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

    Neuronpedia

    INDEX
    Explanations

    the word "uncomfortable" or variations of it

    oai_token-act-pair · gpt-3.5-turbo

    instances of discomfort or unease in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ership
    -0.90
    ework
    -0.89
    onne
    -0.78
    frey
    -0.77
    alez
    -0.75
    ribution
    -0.75
    ebook
    -0.75
    onz
    -0.74
    src
    -0.74
    ramid
    -0.73
    POSITIVE LOGITS
     uncomfortable
    1.25
     discomfort
    1.09
     tiss
    0.92
     awkward
    0.85
     unbeliev
    0.79
     adolesc
    0.78
     earthqu
    0.78
     scrut
    0.78
     cumbers
    0.78
     uproar
    0.76
    Act Density 0.008%

    No Known Activations