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

    Neuronpedia

    INDEX
    Explanations

    identifying a winner

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     亻
    0.40
     estabilidad
    0.36
    itatis
    0.35
    డ్డు
    0.34
    ገድ
    0.33
    🅰
    0.33
     respeto
    0.33
    ನದ
    0.32
     rejest
    0.32
     흔
    0.32
    POSITIVE LOGITS
     winner
    3.08
     winners
    2.97
    winner
    2.63
     Winner
    2.56
    Winner
    2.55
    winners
    2.53
     Winners
    2.39
    Winners
    2.33
     победи
    2.22
     venced
    2.20
    Act Density 0.057%

    No Known Activations