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

    Neuronpedia

    INDEX
    Explanations

    Obscuring characters

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _FF
    -0.07
     gener
    -0.07
    oš
    -0.07
    '))
    -0.06
     repreh
    -0.06
     devoted
    -0.06
     територ
    -0.06
     haline
    -0.06
     WALL
    -0.06
    athing
    -0.06
    POSITIVE LOGITS
    andas
    0.07
    ($(".
    0.06
    [^
    0.06
     nút
    0.06
    centroid
    0.06
    ��
    0.06
     (:
    0.06
     CIM
    0.06
    erialized
    0.06
     •
    0.06
    Act Density 0.002%

    No Known Activations