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

    Neuronpedia

    INDEX
    Explanations

    components as building blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    HandleTypeDef
    0.40
     prisoners
    0.39
     Prisoners
    0.38
     JACKSON
    0.38
     Geraldine
    0.38
     Cour
    0.36
     prisoner
    0.36
     fratern
    0.35
    ারে
    0.35
    ৃতির
    0.35
    POSITIVE LOGITS
     components
    4.28
     component
    4.03
     componentes
    3.98
     Components
    3.92
     komponen
    3.88
     компонен
    3.84
     komponent
    3.84
    components
    3.81
     componente
    3.73
    Components
    3.70
    Act Density 0.210%

    No Known Activations