Explorations into the proposal from the paper "Grokfast, Accelerated Grokking by Amplifying Slow Gradients"
MIT License
Published by lucidrains about 2 months ago
Published by lucidrains 4 months ago
Published by lucidrains 4 months ago
Published by lucidrains 4 months ago
Published by lucidrains 4 months ago
Full Changelog: https://github.com/lucidrains/grokfast-pytorch/commits/0.0.2
Published by lucidrains 4 months ago