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

    Neuronpedia

    INDEX
    Explanations

    you can

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     képes
    0.62
     capable
    0.55
     قادر
    0.55
    capable
    0.54
     capaz
    0.52
     способны
    0.51
     mampu
    0.51
     capability
    0.50
     incapable
    0.49
    Capable
    0.49
    POSITIVE LOGITS
    也可以
    0.69
     можно
    0.68
     bisa
    0.67
     could
    0.66
     můžete
    0.61
     podría
    0.61
     можна
    0.61
     pourrait
    0.60
     könnte
    0.60
     можете
    0.59
    Act Density 0.007%

    No Known Activations