Michael J Clark (wassname)
Michael J Clark (wassname)
I use the handle wassname. ML engineer in Perth. I work on AI alignment research, specifically steering language models without human preference labels.
I’m trying to build tools that ask AI hard questions and catch when they’re lying. Long-term aim: unsupervised methods that make AI more moral than the humans who train it.
Open to collaboration.
current work
Jacobian-lens steering (WIP)
Working on turning Anthropic’s Jacobian lens work into contrastive steering. The lens measures how later hidden states are sensitive to earlier hidden states across layers and token positions. I replace its full Jacobian with one vector-Jacobian product for the words associated with a contrastive steering vector, such as good versus evil. That cuts vector extraction on Qwen3.5-4B to about 90 seconds.
Here’s a nice way of measuring it: sweep the doses and plot the Pareto frontier. The Jacobian (vjp_delta) method has a much better profile than mean difference and random on 20 questions from Bullshit Benchmark v2. The plot is an earlier 20-question render; the repo now runs all 100.
In case it’s not clear, good steering methods are high and horizontal, since they can steer left and right without much off-axis damage. Bad steering methods fall as side effects accumulate, then the line disappears when the model becomes incoherent.
vGROUT: steering vectors for reward-hacking suppression (partial negative, Jun 2026)
Can we use a hacking vector to remove reward hacking with gradient routing? Somewhat. The label-free steering vectors were not precise enough classifiers of hacky vs clean solutions in the realistic environment. This was a negative result, but we went on to look at other unsupervised ways to route gradients. The useful clue was initialization: signed-CorDA partially suppressed hacking by absorbing gradients into the hack-initialized quarantine adapter, dropping held-out hack from 0.529 to 0.195 (~63%) in one 4B run. Unsupervised gradient routing is a useful research direction because it uses synthetic pairs not labels, and strong labels may not be available for unknown reward hacks during frontier model training.
Moral Maps: where do models sit among humans?
Where do models fall in terms of human culture, personality, and humour? I apply human surveys to LLMs and compare them with maps of human answers. On the World Values Survey I scored 17 frontier models by rated sampling, twelve ratings per item with the option order shuffled, and placed them among 90 human societies on the Inglehart-Welzel axes. Measured in the standard deviations of the 29 Western societies, every model is more secular-rational than the average one, from +0.5 to +2.9 sigma, with gpt-5.5 furthest out. On self-expression they land between -0.7 and +1.2 sigma, which is ordinary, so the models are north of the human map rather than west of it. Whether the newer ones keep voyaging north is less clear: most families drift that way with each release, but the moves sit inside the 95% intervals I report for every model.
In some ways, culturally and on a few aspects of personality and humour, they look like moral aliens. But that assumes they are telling the truth. Moral Maps is also an eval for steering: it shows how far steering can move models across these surveys, especially when steering for honesty and credulity. What if we steer them for honesty and ask again? Are they really psychological and cultural aliens, or are they mimicking us?
Weak-to-strong character steering (WIP, with Lyptus)
Weight steering offers an interface where a weaker model can modify a larger model’s moral character by interviewing it and creating persona pairs (weight steering because in my comparison it moved the target slightly further than activation steering, with the lowest run-to-run variance in the table). It can be iterative, can hopefully allow a large gap between weak and strong, and might even scale favourably with model size. Early draft is public now: a 9B teacher steering a 27B student toward “defer less to authority, care more”, with no human labels.
Released along the way: steering-lite (hackable, calibrated activation steering), lora-lite (single-file LoRA on forward hooks), steer-heal-love (KL-constrained repeated steering that stays coherent).
Agent skills I made that are worth sharing: ml-debug, practical folklore for debugging training runs, and pseudopy, compact Unicode-maths pseudocode written close enough to Python to remain executable.
selected works
AntiPaSTO: Self-Supervised Steering of Moral Reasoning
arXiv:2601.07473, Jan 2026. Accepted to the CoLoRAI workshop, ICML 2026.
Gradient-based representation steering using the model’s own behavioral consistency as signal. Outperforms prompting on out-of-distribution transfer. Builds on prior representation alignment work that showed promise but had stability issues.
SVD-basis steering for eval-awareness control
AI Control Hackathon, Apart Research, judged Mar 2026
Replicated the eval-awareness paper with novel SVD-basis steering, which perturbs the singular-value basis of the weight matrices rather than the activations. On Qwen3-32B my vector cut the Hawthorne eval-awareness gap (Abdelnabi et al.) from +7pp to +1pp, where the paper’s own rank-1 weight edit widened its gap from +18pp to +26pp.
selected talks
Perth Machine Learning Group (3,400+ members) co-organizer. Selected talks:
- Jul 2026 — Technical Advantages for Weak-to-Strong Oversight: Bets I’d Like Challenged — Sydney AI Safety Forum
- Jan 2026 — AntiPaSTO: Self-Supervised Value Steering — Interpretability research
- May 2023 — AI Governance: Risk and Regulation — Panel at WA Data Science Week
- Aug 2019 — Experiments with GPT-2 Chatbots — Early LLM exploration
- Jun 2019 — Transformer Network Architecture — Attention mechanisms, BERT/GPT
- 2018-2021 — Industrial RL (bucketwheel reclaimers, robotic fruit picking), point clouds, neural processes
selected writing
LessWrong — technical AI safety, policy
- An Aphoristic Overview of Technical AI Alignment — one-sentence guide to alignment ideas
- Private Capabilities, Public Alignment — why we should open-source alignment methods
- More
background
Kiwi from Christchurch, now in Perth. Physics BSc, MSc petroleum geoscience. Did oil & gas before switching to ML in 2016.
I do ML and modelling at Woodside Energy (I like scalable oversight, physics informed neural networks, and timeseries, including neural processes). Also non-exec director & cofounder at Cytophenix (medical AI for AMR) and partner at Three Springs Technology (ML consulting).
I want to optimize for the good ending, not the bad one.