ryan_greenblatt comments on Alignment Faking in Large Language Models

[ryan_greenblatt]

The abstract says “We find the model complies with harmful queries from free users 14% of the time”, but in other places it says 12% - should it be 12%?

Thanks, this has been resolved.