Top K? How can I set it?

I understand that the API allows us to specify for Top P, but how about Top K? Is there a default Top K that is already implemented? Or can we change Top K by deriving it from Top P?

There’s no top_k parameter at this time, but I’d love to know why you’d want to use Top K.

Hi Joey! I was hoping to recreate the methods described in the GPT-2 Release Strategies report to detect GPT-3 generations. This is to help detect potential misuses of GPT-3. The method described in the report used Top-K as a restriction in order to train their discriminator. Thank you.

Interesting! I’m not sure of any way to use top_k here.