implementation detail · filed under optimization
Per-head splitting of attention and GDN input projections
Splitting attention and GDN input projections at per-head granularity; the evidence does not establish that this is identical to Muon Split.
- source
- 1
- model
- 1
- lab adopt it
- 1
- strongest
- used
How sources treat it
One count per evidence span, weakest treatment to strongest.
used 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
The and GDN input projections are split at per-head granularity, which improves both loss and downstream benchmarks.
usedoptimizationin Qwen3.8-NextQwen
Filed alongside
Other methods under optimization :: optimizer.
MuonPer-Head MuonAdamWCategory-specific assignment of Muon and AdamWSplit fused gradients before orthogonalizationNesterov momentumSinkhorn-balanced updateAdam without weight decay for the N-gram embedding tableAdamW for attention and GDN output gatesAdamW for gated-residual low-rank projectionsAdamW for input embeddings and output headAdamW for the MoE routerAsynchronous Micro-Group pipelineCanzonaCUDA graph capture of the optimizer stepEight-step Newton–Schulz iterationHybrid Muon and AdamW parameter-group optimizer assignmentHybrid Newton–Schulz iterationsHybrid optimization with Muon and AdamMoonlight-style learning-rate scalingMuon orthogonalization accuracy refinementMuon restricted to two-dimensional linear-map weightsMuon SplitMuown