Heykuki News

TopNewBestAskShowJobs
TopNewBestAskShowJobs
1.
Modded-NanoGPT: NanoGPT (124M) quality in 3.25B tokens (github.com/KellerJordan)
81 points
ocean_moist
2 years ago
11 comments
2.
Train to 94% on CIFAR-10 in 3.29 seconds on a single A100 (github.com/KellerJordan)
3 points
kjjnot
2 years ago
1 comment
3.
modded-nanogpt: NanoGPT (124M) in 2 minutes (github.com/KellerJordan)
2 points
tosh
3 months ago
discuss
4.
Muon Optimizer (github.com/KellerJordan)
2 points
pilooch
2 years ago
discuss