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

    Neuronpedia

    INDEX
    Explanations

    sometimes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ار
    1.89
    q
    1.63
    é
    1.52
    ol
    1.36
    ć
    1.36
    encoder
    1.28
    id
    1.27
    ut
    1.24
    ay
    1.23
    ؤں
    1.23
    POSITIVE LOGITS
     اوقات
    1.94
    கூ
    1.73
     ומ
    1.66
    lers
    1.60
    கோ
    1.52
    ني
    1.44
    нде
    1.43
    ました
    1.42
    לט
    1.41
     נ
    1.38
    Act Density 0.086%

    No Known Activations