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

    Neuronpedia

    INDEX
    Explanations

    describing technical or factual details

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     tutte
    0.43
    aviti
    0.43
    iniz
    0.40
    iber
    0.40
     tante
    0.39
     вина
    0.39
    ьогодні
    0.39
    етесь
    0.39
    るので
    0.39
    pegno
    0.38
    POSITIVE LOGITS
    ことも
    0.46
     живот
    0.45
    動物
    0.44
    _{\|
    0.43
    0.43
    ∗</
    0.43
     ကု
    0.42
     ANIM
    0.42
     Иванов
    0.42
    ˨
    0.42
    Act Density 0.002%

    No Known Activations