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

    Neuronpedia

    INDEX
    Explanations

    foreign language words

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     considere
    0.58
     survive
    0.58
     repair
    0.56
     childcare
    0.56
     wrongdoing
    0.54
     surviv
    0.54
     commissioning
    0.53
     sufferings
    0.53
    riamo
    0.53
    zäh
    0.53
    POSITIVE LOGITS
     Bet
    0.56
     Kas
    0.56
     Ancak
    0.56
     Ди
    0.56
     Под
    0.56
     Ри
    0.56
     Э
    0.55
     Ши
    0.55
     Ку
    0.55
     Мо
    0.55
    Act Density 0.000%

    No Known Activations