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

    Neuronpedia

    INDEX
    Explanations

    desired specific things

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    appré
    0.56
    疯狂
    0.56
    {
    0.55
    =
    0.54
    沤
    0.54
    </h3>
    0.52
    극장
    0.52
    frac
    0.51
     შესახებ
    0.51
    acquired
    0.50
    POSITIVE LOGITS
     desired
    0.63
    at
    0.60
    on
    0.59
    in
    0.56
    os
    0.56
     desider
    0.55
     ils
    0.54
     odbior
    0.54
    з
    0.54
    Desired
    0.54
    Act Density 0.451%

    No Known Activations