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

    Neuronpedia

    INDEX
    Explanations

    sweet flavors and treats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ल
    1.20
     begle
    1.11
    та
    1.09
    ش
    1.09
    ли
    1.05
    ет
    1.01
     texte
    0.97
     übertragen
    0.96
    ле
    0.95
     werden
    0.95
    POSITIVE LOGITS
     sweet
    1.37
    in
    1.25
     Sweet
    1.21
    at
    1.14
     sweeter
    1.13
    ：
    1.09
     sweetness
    1.07
    as
    1.02
    sweet
    1.02
     ۳
    1.01
    Act Density 0.015%

    No Known Activations