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

    Neuronpedia

    INDEX
    Explanations

    book titles and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     (
    1.72
    ,
    1.69
    .
    1.68
    -
    1.66
    :
    1.62
    1.56
     and
    1.46
    /
    1.45
    '
    1.44
     in
    1.44
    POSITIVE LOGITS
    EnglishMarks
    1.81
    <unused595>
    1.80
    <unused1115>
    1.78
    <unused1897>
    1.77
    mataspid
    1.76
    MathMarks
    1.76
    ManagerPortal
    1.76
    goài
    1.75
    <unused1025>
    1.74
    <unused227>
    1.74
    Act Density 0.009%

    No Known Activations