List your company

Language Modeling on Clustered Pre-training Data Cluster 7

15.972Perplexity (PPL)Split MoE
15.9399216.1564616.37316.58954Mar 19, 2026

Evaluation results

MethodLinks
Split MoEModel Variant=Split MoE2026.0315.972
BaseModel Variant=Base2026.0316.774