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

    Neuronpedia

    INDEX
    Explanations

    wrong answers or options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    are
    0.60
    in
    0.56
    or
    0.55
     in
    0.53
    be
    0.53
    is
    0.51
    al
    0.50
     are
    0.50
    schlag
    0.50
     and
    0.50
    POSITIVE LOGITS
    趄
    0.54
    瑆
    0.54
     batalha
    0.52
    तक
    0.52
    蒉
    0.51
    UG
    0.50
    ünkü
    0.50
    າມາດ
    0.50
    अप्र
    0.50
    εται
    0.50
    Act Density 0.000%

    No Known Activations