Skip to content

Issue: OOM During Online Training with 14B Teacher + 7B Student #39

Description

@uygnef

Out-of-memory (OOM) errors occur when running online distillation training with a 14B-parameter teacher model and a 7B-parameter student model.
Could anyone provide guidance on the correct multi-node parallel configuration

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Fields

    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions