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

    Neuronpedia

    INDEX
    Explanations

    instances of specific letters followed by numbers, indicating a pattern related to location or categorization

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aptop
    -0.21
    ikes
    -0.19
    ots
    -0.19
    ike
    -0.17
    ocking
    -0.17
    inker
    -0.17
    abs
    -0.16
    ance
    -0.16
    isten
    -0.16
    ife
    -0.15
    POSITIVE LOGITS
    ichten
    0.20
    om
    0.18
    usat
    0.18
    lund
    0.17
    el
    0.17
    orraine
    0.17
    اÙĦØ©
    0.16
    .editor
    0.16
    ucc
    0.16
    assed
    0.16
    Act Density 0.034%

    No Known Activations