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

    Neuronpedia

    INDEX
    Explanations

    words related to time, specifically focusing on periods of days

    oai_token-act-pair · gpt-3.5-turbo

    repeated mentions of "days" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    wig
    -0.71
    allo
    -0.66
    fan
    -0.63
    uana
    -0.63
    olph
    -0.62
    opter
    -0.62
    bender
    -0.62
    rey
    -0.61
    peer
    -0.60
    sonian
    -0.60
    POSITIVE LOGITS
     days
    3.66
     weeks
    2.47
     Days
    2.47
    Days
    2.26
     months
    2.23
     nights
    2.15
     hours
    2.12
    days
    1.95
     years
    1.79
     mornings
    1.78
    Act Density 0.025%

    No Known Activations