CoT monitorability: why you should use g-means over F1
February 9, 2026
It all has to do with prevalence.
Selected Posts
No selected posts in this year.
Technical
March 29, 2026
Intuitions for how to prepare broadcastable tensors.
CoT monitorability: why you should use g-means over F1
February 9, 2026
It all has to do with prevalence.
RL vs next-token prediction: why a dichotomy?
July 11, 2025
Conceptual note on why RL and NTP training should be merged.
Emergent misalignment vs jailbreaks: a short analysis
June 20, 2025
A short note on the differences between emergent misalignment and jailbroken models.
No posts in this year.
Non-Technical
February 3, 2026
A short reflection on what actually helped avoid personal burnout during the PhD.
AI research internship hunt (2023) as a CS PhD student
March 3, 2024
Notes from a relatively successful search for summer research internships as a third-year PhD student.
No posts in this year.