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

    Neuronpedia

    INDEX
    Explanations

    fake passes and runs

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     punishments
    0.75
    പര
    0.74
     concerne
    0.73
     verteilt
    0.72
     constitué
    0.72
    戦略
    0.71
    herence
    0.70
     ಅನುಪ
    0.68
     strategic
    0.66
     constante
    0.66
    POSITIVE LOGITS
     underneath
    0.77
     low
    0.76
    fake
    0.75
     fake
    0.74
    Low
    0.73
    Fake
    0.72
    骗
    0.71
    Silk
    0.68
     Fake
    0.67
    low
    0.67
    Act Density 0.047%

    No Known Activations