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

    Neuronpedia

    INDEX
    Explanations

    composite

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    anes
    -0.07
     mange
    -0.06
     Uganda
    -0.06
     stones
    -0.06
    읍
    -0.06
    uxt
    -0.06
    locked
    -0.06
     Hern
    -0.06
    ुप
    -0.06
    	al
    -0.06
    POSITIVE LOGITS
     dumping
    0.09
    [\
    0.07
     grammar
    0.06
     comprising
    0.06
    dum
    0.06
    ynchronously
    0.06
    이나
    0.06
     homework
    0.06
     amino
    0.06
    нение
    0.06
    Act Density 0.004%

    No Known Activations