optimizers · stanford
Second-order clipped optimizer using a light Hessian diagonal estimate. ~2x faster pretraining vs AdamW.
| Weight file SHA-256 | 281cc06f91910e51a3e02c494cb58bcfa39029fb551dc4a532c66027451f08b1 |
|---|
| Publisher | stanford |
|---|---|
| Origin | United States |
| Licence | MIT |
| Pulls recorded here | 13,503 |
Registry record as of 2026-06-06. Identity, licence and parameter count are properties of a release and do not move; tier can, and the verify link above re-checks it against the live registry rather than this page.