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

    Neuronpedia

    INDEX
    Explanations

    temporal references, such as durations like "few weeks ago" and "years back"

    oai_token-act-pair · gpt-3.5-turbo

    expressions of time frames, specifically references to weeks or months ago

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    acus
    -0.70
     bidder
    -0.65
    emale
    -0.65
    ubis
    -0.62
    oux
    -0.61
    isine
    -0.61
    ãģĻ
    -0.58
     Abyssal
    -0.58
    iola
    -0.58
    ursed
    -0.57
    POSITIVE LOGITS
     ago
    1.40
    pring
    1.21
     Ago
    1.00
    catentry
    0.85
     beforehand
    0.75
    hips
    0.75
     posterior
    0.71
    ilk
    0.70
     back
    0.69
     shy
    0.69
    Act Density 0.076%

    No Known Activations