mosaicml / llm-foundry

LLM training code for Databricks foundation models
https://www.databricks.com/blog/introducing-dbrx-new-state-art-open-llm
Apache License 2.0
3.99k stars 525 forks source link

Configurable submesh #1236

Closed dakinggg closed 4 months ago

dakinggg commented 4 months ago

Makes the submesh for MoE + FSDP more configurable