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

    Neuronpedia

    INDEX
    Explanations

    defeating or overcoming

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    搶
    0.85
    抢
    0.82
    ിലേ
    0.78
     stealing
    0.74
     steal
    0.74
     수상
    0.71
     চুরি
    0.71
     theft
    0.69
    ጵ
    0.67
     चोरी
    0.67
    POSITIVE LOGITS
     defeat
    2.29
     defeated
    2.22
     defeats
    2.19
     defeating
    2.03
    defeated
    2.00
    defeat
    1.97
     beat
    1.77
     derrot
    1.76
     vanqu
    1.74
     derrota
    1.69
    Act Density 0.169%

    No Known Activations