René's URL Explorer Experiment


Title: Alex Makelov

Description: Personal website of Alex Makelov - researcher in mechanistic interpretability of large language models.

direct link

Domain: amakelov.github.io

Links:

Prof. Aleksander Madryhttps://madry.mit.edu/
Prof. Salil Vadhanhttps://salil.seas.harvard.edu/
mandalahttps://github.com/amakelov/mandala
Google Scholarhttps://scholar.google.com/citations?user=haO4sKoAAAAJ&hl=en&oi=ao
Semantic Scholarhttps://www.semanticscholar.org/author/Aleksandar-Makelov/17775913
Twitterhttps://x.com/AMakelov
Persona Features Control Emergent Misalignmenthttps://arxiv.org/abs/2506.19823
Sparse Autoencoders Match Supervised Features for Model Steering on the IOI Taskhttps://openreview.net/attachment?id=JdrVuEQih5&name=pdf
AI Alignment Forum posthttps://www.alignmentforum.org/posts/zj3GKWAnhPgTARByB/saes-discover-meaningful-features-in-the-ioi-task
Towards Principled Evaluations of Sparse Autoencoders for Interpretability and Controlhttps://arxiv.org/abs/2405.08366
mandala: Compositional Memoization for Simple & Powerful Scientific Data Managementhttps://amakelov.github.io/scipy-mandala.pdf
Is This the Subspace You Are Looking for? An Interpretability Illusion for Subspace Activation Patchinghttps://arxiv.org/abs/2311.17030
Backdoor or Feature? A New Perspective on Data Poisoninghttps://openreview.net/pdf?id=4NT3umNU3D0
Towards Deep Learning Models Resistant to Adversarial Attackshttps://arxiv.org/abs/1706.06083
Expansion in Lifts of Graphshttps://dash.harvard.edu/bitstream/handle/1/14398532/MAKELOV-SENIORTHESIS-2015.pdf?sequence=1
Practical dependency tracking for Python function callshttps://amakelov.github.io/blog/deps/
Mandala: Python programs that save, query and version themselveshttps://amakelov.github.io/blog/pl/

Viewport: width=device-width, initial-scale=1


URLs of crawlers that visited me.