Ideas & society2024 · shared on Hacker News
Grokfast: Accelerated Grokking by Amplifying Slow Gradients
Explore Grokfast: Accelerated Grokking by Amplifying Slow Gradients, its original source and Hacker News discussions. Discover community perspectives and save it to your reading list.
GO BEYOND THE ABSTRACT
The conversation around the idea.
Original Hacker News threads. Different perspectives, questions, and things worth thinking about.
Showing up to 100 related threads. A paper may appear in a comment rather than the original submission.
Yjohnsutor ·
Grokfast: Accelerated Grokking by Amplifying Slow Gradients
117 points39 comments