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

    Neuronpedia

    INDEX
    Explanations

    wit

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ”
    0.48
    "
    0.45
    ad
    0.42
    ,
    0.41
     -
    0.40
    %
    0.40
     days
    0.39
     and
    0.38
    ;
    0.38
    words
    0.38
    POSITIVE LOGITS
     фильм
    0.59
     ою
    0.58
    ıp
    0.55
    眇
    0.54
    耔
    0.53
     hendak
    0.52
    蟳
    0.52
     croche
    0.51
     ভারতীয়
    0.50
     capuche
    0.50
    Act Density 0.001%

    No Known Activations