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

    Neuronpedia

    INDEX
    Explanations

    standing up for oneself or others

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     aument
    0.67
     en
    0.64
    Unter
    0.64
    Reduce
    0.61
    Land
    0.60
     Caribe
    0.59
    u
    0.59
     lleg
    0.59
     gris
    0.59
     enviar
    0.59
    POSITIVE LOGITS
     défendre
    0.63
    мол
    0.62
    ukone
    0.62
    نگ
    0.56
     intérêts
    0.56
    ці
    0.55
    ρα
    0.55
    چھ
    0.55
    фта
    0.55
    avaient
    0.55
    Act Density 0.016%

    No Known Activations