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

    Neuronpedia

    INDEX
    Explanations

    direction

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ada
    0.75
    indahkan
    0.70
    ،
    0.69
     disimpan
    0.68
    harap
    0.67
     kekuatan
    0.66
     penampilan
    0.66
    anes
    0.65
    kadang
    0.65
    ation
    0.64
    POSITIVE LOGITS
     direction
    1.38
    direction
    1.29
    方向
    1.20
     Direction
    1.19
     DIRECTION
    1.13
    Direction
    1.10
     direzione
    1.05
     direção
    1.04
     방향
    1.03
     directions
    0.98
    Act Density 0.031%

    No Known Activations