← Back to Optimizer

MuonEq-R

Optimizer
Used in
4 PRs
Best BPB
1.0066
Avg BPB
1.0634

Hyperparameters Across PRs

pr_numberweight_decaymomentumother_params
1326
1334{"row_normalized":true}
1485{"row_normalized_newton_schulz":true}
20710.095