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

    Neuronpedia

    INDEX
    Explanations

    phrases related to taking advantage of situations or other individuals

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating exploitation or taking advantage of situations or individuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    earchers
    -0.88
    ignt
    -0.78
    cci
    -0.75
    usa
    -0.71
    execute
    -0.71
    cade
    -0.71
    allion
    -0.71
    heading
    -0.71
    ordering
    -0.71
     dstg
    -0.69
    POSITIVE LOGITS
     loopholes
    1.10
     opportunities
    1.07
     opportunity
    1.00
     loophole
    0.89
     vulnerabilities
    0.77
     existing
    0.76
     deductions
    0.73
     exemptions
    0.72
     tricks
    0.71
     protections
    0.71
    Act Density 0.070%

    No Known Activations