ARENA Program Progress Tracker(in progress)

Treylon Wofford | Started: ____________ | Target completion: ____________

Legend: ⬜ Not started / In progress → ✅ Completed (swap the emoji as you finish each module)

Week 1 — Transformer Interpretability: Foundations #

Chapter 1, Part 1: Build the core toolkit

• ⬜ 1.1 — Transformers from Scratch: Build a transformer from scratch and load pretrained GPT-2 weights
• ⬜ 1.2 — Intro to Mech Interp: Learn TransformerLens to extract activations, apply hooks & find important attention heads
• ⬜ 1.3.1 — Linear Probes: Train linear probes to detect deception in a model playing the game Coup
• ⬜ 1.3.2 — Function Vectors & Model Steering: Steer model behaviour using activation interventions and the nnsight library
• ⬜ 1.3.3 — Interpretability with SAEs: Use SAEs to decompose LLM activation space, monitor cognition & steer behaviour
• ⬜ 1.3.4 — Activation Oracles: Implement activation oracles to reveal hidden knowledge and uncover forward-predictions

Week 1 notes:

Week 2 — Transformer Interpretability: Circuits & Classic Results #

Chapter 1, Part 2: Reverse-engineering real models

• ⬜ 1.4.1 — Indirect Object Identification (IOI): Reverse-engineer the IOI circuit in GPT-2 small following “Interpretability in the Wild”
• ⬜ 1.4.2 — SAE Circuits: Apply SAEs to circuit analysis, decomposing computations and tracing features through layers
• ⬜ 1.5.1 — Balanced Bracket Classifier: Reverse-engineer the algorithm learned by a bracket-balancing transformer
• ⬜ 1.5.2 — Grokking & Modular Arithmetic: Discover Fourier circuits in modular arithmetic models and observe grokking in action
• ⬜ 1.5.3 — OthelloGPT: Investigate emergent world representations in a GPT model trained on Othello games
• ⬜ 1.5.4 — Superposition & SAEs: Replicate Anthropic’s superposition paper and train SAEs to recover features
• ⬜ Bonus — Monthly Algorithmic Problems: 7 algorithmic challenges to test your interpretability skills in hackathon format

Week 2 notes:

Week 3 — Reinforcement Learning #

Chapter 2: From fundamentals to RLHF

• ⬜ 2.1 — Intro to RL: RL fundamentals: MDPs, policies, value functions, and multi-armed bandits
• ⬜ 2.2 — DQN & VPG: Implement DQN and Vanilla Policy Gradient for CartPole and beyond
• ⬜ 2.3 — PPO: Build a PPO agent from scratch and train it to master CartPole
• ⬜ 2.4 — RLHF: Implement RLHF end-to-end, applying PPO to language model finetuning
• ⬜ 2.5 — MCTS & AlphaZero: Implement MCTS and AlphaZero to train agents for complex games

Week 3 notes:

Week 4 — LLM Evaluations #

Chapter 3: Threat models, datasets, and agents

• ⬜ 3.1 — Intro to Evals: Design threat models and specifications for evaluating model properties
• ⬜ 3.2 — Dataset Generation: Use LLMs to generate and refine high-quality evaluation datasets
• ⬜ 3.3 — Running Evals with Inspect: Run standardised LLM evaluations using UK AISI’s Inspect library
• ⬜ 3.4 — LLM Agents: Build LLM agents with scaffolding to play Wikipedia Racing and other tasks
• ⬜ 3.5 — AI Control: Learn to monitor and control AI systems in a simulated environment

Week 4 notes:

Week 5 — Alignment Science #

Chapter 4: Frontier research topics

• ⬜ 4.1 — Emergent Misalignment: Study emergent misalignment in finetuned models
• ⬜ 4.2 — Science of Misalignment: Two case studies in black-box investigation to understand and characterize seemingly misaligned behaviour
• ⬜ 4.3 — Interpreting Reasoning Models: Apply interpretability techniques to chain-of-thought reasoning models
• ⬜ 4.4 — LLM Psychology & Persona Vectors: Explore persona vectors and psychological properties of language models
• ⬜ 4.5 — Investigator Agents: Use AI agents for investigating model behaviours (including petri & bloom)

Week 5 notes:

Progress Summary #

Website-Ready Highlights #

Move a module here once completed, so you always know what you can honestly claim on your site: