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

    Neuronpedia

    INDEX
    Explanations

    references to hazards or risks related to various contexts

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

    hazard ratio

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    بالإنجليزية
    -0.54
     unwritten
    -0.47
     unspoken
    -0.44
     للاسماء
    -0.44
    Cartney
    -0.42
    OrCreate
    -0.41
    <h1>
    -0.41
     McCartney
    -0.41
    stech
    -0.41
    Биография
    -0.41
    POSITIVE LOGITS
     hazard
    2.22
     Hazard
    2.11
    hazard
    1.94
    Hazard
    1.94
     HAZARD
    1.87
     hazards
    1.74
     Hazards
    1.69
    Hazards
    1.38
     hazardous
    1.24
     Hazardous
    1.21
    Act Density 0.003%

    No Known Activations