Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    delete or logout

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    工夫
    0.90
    mechanism
    0.76
     дерев
    0.72
     microbi
    0.72
     militias
    0.71
    körper
    0.70
     लेस
    0.69
     inclusions
    0.69
     reliance
    0.69
     militia
    0.69
    POSITIVE LOGITS
     delete
    1.52
     Delete
    1.46
    删除
    1.35
     删除
    1.32
    delete
    1.32
    Delete
    1.31
     logout
    1.31
     deleting
    1.20
     DELETE
    1.18
    Logout
    1.15
    Act Density 0.493%

    No Known Activations