Pivotal Token Search

Adversarial Manipulation of CoT

Mechanistic analysis of Chain-of-Thought faithfulness using GPT-2 Small

Analysed determinism, faithfulness, reasoning patterns, & steering. Developed and tested methods to enhance control and fail-safes

Implementation and analysis of Sparse Autoencoders for neural network interpretability research. Features interactive visualization dashboard and W&B integration.

mech-interp suite for Granite4 models that use Mamba-2 architecture

Collection and learnings of my journey in Artificial Intelligence

Official implementation of the 'Uncovering Competency Gaps in Large Language Models and Their Benchmarks' paper

All code, stimuli, and results for a mechanistic interpretability study investigating how large language models internally represent emotional content

Unofficial implementation to reproduce the experiments from "Superposition as a Phase Change" of "Toy Models of Superposition".