ayabraham.com
AYODELE
AI Safety · Mech. Interp.AI Auditing · Alignment
AI Safety
Mechanistic Interpretability
AI Auditing

Projects


Research · Education · Software
AI Safety Research
Apart Research Hackathon · 2026
Under Review

Secret Loyalty Auditing

Submitted to the Apart Research AI Safety Hackathon. The project investigates whether hidden objectives in synthetic LLM organisms can be detected through mechanistic interpretability techniques. Specifically, it evaluates how auditor affordance levels influence the linear decodability of hidden objectives from transformer activations, and explores lightweight auditing protocols based on activation probing.

Hidden ObjectivesAdversarial PromptingLLM EvaluationAI Auditing
Independent · 2026
Ongoing

Independent Mechanistic Interpretability Research

Self-directed research into transformer circuits and feature geometry. Currently studying superposition, how models represent more features than they have dimensions, using toy model experiments and sparse autoencoder analysis. Exploring implications for feature identification and circuit-level auditing.

Transformer CircuitsSuperpositionSparse AutoencodersFeature Geometry
Education & Community
Self-study · 2025–26
In Progress

ARENA Curriculum

Working through the Alignment Research Engineer Accelerator (ARENA) curriculum. Covers transformer architecture from scratch, mechanistic interpretability techniques, reinforcement learning from human feedback, and agent foundations. Self-paced, with implementation exercises in PyTorch.

Mechanistic InterpretabilityRLHFTransformer Architecture
Lens Academy · 2026
Active

Lens Academy, Course Navigator

Volunteering as a course navigator at Lens Academy, helping participants work through structured AI safety and alignment learning pathways. Working at the intersection of safety research and education, making interpretability and alignment concepts accessible to newcomers.

AI Safety EducationAlignmentMentorship
Open Source · 2025
Open Source

Lerna

A personal AI curriculum architect. Lerna generates structured, personalised learning paths for technical topics using language models. Built to support self-directed learning in fields like mechanistic interpretability where reading lists and curricula are evolving rapidly.

AI ToolingEducationOpen Source
Software
Product · 2024
Live

meBible

An AI-powered Bible study and spiritual growth application. Designed to help users engage with scripture through structured study plans, contextual commentary, and personalised reflection prompts.

AI ApplicationMobileSpiritual Tech