List your company

Language Modeling on Clustered Pre-training Data Cluster 9

15.405Perplexity (PPL)Split MoE
15.3922415.4783715.564515.65063Mar 19, 2026

Evaluation results

MethodLinks
Split MoEModel Variant=Split MoE2026.0315.405
BaseModel Variant=Base2026.0315.724