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

    Neuronpedia

    INDEX
    Explanations

    names like John, William, George, Thomas, Robert

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     eyeliner
    0.32
     deodor
    0.29
     checkboxes
    0.29
     backsplash
    0.28
     velcro
    0.28
     Sasuke
    0.28
     KPIs
    0.28
     Fitbit
    0.28
     MetaMask
    0.28
     eucalyptus
    0.27
    POSITIVE LOGITS
     John
    0.31
    and
    0.29
     William
    0.29
     George
    0.29
    an
    0.28
     Thomas
    0.27
    am
    0.27
     Robert
    0.27
    а
    0.26
    the
    0.26
    Act Density 0.024%

    No Known Activations