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

    Neuronpedia

    INDEX
    Explanations

    LaTeX author commands

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     azo
    0.47
    queta
    0.45
    acia
    0.42
    YPE
    0.40
    )-$
    0.40
    stvo
    0.40
     tu
    0.40
    untur
    0.40
     gauze
    0.39
    }'.
    0.39
    POSITIVE LOGITS
    }\\
    0.47
     \\
    0.45
    大学
    0.44
     "\\
    0.43
    腾讯
    0.43
    同一个
    0.42
     \\\
    0.42
    课程
    0.41
    \\
    0.41
    {}
    0.41
    Act Density 0.001%

    No Known Activations