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

    Neuronpedia

    INDEX
    Explanations

    danger and threats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Cav
    0.44
     ova
    0.41
    uci
    0.41
     주로
    0.41
    Label
    0.41
     대부분
    0.41
    Select
    0.40
    Drive
    0.40
     تقريبا
    0.40
    HTMLElement
    0.40
    POSITIVE LOGITS
     worrisome
    1.34
     ominous
    1.18
     worrying
    1.11
     alarming
    1.10
     dangerous
    1.02
     опасность
    1.02
     опас
    1.01
     threat
    0.98
     خطر
    0.98
     zagro
    0.98
    Act Density 0.134%

    No Known Activations