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

    Neuronpedia

    INDEX
    Explanations

    prototype definitions and initializers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     passwd
    -0.90
     panna
    -0.85
    　↓
    -0.78
    carbon
    -0.77
    펀
    -0.76
    컸
    -0.75
     نهایت
    -0.74
     Eier
    -0.73
     naveg
    -0.72
     ROUTES
    -0.72
    POSITIVE LOGITS
     initialize
    1.63
    initialize
    1.60
     иници
    1.02
    初始化
    0.90
     initialization
    0.89
     Initialize
    0.88
     init
    0.87
     defaults
    0.87
     initializing
    0.87
     Initializes
    0.86
    Act Density 0.034%

    No Known Activations