[chowells]

So I was talking to a friend about this, and he thought this was a parody because of all the obviously incorrect results you were highlighting in the Twitter thread. Other people have mentioned the \ escapes going missing, but my friend called out https://twitter.com/gd3kr/status/1545495732265766913 as so hilariously wrong that it couldn't have been anything other than a parody.

Is this really a parody, or is it just another example of people not actually reading GPT-3 output carefully enough to notice it's nonsense?

[gd3kr]

Hey, no it’s not a parody lmao. I shipped it because realised it could have at least some utility while building it. To be fair, the wrong examples would be an oversight on my part, but I thought of them being more of genuine benchmark of what GPT3 was capable of; sort of like an experimental feature. I’m working hard on refining the results by fine tuning DaVinci and getting the output as close to ideal as possible — I think there’s a lot of potential there. In the end, even boosting a user’s productivity marginally is a win in my book, and as a lot of people have pointed out, it’s already doing that to some extent.