https://transformer-circuits.pub/2022/toy_model/index.html Toy Models of SuperpositionNelson Elhage, Tristan Hume, Catherine Olsson, Nicholas Schiefer, Tom Henighan, Shauna Kravec, Zac Hatfield-Dodds, Robert Lasenby, Dawn Drain, Carol Chen, Roger Grosse, Sam McCandlish, Jared Kaplan,
What I Watch: How LLMs store facts
How might LLMs store facts | Chapter 7, Deep Learning3Blue1BrownAug 31, 2024 “Unpacking the multilayer perceptrons in a transformer, and how they may store facts”
What I Watch: compare high dimensional vectors
A new way to compare high dimensional vectorsTunadorableAug 26, 2024 “Surpassing Cosine Similarity for Multidimensional Comparisons: Dimension Insensitive Euclidean Metric (DIEM)”
What I Read: passively learned, causality
What can be passively learned about causality?Simons InstituteAndrew Lampinen (Google DeepMind)Jun 25, 2024 “What could language models learn about causality and experimentation from their passive training?”