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

    Neuronpedia

    INDEX
    Explanations

    book titles and descriptions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    媑
    0.61
    <unused1057>
    0.60
    ”।
    0.55
    。</
    0.55
    睉
    0.55
    <unused1021>
    0.55
    <unused1858>
    0.55
    呥
    0.54
    🗾
    0.54
    <unused2020>
    0.54
    POSITIVE LOGITS
    |
    0.69
     (
    0.59
     non
    0.59
    :
    0.59
     sentiment
    0.56
     handle
    0.55
     L
    0.54
     calc
    0.54
     pre
    0.54
     user
    0.53
    Act Density 0.000%

    No Known Activations