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

    Neuronpedia

    INDEX
    Explanations

    time-related expressions such as durations and specific time intervals

    oai_token-act-pair · gpt-3.5-turbo

    references to time durations, specifically involving years and weeks

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     corridors
    -0.66
    ports
    -0.64
    estine
    -0.63
    acted
    -0.60
    ashtra
    -0.58
    ettings
    -0.58
     grav
    -0.58
    uctions
    -0.57
    jri
    -0.57
    ãĤ¢ãĥ«
    -0.57
    POSITIVE LOGITS
    long
    1.24
     ago
    1.03
    ozy
    0.90
    glass
    0.89
    gap
    0.86
    nn
    0.84
     Ago
    0.83
    dream
    0.82
     apiece
    0.81
    iversary
    0.80
    Act Density 0.094%

    No Known Activations