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

    Neuronpedia

    INDEX
    Explanations

    various specific domains

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     one
    0.57
     of
    0.54
     to
    0.52
     để
    0.46
     a
    0.45
     when
    0.45
     on
    0.45
    ة
    0.45
     be
    0.44
     برای
    0.44
    POSITIVE LOGITS
     pudi
    0.38
    <unused0>
    0.37
    ዀ
    0.35
    数据的
    0.35
    uchsia
    0.35
    σιμοποι
    0.34
    modium
    0.34
    </li>
    0.34
    を採用
    0.34
    可能有
    0.34
    Act Density 0.002%

    No Known Activations