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

    Neuronpedia

    INDEX
    Explanations

    repeated instances of the word "as" indicating comparisons or analogies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    Parms
    -0.16
    vous
    -0.15
    èį·
    -0.15
    riers
    -0.14
    mant
    -0.14
    anner
    -0.14
    ovit
    -0.14
    ç´ł
    -0.14
     kaz
    -0.13
    rape
    -0.13
    POSITIVE LOGITS
    fat
    0.16
    ocz
    0.14
    yw
    0.14
    arr
    0.14
    resh
    0.13
     Yar
    0.13
    bsolute
    0.13
    ersions
    0.13
    æ¡
    0.13
    err
    0.13
    Act Density 0.082%

    No Known Activations