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

    Neuronpedia

    INDEX
    Explanations

    text after punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Ი
    1.46
    Ნ
    1.42
    Ყ
    1.37
    <unused2167>
    1.35
    <unused280>
    1.28
    <unused219>
    1.27
    <unused1919>
    1.27
    Ბ
    1.25
    <unused1903>
    1.25
    <unused1654>
    1.24
    POSITIVE LOGITS
    ,
    1.12
    :
    0.89
    .
    0.89
    ́
    0.86
    -
    0.83
     e
    0.81
    on
    0.79
     (
    0.77
    е
    0.77
     е
    0.75
    Act Density 0.376%

    No Known Activations