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

    Neuronpedia

    INDEX
    Explanations

    announcements and award categories

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Panel
    -0.89
     Summit
    -0.85
    ://$
    -0.81
     every
    -0.81
     tangible
    -0.81
     three
    -0.80
     empfangen
    -0.77
     rudimentary
    -0.77
     any
    -0.75
    tipos
    -0.75
    POSITIVE LOGITS
     Dominus
    0.84
     ê
    0.82
     déterminer
    0.81
    reversible
    0.80
     transparente
    0.80
    GameManager
    0.79
     salva
    0.79
    bruck
    0.79
    帮你
    0.77
    ושים
    0.77
    Act Density 0.037%

    No Known Activations