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

    Neuronpedia

    INDEX
    Explanations

    re-organized, simplifies, reveals

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    antly
    1.16
    m
    1.16
    ivé
    1.07
     tươi
    1.06
     riguardo
    1.03
    ell
    1.02
    یر
    1.01
     iria
    1.01
    that
    1.00
    пло
    1.00
    POSITIVE LOGITS
    ти
    1.41
    ร
    1.30
    र
    1.27
    те
    1.26
    ר
    1.25
    ка
    1.23
    де
    1.22
    ке
    1.22
    ر
    1.18
    ర్థిక
    1.16
    Act Density 0.181%

    No Known Activations