Persona vectors (Chen et al. 2025) and emotion representations (Sofroniew et al. 2026) have been studied independently, but plausibly overlap in activation space. We'll measure their geometric and causal relationship to determine whether persona drift and emotional-state changes are mechanistically distinct failure modes; and whether interventions on one silently move the other.
About the project
Motivation. Two recent lines of interpretability work have identified linear structure governing model behavior: persona vectors, which capture character traits like sycophancy or malice (Chen et al., "Persona Vectors," arXiv:2507.21509), and emotion concept representations (Sofroniew et al., arXiv:2604.07729). Both are extracted with similar contrastive techniques, both support steering, and both are proposed as monitoring targets. But, so far, no one has made clear their relationship. When a model "becomes anxious," is that a persona shift, an emotion representation activating, or both–entangled in a shared subspace? If the directions overlap substantially, then persona-monitoring systems will produce false positives on transient emotional content, and emotion-steering interventions may cause unintended persona drift (or vice versa).
Core research questions. (1) What is the geometric relationship between persona directions and emotion directions extracted from the same model: cosine similarity, overlap of subspaces, effective dimensionality of the shared component? (2) Are the two causally separable: can we steer emotion while holding persona evals fixed, and steer persona while holding emotion probes fixed? (3) Do the relationships found in one model family generalize to others?
Plan. Weeks 1–3: replicate both extraction pipelines (persona vectors via contrastive system prompts and trait evals; emotion directions via contrastive datasets and probing) on a single open-weight model (e.g., Llama-3.1-8B or Qwen2.5-7B). Weeks 4–7: build controlled datasets that cross persona × emotion (e.g., an anxious helpful assistant vs. a calm sycophantic one), then analyze subspace geometry: pairwise similarities, principal angles between subspaces, variance explained by shared components. Weeks 8–10: causal experiments — steer along each direction and measure effects on both persona evals and emotion probes; test whether projecting out the shared subspace preserves the intended intervention. Weeks 11–12: replication on a second model and writeup.
Outputs. An Alignment Forum / LessWrong post at minimum; a workshop paper (e.g., an interpretability workshop at ICLR/NeurIPS) if results are strong. All code open-sourced.
Why this is tractable in 12 weeks. Both methods are published with clear recipes, the models are small enough for modest compute, and I am actively working on this question myself — mentees join a running project with existing code and hypotheses, not a cold start.
Theory of change
Persona drift is a documented failure mode of deployed assistants, and activation-level monitoring and steering are among our main proposed defenses. But monitoring and steering tools are only trustworthy if we know what the directions they use actually represent. If persona and emotion representations are entangled, safety interventions built on either will have unmeasured side effects: suppressing a "malicious persona" direction might also flatten legitimate emotional understanding, and emotion-based welfare monitoring might misfire on persona shifts. Clarifying this distinction directly improves the reliability of representation-based safety tooling. It also connects to my prior work on when activation-level detection does and doesn't support successful intervention (arXiv:2511.16699).
Your role
Mentees will be hands-on researchers. After a guided replication phase, each mentee will own a workstream (dataset construction, geometric analysis, or causal experiments) with substantial autonomy over implementation decisions. We'll make design decisions together in weekly meetings. Mentees will be co-authors on any resulting publication, with authorship order reflecting contribution.
Prerequisites
- Highly proficient in Python and comfortable with PyTorch.
- Hands-on experience with transformer internals: at minimum, completed substantial parts of the ARENA interpretability curriculum or equivalent (extracted activations, run hooks, done activation patching or probing on a real model).
- Comfortable reading ML papers and turning them into code without step-by-step instructions.
- (Nice to have, not required:) experience with TransformerLens or nnsight; prior steering-vector work.
Location preference
No geographic requirement. Must be available for a weekly team meeting compatible with US Eastern Time (roughly 9am–9pm ET).
Application question(s)
- Skim Chen et al., "Persona Vectors" (arXiv:2507.21509). Propose one concrete confound in their extraction method that could cause a "persona direction" to actually capture something else, and describe an experiment to test for it. (250 words)
- Suppose steering along an "anxiety" direction also increases sycophancy evals. Give two mechanistically distinct explanations for this result and how you'd distinguish them empirically. (200 words)
- Link to code you've written involving model internals (a repo, notebook, or ARENA exercise solutions).
About the mentor
I'm a philosophy PhD student at the University of Rochester working at the intersection of philosophy of mind and mechanistic interpretability. My research asks what it would take for interpretability claims to be epistemically robust. Before the PhD I did a BA in computer science, and I've spent the past year deep in the technical side: the ARENA curriculum through CBAI's CAMBRIA bootcamp, replicating activation-oracle results, and working with QK/OV circuits, superposition, and steering vectors. I also co-created HUMAN, a neurosymbolic character-training architecture, and am building Flourish, a behavioral virtue benchmark.
As a mentor I care about two things: rigor in what we claim our experiments show, and shipping. I've supervised undergraduates independently and as a TA, and run an AI safety group (CNY AI Safety). I'm especially excited to work with mentees who want to connect conceptual questions about model cognition to concrete interpretability experiments. To work with me, you don't need a philosophy background, just curiosity and willingness to engage, and write clean PyTorch.