Skip to content

关于并行策略的咨询 #1

Description

@HOU-LONG

请问,该项目是否对标Megatron,支持上下文并行、专家并行等策略呢?您是否在1M的长上下文数据中训练过序列模型呢?

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions