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

    Neuronpedia

    INDEX
    Explanations

    know

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Right
    1.19
    Right
    1.12
     right
    1.03
    right
    0.95
    右
    0.94
     RIGHT
    0.84
    RIGHT
    0.79
     मजब
    0.70
    arnas
    0.67
     direita
    0.66
    POSITIVE LOGITS
     how
    0.80
    ？
    0.78
     How
    0.77
     ¿
    0.77
     ?";
    0.74
    ?
    0.74
     cómo
    0.74
    !?
    0.73
     troupe
    0.72
    ?!"
    0.71
    Act Density 0.006%

    No Known Activations