List your company

Language Modeling on Clustered Pre-training Data Cluster 8

14.96Perplexity (PPL)Split MoE
14.9565614.9797815.00315.02622Mar 19, 2026

Evaluation results

MethodLinks
Split MoEModel Variant=Split MoE2026.0314.96
BaseModel Variant=Base2026.0315.046