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

    Neuronpedia

    INDEX
    Explanations

    contractions

    oai_token-act-pair · gpt-3.5-turbo

    contractions of "it is" and similar phrases indicating statements or facts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     devise
    -0.64
     unsuccessfully
    -0.62
    ociate
    -0.62
    naires
    -0.61
    disabled
    -0.61
    rouse
    -0.61
     monop
    -0.60
    ]+
    -0.60
    ansas
    -0.60
     handle
    -0.59
    POSITIVE LOGITS
     raining
    1.00
     tempting
    0.92
    estern
    0.92
     dusk
    0.91
     dawn
    0.87
     imperative
    0.86
     inevitable
    0.82
     bitters
    0.80
     easy
    0.79
     TIME
    0.78
    Act Density 0.152%

    No Known Activations