Screenshot of this question was making the rounds last week. But this article covers testing against all the well-known models out there.

Also includes outtakes on the ‘reasoning’ models.

  • MangoCats@feddit.it
    link
    fedilink
    English
    arrow-up
    1
    ·
    13 hours ago

    The very interesting part will be how successful they are at training the training data selectors to choose high quality data sources.

    • Zos_Kia@jlai.lu
      link
      fedilink
      English
      arrow-up
      1
      ·
      4 hours ago

      I think a lot of it is still done by hand, and there is also synthetic data distilled from larger models of course.