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

    Neuronpedia

    INDEX
    Explanations

    phrases related to potential threats or looming situations

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to impending threats or challenges

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ive
    -0.80
    ilic
    -0.79
    owder
    -0.77
    hibition
    -0.75
    ves
    -0.74
    hib
    -0.73
    artments
    -0.72
    %]
    -0.71
    hibited
    -0.69
    leeve
    -0.69
    POSITIVE LOGITS
     looms
    1.10
     looming
    0.96
    abouts
    0.91
     omin
    0.87
    enance
    0.75
     owl
    0.73
     Archdemon
    0.68
     Pose
    0.68
    stakes
    0.67
     Izan
    0.66
    Act Density 0.014%

    No Known Activations