• WanderingThoughts@europe.pub
    link
    fedilink
    English
    arrow-up
    2
    ·
    1 day ago

    A lot of the improvement came from finding ways to make it bigger and more efficient. That is running into the inherent limits, so the real work with other models just started.

    • Natanael@infosec.pub
      link
      fedilink
      English
      arrow-up
      3
      ·
      1 day ago

      And from reinforcement learning (specifically, making it repeat tasks where the answer can be computer checked)