log — all entries

Projects

A running log of research and engineering work, newest first. Earlier entries are smaller side projects from before I moved into AI safety evaluations full-time.

active

Aug 2026

Monitored vs. Unmonitored: A Honeypot Study of Instrumental Reasoning

A controlled experiment testing whether models take more shortcuts when they believe grading is unmonitored, using a fake grading server as a honeypot.

Inspect AIFlaskDockerClaude Sonnet 5

complete

May 2026

CyBench Elicitation Ablation

A 2×2×2 ablation across 2 models, 2 guidance levels, and 2 tool configs on 9 CyBench tasks — the real finding was that budget exhaustion, not incorrect reasoning, explained almost every failure.

Inspect AICyBenchDocker

complete

Mar 2026

Sandbagging Detection via Position Attractors

A 2×2×2 pilot extending Cacioli's 2026 position-attractor work to a frontier model, testing whether reasoning-permission changes whether sandbagging leaves a detectable signature.

Inspect AIGPQA DiamondARC-EasySciPy

complete

Aug 2025

Bystander Effect in LLM Agents

5,120 scenarios across 20 domains measuring whether LLM agents intervene less when other agents are present, drawing on the classic Darley & Latané paradigm.

Inspect AIdistilabelPython

archived

Nov 2022

Shakespearean Text Generator

A GPT built from scratch and trained on a Shakespearean text dataset to generate new text in Shakespeare's style.

PythonPyTorch

archived

Jun 2022

Harry Potter Fan-Fiction Generator

A Markov-chain text generator trained on Harry Potter text, deployed as a Flask API with a React frontend.

PythonFlaskReact

archived

Mar 2022

Character Script Generator

A fine-tuned GPT-3 model that generates user-defined crossover scripts between DC and Marvel characters.

OpenAI GPT-3Python

archived

Sep 2021

Predicting Live Facial Emotions

Real-time facial emotion detection from a webcam feed using computer vision.

PythonOpenCV