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

    Neuronpedia

    INDEX
    Explanations

    in

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     rationale
    -0.07
    仰
    -0.07
    ’B
    -0.07
    _EDIT
    -0.07
    Reuse
    -0.06
    trand
    -0.06
     unve
    -0.06
    acet
    -0.06
    <w
    -0.06
     DM
    -0.06
    POSITIVE LOGITS
    0.07
    ',(
    0.06
    ("\
    0.06
    ügen
    0.06
     توسط
    0.06
    uitable
    0.06
    much
    0.06
    .."
    0.06
    ivation
    0.06
     +(
    0.06
    Act Density 0.003%

    No Known Activations