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

    Neuronpedia

    INDEX
    Explanations

    introducing contrast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     CLASS
    0.31
    "
    0.29
     derivados
    0.28
     Raising
    0.28
     جوړونکي
    0.28
    OO
    0.27
    途径
    0.27
     MAY
    0.26
     NOTE
    0.26
    ريب
    0.26
    POSITIVE LOGITS
     despite
    0.66
    despite
    0.66
     trotz
    0.64
     Despite
    0.61
    Despite
    0.61
     несмотря
    0.59
    pesar
    0.55
     being
    0.55
     nonostante
    0.53
     apesar
    0.53
    Act Density 0.002%

    No Known Activations