Hacker News new | ask | show | jobs
by karencarits 1349 days ago
"Default" makes it sound like a deliberate decision or setting, but that is not how these models work. But I guess it would be trivial to actually make a setting to autmatically add specific terms (gender, race, style, ...) to all prompts if that is a desired feature
1 comments

Please no. I am all for neutrality, but the underlying cause is the training dataset. Change that if you want different results, but do not alter artificially.