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

    Neuronpedia

    INDEX
    Explanations

    percent brightness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Nisan
    0.46
    questa
    0.45
    ريكي
    0.43
     darwin
    0.43
    ónde
    0.43
     این
    0.42
     Evan
    0.41
    狠狠
    0.41
     jego
    0.41
     đáp
    0.41
    POSITIVE LOGITS
    Re
    0.45
    Slash
    0.44
    Initialization
    0.44
    Slip
    0.43
    IR
    0.41
    {
    0.41
    AE
    0.40
    Orientation
    0.40
    Drag
    0.40
    Potential
    0.40
    Act Density 0.000%

    No Known Activations