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

    Neuronpedia

    INDEX
    Explanations

    sentences that contain a statement about what does not occur or apply

    oai_token-act-pair · gpt-3.5-turbo

    statements indicating negation or the absence of something

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     palms
    -0.72
    hog
    -0.69
    case
    -0.69
    )=(
    -0.65
     Hok
    -0.63
    Reviewer
    -0.63
    cised
    -0.63
     Ages
    -0.63
     Handling
    -0.61
     Methods
    -0.61
    POSITIVE LOGITS
    ppel
    1.08
    omsday
    0.99
    herty
    0.94
    oms
    0.92
     indeed
    0.88
    vet
    0.87
    pez
    0.86
     not
    0.85
    ozy
    0.85
    anos
    0.81
    Act Density 0.117%

    No Known Activations