MondSemmel comments on New Scaling Laws for Large Language Models

MondSemmel 3 Apr 2022 13:50 UTC
7 points
Typos:
- trained a 70-billion parameter model (“Chinchilla”) using the same compute as had been used for the 280-parameter Gopher. → 280-billion parameter Gopher
- Number go up → goes up
- Pattern 9 Apr 2022 18:33 UTC
  4 points
  0
  Parent
  Not sure why this was downvoted.
  - Lucas Adams 7 Sep 2022 22:25 UTC
    4 points
    0
    Parent
    Number go up is a meme not a typo