Skip to content

What modules of the model are fine-tuned by the “Chart Specific Annealing Tuning” mentioned in the paper? #13

Description

@Chinouth2144
No description provided.

Activity

  1. Coobiw commented on May 29, 2025

    @Coobiw
    Contributor

    Hi, thanks for you attention. In chart-specific annealing tuning, the moe connector(including learnable router and the experts) and LLM LoRA are trainable.

  2. Coobiw commented on May 29, 2025

    @Coobiw
    Contributor

    Detailed settings are the same as sft.sh.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

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