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

    Neuronpedia

    INDEX
    Explanations

    excess or exaggeration

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     przeprowad
    0.92
     entspricht
    0.84
     tronco
    0.79
     wafers
    0.78
     unmistakable
    0.76
     jakieś
    0.76
     godiš
    0.75
     combatir
    0.75
    ziała
    0.75
     hauptsächlich
    0.75
    POSITIVE LOGITS
    whelming
    1.41
    whel
    1.34
    lapping
    1.21
    emphas
    1.20
    reliance
    1.20
    abundance
    1.19
    used
    1.18
    blown
    1.15
    ripe
    1.14
    tones
    1.14
    Act Density 0.133%

    No Known Activations