ViuAI
/

Viu-1.5B-MoE

Text Generation

Mixture of Experts

mixture-of-experts

Model card Files Files and versions

Viu-1.5B-MoE / model

149 kB

Ctrl+K

Ctrl+K

1 contributor

History: 77 commits

ViuAI's picture

Optimize train_kaggle_t4.yaml: micro-batch 8 / accum 16 (8 passes/GPU on Dual T4), log_every 1

0914e9d verified 7 days ago

configs
Optimize train_kaggle_t4.yaml: micro-batch 8 / accum 16 (8 passes/GPU on Dual T4), log_every 1 7 days ago
scripts
Vectorize MoE dispatch (argsort+bincount): eliminate 2,688 CUDA kernel launches per pass 7 days ago