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

    Neuronpedia

    INDEX
    Explanations

    all layers of skin

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    </h2>
    0.43
     */
    0.40
     script
    0.40
    дь
    0.38
     দৃষ্ট
    0.38
    0.37
     Deals
    0.37
     Hook
    0.36
     vw
    0.36
    Delete
    0.36
    POSITIVE LOGITS
     nenhum
    0.44
     stabilise
    0.43
     стабили
    0.42
     waging
    0.41
     realiz
    0.40
    ື້ນ
    0.40
    <unused91>
    0.40
     expedi
    0.39
     imperatives
    0.38
     stabilizers
    0.38
    Act Density 0.000%

    No Known Activations