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

    Neuronpedia

    INDEX
    Explanations

    specific numbers and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     but
    -0.93
     it
    -0.91
    ͎
    -0.90
     maart
    -0.90
     ngayon
    -0.88
     bawat
    -0.88
    腴
    -0.86
     why
    -0.85
     SQLITE
    -0.85
     iyong
    -0.84
    POSITIVE LOGITS
     první
    1.03
     aseguran
    0.98
     consideramos
    0.97
     necesitan
    0.96
     deaktivi
    0.94
    <th>
    0.93
     preferências
    0.88
     tenues
    0.88
     primeiras
    0.87
     admiten
    0.87
    Act Density 0.010%

    No Known Activations