From Muon to Gradient Clipping: Some Thoughts on QK Stability

Other: Optimization Techniques(mastergodzilla.github.io)view on HackerNews
Muon optimizerQK weight matricesTransformeroptimizationstabilitygradient descentspectral normpseudoinversebilinear attentionlow-rank approximation.

Author: Eridanus2

Date: 7/13/2026

Article Summary:
A researcher explores the limitations of the Muon optimizer when applied to QK (Query and Key) weight matrices in Transformers, and proposes a modified update rule that follows Muon's philosophy while addressing the stability issues.