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

    Neuronpedia

    INDEX
    Explanations

    phrases related to bypassing security measures

    oai_token-act-pair · gpt-3.5-turbo

    terms related to evading or bypassing rules and regulations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Sad
    -0.67
    uesday
    -0.67
    killer
    -0.66
    soon
    -0.66
    mad
    -0.64
    iaries
    -0.64
     breaths
    -0.64
    oom
    -0.63
    asp
    -0.62
    aeus
    -0.62
    POSITIVE LOGITS
     detection
    1.20
     prohibitions
    0.85
     FOIA
    0.83
     restrictions
    0.81
     regulations
    0.81
     censorship
    0.81
     bounds
    0.79
     Regulation
    0.78
     pesky
    0.78
     Detection
    0.77
    Act Density 0.126%

    No Known Activations