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

    Neuronpedia

    INDEX
    Explanations

    say be/been

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ندار
    0.79
     anschauen
    0.77
     получать
    0.75
     मिलेगी
    0.73
     başlay
    0.71
     mengakses
    0.71
    менять
    0.71
     говорить
    0.71
     irão
    0.70
    会出现
    0.69
    POSITIVE LOGITS
     Being
    1.61
     being
    1.57
    Being
    1.46
    being
    1.37
     bee
    1.24
     BEING
    1.20
    被
    1.16
     Be
    1.14
     bein
    1.12
    être
    1.12
    Act Density 0.268%

    No Known Activations