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

    Neuronpedia

    INDEX
    Explanations

    drama and magic

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     crushed
    0.88
    re
    0.88
    for
    0.76
    留下
    0.74
     בק
    0.74
     for
    0.73
     Cly
    0.72
    fresh
    0.71
     better
    0.70
    where
    0.70
    POSITIVE LOGITS
     állapot
    1.27
    asText
    1.23
    𒈹
    1.21
     ہُما
    1.20
    <unused182>
    1.19
    𒈪
    1.19
    mataspid
    1.18
     nirvachan
    1.18
     erbjuder
    1.18
    𐰺
    1.18
    Act Density 0.208%

    No Known Activations