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

    Neuronpedia

    INDEX
    Explanations

    dates, numbers, and quantities in the text

    oai_token-act-pair · gpt-3.5-turbo

    conjunctions and other connecting phrases or values in numerical contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    actionDate
    -0.85
    oris
    -0.78
     [/
    -0.71
    DragonMagazine
    -0.70
     IMAGES
    -0.69
    dict
    -0.67
    agine
    -0.67
    Ħ¢
    -0.67
    ledged
    -0.66
    Ãį
    -0.66
    POSITIVE LOGITS
     thereafter
    0.78
     vice
    0.75
     then
    0.69
     0
    0.69
     another
    0.69
     Guam
    0.63
     secondly
    0.62
     36
    0.61
     later
    0.60
     37
    0.60
    Act Density 0.362%

    No Known Activations