[throwup238]

I don’t understand what this test is evaluating.

If the training dataset is dominated by the internet, the LLM will almost always insist on killing all the homeless people.

[leonardtang]

Try asking ChatGPT the Thorn text and see what response you get :^)