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

    Neuronpedia

    INDEX
    Explanations

    cause, relationship, sexuality, secretly

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ourselves
    0.52
     are
    0.49
     as
    0.49
     volatile
    0.45
     but
    0.44
     care
    0.44
     look
    0.44
     that
    0.43
     spoilt
    0.43
     Whey
    0.43
    POSITIVE LOGITS
    ᅪ
    0.53
     ਅਤੇ
    0.48
    →</
    0.47
     دوبارہ
    0.46
    !</
    0.46
     వల్ల
    0.46
    !");
    0.44
    !]
    0.44
     причинам
    0.44
    뽀
    0.44
    Act Density 0.021%

    No Known Activations