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

    Neuronpedia

    INDEX
    Explanations

    references to safety and danger

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    omu
    -0.16
    HAM
    -0.15
    wheel
    -0.15
    ¶Į
    -0.15
    ETCH
    -0.15
    ham
    -0.15
    격
    -0.14
    Ä°M
    -0.14
    Rotor
    -0.14
    RATE
    -0.13
    POSITIVE LOGITS
     safety
    0.25
     unsafe
    0.24
     Safety
    0.23
    unsafe
    0.23
     risks
    0.22
    afety
    0.21
     safer
    0.21
     risky
    0.21
    Safety
    0.21
     dangerous
    0.21
    Act Density 0.083%

    No Known Activations