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

    Neuronpedia

    INDEX
    Explanations

    function purpose, arguments, return value

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     purposes
    0.55
     Purposes
    0.50
    Themes
    0.41
    perluan
    0.41
     razones
    0.40
    urpose
    0.39
    ilir
    0.38
    contentText
    0.38
    content
    0.38
     reasons
    0.38
    POSITIVE LOGITS
     Args
    0.68
     accepts
    0.65
     Inputs
    0.65
     принимает
    0.64
     inputs
    0.61
    返回值
    0.61
     accept
    0.57
     прийма
    0.57
     返回
    0.54
     RETURNS
    0.53
    Act Density 0.008%

    No Known Activations