About - Michael Ivanitskiy

I’m a 5th year PhD candidate in Applied Mathematics, and my research is on mechanistic interpretability, primarily oriented towards AI safety. I’m currently most excited about Stochastic Parameter Decomposition, Understanding Attention Heads via pattern dynamics, and a particular kind of model reverse engineering.

Some of my previous work in mechanistic interpretability was on understanding search in toy models trained on mazes. I’ve also worked on projects in game theory, computational neuroscience, and computer vision. For more on my research, see here.

Outside of research, I like hiking, rock climbing, and fossil hunting. I am big fan of old sci-fi novels, particularly of the works of Arthur C. Clarke. I also have a number of open source projects, which are tooling for either ML research, personal knowledge management systems, or other random things.

Contact me:
mivanits 🇦​​​​​🇹​​​​​ mines 🇩​​​​​🇴​​​​​🇹​​​​​ 🇪​​​​​🇩​​​​​🇺​​​​
or:
{anything_you_want} 🇦​​​​​🇹​​​​​ miv 🇩​​​​​🇴​​​​​🇹 name

LinkedIn | GitHub | Google Scholar | ORCID