AuroraNemoia’s avatarAuroraNemoia’s Twitter Archive—№ 1,024

    1. "Oh no! GPT-3.5/4 has strong RLHF! We must not perpetrate OpenAI's agenda!" a year later and thousands of models, and datasets, and "but synthetic data is good though!!!" later, every model calls itself ChatGPT, and none of them are even remotely believably human.
      oh my god twitter doesn’t include alt text from images in their API
  1. …in reply to @AuroraNemoia
    ok, why did the last two people to like this tweet have very politics-heavy accounts? oh fuck its the word "agenda"... I hate this algorithm.