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

    Neuronpedia

    INDEX
    Explanations

    iPhone Pro Max, Pixel series

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    0.48
     Win
    0.45
    Deputy
    0.45
    У
    0.45
     MATH
    0.44
    间
    0.44
     lung
    0.44
     passive
    0.43
    Passive
    0.43
    BERG
    0.43
    POSITIVE LOGITS
     kraja
    0.54
    stava
    0.53
    createSprite
    0.51
    ја
    0.50
    िलेश
    0.48
     руководителя
    0.48
    കേന്ദ്രം
    0.47
    시켜
    0.47
    літы
    0.46
    شرين
    0.46
    Act Density 0.008%

    No Known Activations