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

    Neuronpedia

    INDEX
    Explanations

    like comparing anomalies

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ཎ
    0.50
     όλα
    0.43
     rozwiąz
    0.42
    và
    0.42
    Licenses
    0.41
    0.41
     επίσης
    0.41
     alebo
    0.40
     iaitu
    0.40
     দেওয়া
    0.38
    POSITIVE LOGITS
    و
    0.91
    ing
    0.78
    ة
    0.77
    ו
    0.66
     can
    0.63
    us
    0.61
    ang
    0.60
    ed
    0.59
    ore
    0.59
    о
    0.55
    Act Density 0.000%

    No Known Activations