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

    Neuronpedia

    INDEX
    Explanations

    phrases related to protection or safety

    oai_token-act-pair · gpt-3.5-turbo

    keywords related to protection and safety

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     neoc
    -0.75
     nodd
    -0.74
    ickson
    -0.74
     throat
    -0.72
     cinem
    -0.71
    eyed
    -0.70
    geries
    -0.70
    ikawa
    -0.70
     carn
    -0.69
     hairst
    -0.68
    POSITIVE LOGITS
     Protect
    2.91
    Protect
    2.76
     Safe
    2.00
    Safe
    1.82
    protect
    1.72
     Help
    1.50
     Secure
    1.49
     Prevent
    1.47
    Free
    1.39
     Protector
    1.37
    Act Density 0.041%

    No Known Activations