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

    Neuronpedia

    INDEX
    Explanations

    likely teasing language

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     soutien
    0.52
    ząc
    0.50
    峣
    0.50
     быстрее
    0.49
    вающий
    0.47
    为
    0.47
    创建一个
    0.45
    馏
    0.45
     እና
    0.44
    轅
    0.44
    POSITIVE LOGITS
     phenomena
    0.45
     aforementioned
    0.45
     ubiquitous
    0.45
     acoustics
    0.45
     paraphernalia
    0.44
    <unused60>
    0.44
     technologies
    0.43
     wares
    0.42
     idiosyncratic
    0.41
     addicted
    0.40
    Act Density 0.008%

    No Known Activations