Skip to content

Directly locating activations instead of tracing #1

Description

@luolearning

Hi — I saw you're doing residual stream tracing and token-level activation analysis.

I’ve been experimenting with a small plug-in that makes internal activations directly locatable, rather than inferred through tracing or delta comparisons.

Curious if this connects to what you're seeing.

https://github.com/luolearning/luoshu_kit

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions