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

    Neuronpedia

    INDEX
    Explanations

    cause confusion

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Mileage
    0.94
     croche
    0.90
     skyrocketed
    0.88
     땅
    0.86
     Millions
    0.85
     gasoline
    0.84
     메모
    0.84
     Dancer
    0.84
     Sticker
    0.83
     slang
    0.83
    POSITIVE LOGITS
     reasonably
    1.27
    reasonably
    1.22
     mencegah
    1.21
    reasonable
    1.21
     reasonable
    1.12
    harm
    1.03
     harm
    0.99
     निवार
    0.95
     جلوگیری
    0.93
     dissu
    0.91
    Act Density 0.623%

    No Known Activations