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

    Neuronpedia

    INDEX
    Explanations

    code punctuation and string parts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     or
    -2.52
     those
    -2.50
     on
    -2.39
     an
    -2.34
     When
    -2.33
    s
    -2.33
     to
    -2.31
     What
    -2.31
     if
    -2.30
     even
    -2.30
    POSITIVE LOGITS
    沵
    3.17
     boister
    2.55
    摀
    2.42
    2.33
    ⑊
    2.33
    2.33
     délais
    2.28
    2.27
     WHICH
    2.25
     appeler
    2.25
    Act Density 0.003%

    No Known Activations