Sitemap

A list of all the posts and pages found on the site. For you robots out there, there is an XML version available for digesting as well.

Pages

About •

Research scientist applying experimental physics methodology to AI.

CV •

Research scientist applying experimental physics methodology to AI

Projects •

Research scientist applying experimental physics methodology to AI

Posts

Training Dynamics of Transformer Attention Heads

15 minute read

Published:

A time-dependent study of W_QK statistics across training checkpoints in the Pythia model suite: how spectral structure, stable rank, and head diversity evolve during pretraining.

Singular Value Structure of Transformer Attention Heads

20 minute read

Published:

An empirical study of the singular value spectra of W_QK matrices across transformer architectures: spectral distributions, participation ratios, and what they reveal about learned attention geometry.

portfolio

publications

Publications •

Published in , 1900

Publications

The following is a selection of my publication record. For the complete list please see:

talks

teaching