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

    Neuronpedia

    INDEX
    Explanations

    redefine future, complements intelligence, interpreting complex

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused413>
    0.79
    <unused1813>
    0.76
    <unused364>
    0.76
    ciąż
    0.75
     considérer
    0.75
    Flora
    0.74
    <unused391>
    0.74
    <unused713>
    0.73
    <unused1771>
    0.72
    <unused392>
    0.72
    POSITIVE LOGITS
     and
    1.19
     aspects
    1.04
     or
    1.01
    /
    1.01
    ively
    0.98
     situations
    0.92
     how
    0.92
     any
    0.91
     not
    0.89
    ably
    0.88
    Act Density 0.991%

    No Known Activations