Open Menu
AllLocalCommunitiesAbout
lotide
AllLocalCommunitiesAbout
Login

Apple AI Released a 7B Open-Source Language Model Trained on 2.5T Tokens on Open Datasets.

⁨158⁩ ⁨likes⁩

Submitted ⁨⁨3⁩ ⁨months⁩ ago⁩ by ⁨ModerateImprovement@sh.itjust.works⁩ to ⁨technology@lemmy.world⁩

https://www.marktechpost.com/2024/07/21/apple-ai-released-a-7b-open-source-language-model-trained-on-2-5t-tokens-on-open-datasets/

source

Comments

Sort:hotnewtop
  • A_A@lemmy.world ⁨3⁩ ⁨months⁩ ago

    They managed a substantial incremental improvement over previous models by first creating a better set of data as their starting point.

    source
  • JohnDClay@sh.itjust.works ⁨3⁩ ⁨months⁩ ago

    Is this the one for ‘research only’ that is trained on YouTube transcripts including mkbhd?

    source
    • SuckMyWang@lemmy.world ⁨3⁩ ⁨months⁩ ago

      As someone who knows nothing about this stuff, yes.

      source
      • cheese_greater@lemmy.world ⁨3⁩ ⁨months⁩ ago

        Happy cake day, Wang suck dude

        source
      • reddwarf@feddit.nl ⁨3⁩ ⁨months⁩ ago

        As someone who know f-all about AI I support your endorsement, if only by being impressed the training took 2.5 tokes so you know that AI smokes like a banger!

        source
  • iridium@piefed.social ⁨3⁩ ⁨months⁩ ago

    Hopefully they'll be able to put something together that can run locally, so they can finally stop using this "Open"AI bullshit

    source