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

    Neuronpedia

    INDEX
    Explanations

    phrases related to potential risks and dangers

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    danger, risk, compromise, prejudice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    AndEndTag
    -0.65
     Normdatei
    -0.52
    CodeAttribute
    -0.48
     Вікі
    -0.48
    UrlResolution
    -0.47
     mergeFrom
    -0.47
    weil
    -0.46
    isContained
    -0.46
    อิง
    -0.45
     yym
    -0.44
    POSITIVE LOGITS
     jeopardize
    1.07
     jeopardi
    1.06
     endanger
    0.98
     jeopardy
    0.88
     endangering
    0.85
     gefähr
    0.67
     endangered
    0.66
     danger
    0.66
    angering
    0.62
     risk
    0.62
    Act Density 0.080%

    No Known Activations