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

    Neuronpedia

    INDEX
    Explanations

    efficiency, concurrency, simplicity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    너무
    0.45
    可以
    0.44
    Не
    0.43
    이나
    0.43
    敕
    0.42
    出来的
    0.42
     rosem
    0.42
    the
    0.41
    문제
    0.41
    ﺹ
    0.41
    POSITIVE LOGITS
    ik
    0.54
    ية
    0.52
    ist
    0.49
    ES
    0.49
    uq
    0.45
    ic
    0.45
    лығы
    0.44
    uja
    0.43
    ق
    0.43
    ulter
    0.43
    Act Density 1.181%

    No Known Activations