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

    Neuronpedia

    INDEX
    Explanations

    adversarial terms and contrasting situations

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "despite" and related phrases indicating contrast or opposition

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    soDeliveryDate
    -0.96
    emale
    -0.86
    rael
    -0.78
    uesday
    -0.73
    lda
    -0.72
    anca
    -0.71
    ovo
    -0.71
    ardless
    -0.69
    videos
    -0.68
    soever
    -0.67
    POSITIVE LOGITS
     warnings
    1.09
     setbacks
    1.02
     caveats
    1.00
     assurances
    0.98
     fact
    0.96
     setback
    0.96
     limitations
    0.92
     shortcomings
    0.91
     seeming
    0.85
     efforts
    0.85
    Act Density 0.146%

    No Known Activations