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

    Neuronpedia

    INDEX
    Explanations

    instructions or how-to guides

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ensuring
    0.52
     ensures
    0.51
     confirms
    0.45
     ensure
    0.44
     necessary
    0.44
     confirmer
    0.44
     diminishes
    0.43
    ttä
    0.43
     memastikan
    0.42
     must
    0.42
    POSITIVE LOGITS
     можно
    0.68
     يمكنك
    0.64
     Canva
    0.63
     можна
    0.63
    你可以
    0.63
     puedes
    0.60
    Airbnb
    0.59
     możesz
    0.59
     алыңыз
    0.59
     онлайн
    0.58
    Act Density 0.544%

    No Known Activations