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

    Neuronpedia

    INDEX
    Explanations

    unprocessed data, lists, or removing malicious

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    払い
    1.17
    гаа
    1.14
    훙
    1.10
    囩
    1.04
    Σ
    1.02
    팅
    1.01
     primitives
    1.01
     biosynthesis
    1.00
     permissive
    1.00
    ﺮ
    0.99
    POSITIVE LOGITS
    на
    1.71
    م
    1.48
    at
    1.47
    ığı
    1.30
    ف
    1.27
    й
    1.27
    นด์
    1.25
    ртом
    1.23
    m
    1.23
    FET
    1.19
    Act Density 0.000%

    No Known Activations