Jansen Nye
CS/ML undergrad working on mechanistic interpretability
Hey! I'm a senior at BYU studying computer science with a machine learning emphasis (graduating April 2027). At DRAGN Labs, I work with Dr. Nancy Fulda on language model interpretability, self-concept, and reward hacking. I spent summer 2026 as a software engineering intern at Adobe.
Research
Can the Assistant Axis or Schemer Axis Predict Reward-Hacking?
Both the Assistant Axis (Lu et al.) and my own hypothesized Schemer Axis show mixed results predicting reward-hacking in coding tasks on GPT-OSS-120B.
Read the studyWhat Do Self-Other Overlap Adapters Actually Do?
Across five models, I separate the collapse of representations at the SOO training layer from the adapter updates that actually change behavior. The behavioral effect lies in earlier layers, with distinct mechanisms across models.
Read the studySelected projects
Currently
Coursework
BYU- Computer Vision
- Mathematics of Data Science
- Probability & Statistics
- Data Science Capstone
Reading
Reach Out
You can reach me at jansennye989@gmail.com! I'm happy to chat anytime.