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

    Neuronpedia

    INDEX
    Explanations

    a través de

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     de
    0.72
    情况
    0.72
     delle
    0.72
    ança
    0.70
     Abbiamo
    0.70
     sendo
    0.69
     със
    0.69
    问题
    0.69
    ईल
    0.69
     چون
    0.67
    POSITIVE LOGITS
     través
    1.17
     menudo
    1.03
     medida
    1.00
     partir
    0.98
     propósito
    0.98
     voluntad
    0.97
     fondo
    0.96
     modo
    0.95
     manera
    0.95
     largo
    0.94
    Act Density 0.067%

    No Known Activations