List your company

Language Modeling on Public Pretraining Dataset (train)

1.33LossAdam
1.329681.331841.3341.33616Apr 10, 2026

Evaluation results

MethodLinks
AdamModel scale=3B2026.041.33
AdamModel scale=1B2026.041.331
NexusModel scale=1B2026.041.338
NexusModel scale=3B2026.041.338