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

    Neuronpedia

    INDEX
    Explanations

    words related to confidence and positive attributes

    oai_token-act-pair · gpt-3.5-turbo

    references to feelings of confidence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    pmwiki
    -0.88
    sites
    -0.76
    ĻĤ
    -0.73
    HEAD
    -0.70
    xon
    -0.69
    oxide
    -0.66
    Dispatch
    -0.65
    nice
    -0.64
     Antiqu
    -0.64
    bye
    -0.64
    POSITIVE LOGITS
    worthiness
    1.22
     intervals
    0.91
    lessly
    0.87
    fulness
    0.87
    lessness
    0.85
    worthy
    0.84
    iliate
    0.77
     interval
    0.77
     assurance
    0.75
     confidence
    0.75
    Act Density 0.038%

    No Known Activations