x

AI ALIGNMENT FORUM
AF

Liam Carroll

Ω131000

Mathematician, musician, hiking guide. Researcher residing in Australia jointly funded by Timaeus and the Gradient Institute. I am focused on understanding the generalisation properties of models based on Singular Learning Theory and Developmental Interpretability and relating this to alignment.

Website: https://www.liamcarroll.au/

Sequences

Posts

Sorted by New

Wikitag Contributions

Comments

Sorted by

Liam Carroll — AI Alignment Forum

No Comments Found

No wikitag contributions to display.

Distilling Singular Learning Theory

39Stagewise Development in Neural Networks

2y

1

33Growth and Form in a Toy Model of Superposition

2y

0

16DSLT 4. Phase Transitions in Neural Networks

2y

0

14DSLT 3. Neural Networks are Singular

2y

0

14DSLT 2. Why Neural Networks obey Occam's Razor

2y

0

29DSLT 0. Distilling Singular Learning Theory

2y

0

22DSLT 1. The RLCT Measures the Effective Dimension of Neural Networks

2y

0