optimizers · microsoft
Spike-Aware AdaM — clips gradient spikes preventing training instability in LLMs.
| Weight file SHA-256 | c66f41cd29e113c023f0a1aae9d15af387bb77ef9418521e59ee9ddfd56fec7c |
|---|
| Publisher | microsoft |
|---|---|
| Origin | United States |
| Licence | MIT |
| Pulls recorded here | 7,002 |
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.