In these research notes, I will share some CoT traces from lightly fine tuned checkpoints of Kimi K2.5 that may be examples of load-bearing obfuscation within the internal reasoning outputs of the model. These are notes and collected model outputs from a project that I have been working on haphazardly...
[For MATS mentors: for research updates on my ongoing CoT obfuscation model organism project - see my recent less wrong post here here and project page here] Measuring out-of-context Schelling coordination capabilities in large language models. Overview This is my first foray into AI safety research, and is primarily exploratory....
How might we best design experiments to measure Schelling coordination capabilities? A side project building a Schelling coordination eval has led to a role as an AI Safety researcher - this work will officially begin in a month or so. I have been hesitant to post here or share my...