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

    Neuronpedia

    INDEX
    Explanations

    code

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     управления
    -0.06
    恨
    -0.06
     sentencing
    -0.06
    mA
    -0.06
     кры
    -0.06
     marginal
    -0.06
    suffix
    -0.06
     웨
    -0.06
     kop
    -0.06
    ové
    -0.06
    POSITIVE LOGITS
    habit
    0.07
     выс
    0.07
    NF
    0.06
     sque
    0.06
    ession
    0.06
    -red
    0.06
    θα
    0.06
     PROT
    0.06
    racat
    0.06
    ."'
    0.06
    Act Density 0.071%

    No Known Activations