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

    Neuronpedia

    INDEX
    Explanations

    comparisons indicating an increase in intensity or significance

    oai_token-act-pair · gpt-3.5-turbo

    phrases emphasizing incremental increases or intensifications of qualities or situations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    gio
    -0.88
    dinand
    -0.73
    meet
    -0.70
    poke
    -0.69
    medium
    -0.68
    ãĤ¨ãĥ«
    -0.67
    eways
    -0.67
    atan
    -0.67
    istrates
    -0.67
    icides
    -0.67
    POSITIVE LOGITS
     impressive
    1.05
     damning
    1.04
     egregious
    1.02
     astounding
    0.99
     bizarre
    0.96
     confounding
    0.96
     frustrating
    0.96
     precarious
    0.96
     perplex
    0.95
     stringent
    0.95
    Act Density 0.085%

    No Known Activations