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

    Neuronpedia

    INDEX
    Explanations

    specific times in a day formatted with a colon

    oai_token-act-pair · gpt-3.5-turbo

    time expressions formatted as hours and minutes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    dinand
    -0.75
     behav
    -0.68
     extermin
    -0.67
     culprit
    -0.66
     trouble
    -0.64
    ĸļ
    -0.64
    ãĤ©
    -0.62
     lov
    -0.61
     authority
    -0.61
     brute
    -0.60
    POSITIVE LOGITS
    00
    1.16
    30
    1.15
    59
    1.13
    45
    1.09
    09
    1.05
    58
    1.04
    56
    1.03
    53
    1.03
    54
    1.03
    55
    1.02
    Act Density 0.048%

    No Known Activations