiv:2405.15943).
-
Updated
Jun 25, 2026 - Jupyter Notebook
Mechanistic interpretability: belief-state geometry in a transformer's residual stream. From-scratch replication of Shai et al. 2024 (arXiv:2405.15943).
Hankel matrix analysis of belief state geometry in transformers - companion code for Hankel world-models blog series
Add a description, image, and links to the belief-states topic page so that developers can more easily learn about it.
To associate your repository with the belief-states topic, visit your repo's landing page and select "manage topics."