On: Teaching GPT-3 to Identify Nonsense

[arram]

I briefly tested it with 'That doesn't make sense.' which worked, but switched back to yo because I'm looking at the probability of the first word and 'That' is too common to assume that the rest of the completion is 'doesn't make sense.'

[benlivengood → arram]

Maybe try 'Nonsense!' as the response?

[skybrian → arram]

It would be interesting to see how reliability changes based on the exact response you asked for.

[arram → skybrian]

My next experiment.

[lt → arram]

I'd try with a few different responses according to the type of nonsense question: "That doesn't make any sense", "How could I know that?"/"No one knows that", "That's a matter of opinion", etc.