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

    Neuronpedia

    INDEX
    Explanations

    batch normalization

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ೋನ್
    1.04
     corriente
    1.00
    蛊
    0.99
    凵
    0.97
     menawarkan
    0.96
    Allemagne
    0.95
     uomo
    0.94
    别的
    0.94
    тік
    0.94
    IERC
    0.93
    POSITIVE LOGITS
    s
    1.30
    ter
    1.29
    y
    1.28
    st
    1.27
    x
    1.25
    il
    1.21
    й
    1.19
    я
    1.18
    re
    1.13
    j
    1.11
    Act Density 0.000%

    No Known Activations