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

    Neuronpedia

    INDEX
    Explanations

    cling to surfaces

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ти
    0.89
    ем
    0.85
    های
    0.83
     ευ
    0.82
    ри
    0.82
    alne
    0.79
     ஆண்டுகள்
    0.75
    cientes
    0.74
    ра
    0.73
    galkan
    0.73
    POSITIVE LOGITS
    </h3>
    1.06
    </b>
    1.05
    1.05
    ↵↵
    1.00
     is
    0.96
     a
    0.94
    </h2>
    0.93
    </i>
    0.93
    </a>
    0.91
     l
    0.90
    Act Density 0.003%

    No Known Activations