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

    Neuronpedia

    INDEX
    Explanations

    if that

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     不过
    0.44
     ذریع
    0.43
    িনবার্গ
    0.42
     pięk
    0.42
     คง
    0.41
     cómod
    0.40
     conç
    0.40
     puisqu
    0.40
    ђено
    0.39
     conçu
    0.38
    POSITIVE LOGITS
     there
    1.11
    there
    0.94
     someone
    0.90
     it
    0.89
    該
    0.79
     해당
    0.78
     somebody
    0.77
     the
    0.75
    someone
    0.73
     هناك
    0.72
    Act Density 0.010%

    No Known Activations