sharing an early conversation i had with k3 where i asked it to make a 3b1b-style video explaining its own arch. tbh this is my first time seeing a model use its vision skills like this (might as well explain its esp. high perf on frontend design) more concretely: i attached

nathan chen @nathancgy4 ·
sharing an early conversation i had with k3 where i asked it to make a 3b1b-style video explaining…

更多来自 @nathancgy4

  1. #1

    pretraining & scaling were here all along :) (or what fun will there be!)

    nathan chen @nathancgy4 ·
    pretraining & scaling were here all along :) (or what fun will there be!)
  2. #2

    man ignore the benchmarks and the race i just want there to always remain a frontier level intelligence model that is open and safe, available to everyone regardless of region it…

    nathan chen @nathancgy4 ·
    man ignore the benchmarks and the race i just want there to always remain a fro…
  3. #3

    what it feels like to have an upcoming math test in 6 hours yet finally setting off your first-ever 48b model training run 2am midnight

    nathan chen @nathancgy4 ·
    what it feels like to have an upcoming math test in 6 hours yet finally setting…

查看 @nathancgy4 的全部帖子