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

    Neuronpedia

    INDEX
    Explanations

    http links

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     smuggling
    1.21
     lleve
    1.20
     enfoque
    1.18
     destroy
    1.17
     shades
    1.10
    𝑵
    1.10
    旛
    1.09
     regimens
    1.08
     culmin
    1.08
    国防
    1.07
    POSITIVE LOGITS
    д
    1.27
    νά
    1.21
    ์
    1.14
    ح
    1.12
    <0x80>
    1.09
    ons
    0.98
    et
    0.95
    рби
    0.95
    t
    0.95
    р
    0.93
    Act Density 0.124%

    No Known Activations