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

    Neuronpedia

    INDEX
    Explanations

    expect errors or silos

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    逶
    0.48
    Technische
    0.48
     graus
    0.46
     выбран
    0.44
     wundersch
    0.44
     разнообраз
    0.43
     найбіль
    0.43
    绽
    0.43
     நடைபெ
    0.42
     കൊണ്ട
    0.42
    POSITIVE LOGITS
     billing
    0.46
     aftermarket
    0.45
     grocery
    0.44
     automaker
    0.43
    _
    0.41
     says
    0.40
     diathermy
    0.40
     postpaid
    0.40
    ↵
    0.40
     consumer
    0.39
    Act Density 0.009%

    No Known Activations