Anthony Hughes

Message

Working on the science and engineering of oversight and monitoring.

Homepage: https://anthonyhughes.github.io/

GitHub: https://github.com/anthonyhughes/

X: https://github.com/anthonyhughes/

Anthony Hughes

Working on the science and engineering of oversight and monitoring.

Homepage: https://anthonyhughes.github.io/

GitHub: https://github.com/anthonyhughes/

X: https://github.com/anthonyhughes/

We need good evals for activation faithfulness

Tl;dr: There should be a clear, well-designed eval on activation faithfulness: whether the model activations a monitor reads remain a trustworthy surface when the model suspects/knows it is being monitored. We provide an appendix with preliminary results towards such an eval in case people want to build off of it,...

Sep 11•28

Can we find whether models have been backdoored?

This is the first post in a two-part sequence regarding the state of defending from data poisoning attacks. We describe some methods for determining whether a model has been backdoored and how to find the trigger. In the next post, we will discuss all of the ways we think the...

Jul 6•22