Slacker News Slacker News logo featuring a lazy sloth with a folded newspaper hat
  • top
  • new
  • show
  • ask
  • jobs
Library

Comment by andai

21 hours ago

What's the downside? Don't they stop when they hit diminishing returns?

2 comments

andai

Reply

hgoel  2 hours ago

Wouldn't the model start overfitting at some point? Degrading generalization for accuracy on the training set.

Ifkaluva  18 hours ago

You’d think so, but I haven’t seen it explicitly discussed in their papers, and nobody else that I know of trains on that many tokens

Slacker News

Product

  • API Reference
  • Hacker News RSS
  • Source on GitHub

Community

  • Support Ukraine
  • Equal Justice Initiative
  • GiveWell Charities