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

    Neuronpedia

    INDEX
    Explanations

    chicken and meat quantities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tea
    1.02
     Tea
    1.01
     candies
    0.96
    咖啡
    0.96
     caffeine
    0.95
     coffee
    0.95
     Kaffee
    0.95
     Cups
    0.94
     sugary
    0.94
     cocoa
    0.94
    POSITIVE LOGITS
     meat
    1.55
     Meat
    1.47
     meats
    1.46
     chicken
    1.33
     carne
    1.30
    肉
    1.27
     चिकन
    1.26
    Meat
    1.26
     beef
    1.26
    Beef
    1.25
    Act Density 0.605%

    No Known Activations