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

    Neuronpedia

    INDEX
    Explanations

    years or dates represented by four digits, such as 20XX

    oai_token-act-pair · gpt-3.5-turbo

    references to the 20th century

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    plom
    -0.80
    pora
    -0.74
    afort
    -0.74
    ño
    -0.71
    kell
    -0.70
    hei
    -0.70
    uana
    -0.69
    ablo
    -0.69
    henko
    -0.69
    abolic
    -0.68
    POSITIVE LOGITS
    âĸĪâĸĪ
    1.00
    committee
    0.85
    th
    0.83
    skirts
    0.80
    ISH
    0.78
    isher
    0.77
    ishers
    0.77
    mph
    0.75
    SW
    0.75
     minutes
    0.74
    Act Density 0.060%

    No Known Activations