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

    Neuronpedia

    INDEX
    Explanations

    adorned, Cafe, er/**/*

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    "
    1.38
     is
    1.32
    ко
    1.30
    <0x0D>
    1.29
    م
    1.26
    ig
    1.25
    ে
    1.23
    de
    1.16
    ли
    1.14
    ’
    1.12
    POSITIVE LOGITS
    ล
    1.37
    ל
    1.20
     in
    1.11
    ाना
    1.09
    тни
    1.07
    る
    1.07
    𝒈
    1.03
     در
    0.99
    ार्ड
    0.96
    u
    0.96
    Act Density 0.000%

    No Known Activations