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

    Neuronpedia

    INDEX
    Explanations

    specific numbers or quantities that are related to programs or projects, such as "four"

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "four" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    UTION
    -0.69
    idence
    -0.65
    anwhile
    -0.59
     Rica
    -0.59
    geist
    -0.57
    vier
    -0.57
    brance
    -0.56
     compr
    -0.56
    atcher
    -0.56
    Fed
    -0.56
    POSITIVE LOGITS
    teenth
    2.09
    teen
    1.93
    some
    1.42
    eenth
    1.17
    square
    1.17
    fold
    1.08
    een
    1.02
    acity
    0.98
     hundred
    0.96
     Seasons
    0.96
    Act Density 0.051%

    No Known Activations