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

    Neuronpedia

    INDEX
    Explanations

    list items or summaries

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <eos>
    0.73
    '।
    0.70
    。</
    0.69
    such
    0.68
     or
    0.64
    ende
    0.61
    of
    0.60
    presumably
    0.60
    sof
    0.60
    。",
    0.60
    POSITIVE LOGITS
     Plus
    1.23
     Bonus
    1.12
     Spoiler
    1.09
     Quote
    1.06
    ):
    1.05
     Rated
    1.02
     Ingredient
    1.01
     Ingredients
    1.01
     👇
    1.01
     Average
    1.00
    Act Density 0.396%

    No Known Activations