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

    Neuronpedia

    INDEX
    Explanations

    purpose is

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     کی۔
    0.82
     Containing
    0.76
     کیا۔
    0.76
    만의
    0.72
    .);
    0.69
    стный
    0.68
    '};
    0.67
    Represent
    0.66
    ";}
    0.66
    .).
    0.66
    POSITIVE LOGITS
     lies
    1.46
     заключается
    1.40
     revolves
    1.36
     වන්නේ
    1.21
     here
    1.21
     заключа
    1.21
     lie
    1.13
    คือ
    1.11
     aquí
    1.10
     revolve
    1.09
    Act Density 0.313%

    No Known Activations