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

    Neuronpedia

    INDEX
    Explanations

    the concept of danger and threats to safety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     toString
    -0.54
     CommonModule
    -0.50
    فية
    -0.47
    ملة
    -0.46
    出版年
    -0.46
    雀
    -0.45
     Kälte
    -0.45
    cheon
    -0.44
    riedenheit
    -0.43
     tears
    -0.43
    POSITIVE LOGITS
     dangerous
    1.52
     unsafe
    1.52
     safer
    1.46
     danger
    1.42
    safety
    1.39
     safety
    1.39
    dangerous
    1.38
     peligroso
    1.37
     safest
    1.34
     pelig
    1.32
    Act Density 0.283%

    No Known Activations