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

    Neuronpedia

    INDEX
    Explanations

    wuv, shakin', wags, wanes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     and
    0.73
     or
    0.69
    на
    0.57
    га
    0.50
    :
    0.50
     for
    0.49
    5
    0.49
    ة
    0.48
    ية
    0.48
     are
    0.47
    POSITIVE LOGITS
    0.53
    t
    0.50
    dalam
    0.44
    。
    0.41
    d
    0.41
    た
    0.41
    grupo
    0.40
     थांब
    0.40
    ।
    0.40
    pisah
    0.39
    Act Density 0.062%

    No Known Activations