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

    Neuronpedia

    INDEX
    Explanations

    flawless, pristine, spotless

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     coarser
    0.96
     inadequ
    0.89
    唆
    0.83
     louder
    0.82
     mauvais
    0.80
     сексуа
    0.80
     divisive
    0.80
     διαφορε
    0.79
     weaker
    0.77
    纣
    0.76
    POSITIVE LOGITS
     flawless
    1.80
     pristine
    1.76
     spotless
    1.73
     perfection
    1.56
     Perfection
    1.56
     immaculate
    1.56
     perfect
    1.48
     Perfect
    1.44
     perfek
    1.38
     perfetta
    1.37
    Act Density 0.236%

    No Known Activations