TexMexBazooka

It's a serious problem because people that naturally write in AAE are most likely getting worse results

Person using LLM built on grammatical rules of the English language has subpar results when operating outside of those rules. More at 6.

LLM’s don’t have a bias against anyone, it’s literally just data. And those models are by and large fed with traditionally grammatically correct data. They don’t understand dialects, you’re looking soooooo hard for something to be offended over

What always happens when a dipshit with a business degree starts making decisions about technical problems

Huh, TIL

Every accusation is a confession. Always.

Did they have the skeleton PVP curse?

I wonder if it could somehow pull that data from somewhere like ground news somehow

The drunk kind

Pffffffffft

Cackling in the bar bathroom now, thanks

If some random instance with one user makes like 100s of communities, will it count in the graph?

Yeah it will, that’s the problem with using community count as a measure for the activity of a given instance.

Hexbear is interesting because like you said, they’re very self contained.

If you read the rest of that sentence, you’ll (maybe) figure out that there isn’t any comparative statements there- just pointing out that hexbear mostly originated from users banned from Reddit, for being toxic and disruptive.

There’s a few OC users there