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

    Neuronpedia

    INDEX
    Explanations

    references to reminiscing or reflecting on past events

    oai_token-act-pair · gpt-3.5-turbo

    instances of the phrase "looking back."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     viz
    -0.76
     Hague
    -0.67
    ises
    -0.61
     Liver
    -0.61
    anto
    -0.60
    furt
    -0.59
    icular
    -0.58
    ijn
    -0.58
     pollut
    -0.58
    entric
    -0.57
    POSITIVE LOGITS
     nostalg
    1.17
    wards
    1.01
    gam
    0.95
    dated
    0.95
    packs
    0.92
    trace
    0.88
    stab
    0.86
    glass
    0.85
    WARD
    0.85
    stories
    0.79
    Act Density 0.022%

    No Known Activations