David Johnston comments on Trying to Make a Treacherous Mesa-Optimizer

David Johnston 9 Nov 2022 19:56 UTC
3 points
0
Do you have code available? Just want to better understand the 4 properties of your deceptive agent.
- MadHatter 9 Nov 2022 20:00 UTC
  2 points
  0
  Parent
  Yes, it’s linked in the text, but here it is again: https://colab.research.google.com/drive/1ABVSMi0oCQP_daml-ZdWaVoKmwhQpTvI?usp=sharing
  - David Johnston 9 Nov 2022 20:10 UTC
    1 point
    0
    Parent
    Thanks, I see it now