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

    Neuronpedia

    INDEX
    Explanations

    percentages

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ironically
    0.65
    anjut
    0.64
    Christopher
    0.64
    นาน
    0.63
     Christopher
    0.62
    rivial
    0.61
    imir
    0.60
     Honors
    0.59
    诫
    0.59
    inii
    0.59
    POSITIVE LOGITS
    %
    4.28
     percentage
    4.19
     percent
    3.99
     %
    3.95
     percentages
    3.74
    %,
    3.74
     Percentage
    3.73
    Percentage
    3.71
    percentage
    3.61
     Percent
    3.54
    Act Density 0.602%

    No Known Activations