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

    Neuronpedia

    INDEX
    Explanations

    Dear followed by recipient

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    出現
    0.45
     graf
    0.43
     wares
    0.41
    >)`](
    0.40
     اسئله
    0.39
     absur
    0.38
    수의
    0.38
     syll
    0.38
     rampant
    0.38
    ပြု
    0.38
    POSITIVE LOGITS
    Dear
    0.71
     Dear
    0.62
    dear
    0.55
     dear
    0.53
     esteemed
    0.48
     whoever
    0.47
     Family
    0.46
     merhaba
    0.46
    안녕하세요
    0.44
     President
    0.44
    Act Density 0.093%

    No Known Activations