CS Student @EPFL Interested in AI safety, mostly interpretability
Current theme: default
Less Wrong (text)
Less Wrong (link)