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

    Neuronpedia

    INDEX
    Explanations

    numerical references to periods of time, particularly years

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "years" and anniversaries of events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iqu
    -0.71
     Mane
    -0.66
    rouse
    -0.65
    SET
    -0.63
    icum
    -0.63
    Leaks
    -0.62
    ople
    -0.61
    anche
    -0.59
    atered
    -0.58
     Scrib
    -0.57
    POSITIVE LOGITS
     ago
    1.18
     Ago
    1.01
     olds
    0.89
     old
    0.85
     hindsight
    0.82
     anniversary
    0.78
     honored
    0.77
     honoured
    0.75
     Anniversary
    0.71
     since
    0.70
    Act Density 0.076%

    No Known Activations