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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating protection from various harmful influences or threats

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acific
    -0.17
    phem
    -0.16
    odom
    -0.15
    bane
    -0.15
    èĥ¶
    -0.15
    lÃŃÄį
    -0.15
    atoria
    -0.14
    /fwlink
    -0.14
    apas
    -0.14
    ideo
    -0.14
    POSITIVE LOGITS
     harm
    0.20
     harms
    0.19
     dangers
    0.18
     scrutiny
    0.18
     attack
    0.17
     further
    0.16
     becoming
    0.16
    æĿ¥èĩª
    0.16
     cov
    0.16
     danger
    0.15
    Act Density 0.086%

    No Known Activations