Copilot Performs Better on Untyped Languages Than Other Languages

In terms of performance on different programming languages, we have seen that copilot seems to perform better on JavaScript and Python than other languages. Acceptance rate is one of our key metrics. That's what fraction of the suggestions that copilot shows does the user accept. We look at a breakdown by language and sometimes we see that even less popular languages sometimes have a higher acceptance rate than the mean or the median. Not sure why, but someone asked this a while back. They had assumed that copilot wouldn't perform well on Haskell because there's probably not a lot of Haskell code in the training set. I went and looked and actually copilot performs better than average on Haskell. And we don