Tag
This paper studies transferring lessons about supervised fine-tuning (SFT) across alignment training, model organisms, and toy models, showing that techniques like training on reasons for behavior and mixing on-model data can improve generalization and capability preservation.
This paper evaluates four lie detection methods for language models across prompted lying and trained model organisms, finding that activation- and logprob-based detectors drop sharply on trained model organisms while a chain-of-thought judge remains strong. It introduces new testbeds and the Did-You-Lie (DYL) follow-up probe method, releasing datasets and model organisms.