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

    Neuronpedia

    INDEX
    Explanations

    occurrences of the word "Every" or related expressions emphasizing frequency or recurrence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    midi
    -0.16
    itive
    -0.15
    åįĪ
    -0.14
     whatever
    -0.14
    orgot
    -0.14
    ophil
    -0.14
    culos
    -0.14
    ade
    -0.14
    ovsky
    -0.14
    lete
    -0.13
    POSITIVE LOGITS
    ones
    0.32
    THING
    0.29
    things
    0.27
    .single
    0.27
    where
    0.26
    thin
    0.26
    -single
    0.25
    ONES
    0.25
    BODY
    0.25
    -other
    0.23
    Act Density 0.050%

    No Known Activations