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

    Neuronpedia

    INDEX
    Explanations

    decoupling and predictability

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     amigas
    0.41
    ling
    0.40
    kräfte
    0.39
     आप
    0.39
    nuevo
    0.38
    eman
    0.38
    ocy
    0.38
    اقات
    0.38
     нормы
    0.38
    वानी
    0.37
    POSITIVE LOGITS
     ترى
    0.43
     about
    0.42
     tentang
    0.41
     aporta
    0.41
    يته
    0.41
     براي
    0.41
     về
    0.40
     benöt
    0.40
     chave
    0.40
     acerca
    0.40
    Act Density 0.006%

    No Known Activations