Public KG record
MoE
Recorded overview
layer carries 512 experts, activating 10 routed plus 1 shared per token
An evidence-backed entity from the project knowledge graph; detailed tool metadata is still being added.
Recorded details
- Capabilities: Sparse activation of experts
- Use cases: MoE模型预训练和后训练
Evidence coverage
The KG currently records 8 known information dimensions and has 11 undocumented dimensions, linked to 19 public source articles. These counts describe record coverage, not model performance or verified fit.
Related public records
- DSpark — related to
- MoE模型预训练和后训练 — applied to
- DSpark — related to
- sparse routing — uses
- Clef-omni — uses