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

    Neuronpedia

    INDEX
    Explanations

    programming code separators

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ه
    1.79
    יות
    1.72
    }";
    1.70
    ्य
    1.69
    ا
    1.69
    }");
    1.65
    ב
    1.61
    ب
    1.55
    ز
    1.52
    ة
    1.51
    POSITIVE LOGITS
    ée
    1.61
    ur
    1.47
    ik
    1.36
    il
    1.34
    お待ち
    1.32
    ır
    1.30
    𝕟
    1.29
    此外
    1.27
    有一个
    1.22
    ā
    1.22
    Act Density 0.058%

    No Known Activations