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

    Neuronpedia

    INDEX
    Explanations

    compound phrases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused2197>
    1.15
    <unused2221>
    1.14
     oln
    1.08
    1.07
    <unused1258>
    1.04
     AppBsky
    1.01
    .??"]
    0.99
    <unused1468>
    0.98
    𒈪
    0.98
    𐰞
    0.97
    POSITIVE LOGITS
     (
    0.85
     "
    0.81
     '
    0.77
     of
    0.77
     “
    0.76
     provided
    0.75
     for
    0.75
     in
    0.74
     with
    0.73
     वो
    0.73
    Act Density 2.093%

    No Known Activations