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

    Neuronpedia

    INDEX
    Explanations

    phrases related to a strong degree of certainty or emphasis

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    atures
    -1.20
    yne
    -1.16
    neapolis
    -1.07
    imeters
    -0.99
    eways
    -0.95
    GW
    -0.95
    onut
    -0.94
    İĭ
    -0.93
    itures
    -0.91
     Correction
    -0.91
    POSITIVE LOGITS
     appreciated
    1.28
     obliged
    1.13
     appreci
    1.06
     resemble
    0.99
     distinguish
    0.97
     interested
    0.96
     welcome
    0.95
     influenced
    0.95
     regarded
    0.94
    ________________
    0.94
    Act Density 0.507%

    No Known Activations