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

    Neuronpedia

    INDEX
    Explanations

    stay

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    徙
    -0.07
    肉类
    -0.07
     allo
    -0.07
     of
    -0.07
    布朗
    -0.07
     mưa
    -0.07
    ()%
    -0.06
     ör
    -0.06
    贝尔
    -0.06
    锎
    -0.06
    POSITIVE LOGITS
    自负
    0.08
     GRAT
    0.07
    大家
    0.07
     castle
    0.07
     staple
    0.07
     stay
    0.06
    …I
    0.06
     devote
    0.06
    layers
    0.06
     reservations
    0.06
    Act Density 0.022%

    No Known Activations