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

    Neuronpedia

    INDEX
    Explanations

    verbs associated with decision-making or judgment

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Clockwork
    -0.71
    otide
    -0.69
     patiently
    -0.69
     floated
    -0.69
     maneu
    -0.66
     reckoned
    -0.65
     doubtless
    -0.65
     towed
    -0.62
    lease
    -0.61
    inis
    -0.60
    POSITIVE LOGITS
     disrespect
    0.87
     enough
    0.80
     discrimination
    0.74
     somehow
    0.74
     unfairly
    0.73
     anything
    0.70
     racism
    0.70
     \'
    0.70
     discriminatory
    0.68
     harm
    0.68
    Act Density 0.592%

    No Known Activations