specific method · filed under model architecture
SimpleGDN
A minimalist linearization strategy intended to reuse pretrained weights for continual-training adaptation; the evidence identifies it as an improvement over Gated DeltaNet but does not detail its mechanism.
- source
- 1
- model
- 1
- labs adopt it
- 0
- strongest
- evaluated
How sources treat it
One count per evidence span, weakest treatment to strongest.
evaluated 1
Documented in
Evidence
1 span quoted from the sources, strongest treatment first.
SimpleGDN: A minimalist linearization strategy designed for maximal reuse of pre-trained weights, improving upon GDN for continual-training adaptation.
evaluatedmodel architecturein GLM-5Z.ai
Filed alongside
Other methods under model architecture :: token mixer :: linear attention & state space :: gated delta network.
Gated DeltaNetKimi Delta AttentionGated DeltaNet–sparse MoE hybridKDAGated DeltaNet with bounded sigmoid output gateGated DeltaNet with reduced KV-head configurationHybrid linear attentionKimi Delta Attention and Attention Residuals architectureKimi Delta Attention with input-dependent full-rank output gateKimi Delta Attention with lower-bounded log-decay