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

    Neuronpedia

    INDEX
    Explanations

    even if

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ᒐ
    0.51
    いましたが
    0.47
    基本的に
    0.44
    きましたが
    0.41
    ioned
    0.41
    clature
    0.41
    ieuse
    0.41
     luckily
    0.41
     accessibility
    0.40
    pèce
    0.40
    POSITIVE LOGITS
    只是
    0.55
     imperfect
    0.54
     sadece
    0.53
    แค่
    0.50
    哪怕
    0.48
     simplemente
    0.48
     или
    0.47
     merely
    0.47
    Even
    0.47
     just
    0.46
    Act Density 0.097%

    No Known Activations