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

    Neuronpedia

    INDEX
    Explanations

    rep code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ed
    1.73
    el
    1.48
    a
    1.42
    es
    1.36
    2
    1.34
    ent
    1.30
    os
    1.26
    o
    1.21
    en
    1.20
    an
    1.17
    POSITIVE LOGITS
    ب
    1.24
    т
    1.23
    。
    1.23
    он
    1.20
    е
    1.20
    ۔
    1.17
    ।
    1.16
    1.14
    ку
    1.12
    ।'
    1.12
    Act Density 0.000%

    No Known Activations