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

    Neuronpedia

    INDEX
    Explanations

    drastically change

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     diferencia
    0.46
     mancanza
    0.43
     achievements
    0.41
     full
    0.39
     Difference
    0.39
     poseb
    0.38
     recuperación
    0.38
    خذ
    0.38
     reducción
    0.38
     diferença
    0.38
    POSITIVE LOGITS
     drastically
    0.77
     radically
    0.71
     parameters
    0.70
     fortunes
    0.66
     dramatically
    0.63
     perceptions
    0.63
     priorities
    0.62
     perception
    0.61
     fundamentally
    0.61
     기존
    0.61
    Act Density 0.142%

    No Known Activations