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

    Neuronpedia

    INDEX
    Explanations

    negative qualifiers or negations in statements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ote
    -0.14
    ermo
    -0.14
    uzzle
    -0.14
    æĹ¢
    -0.14
    ç¸
    -0.14
    äch
    -0.14
    æĮ¯ãĤĬ
    -0.14
    nowrap
    -0.13
    fid
    -0.13
    -0.13
    POSITIVE LOGITS
     sure
    0.34
     necessarily
    0.30
    sure
    0.27
     Sure
    0.26
    withstanding
    0.24
     because
    0.23
    Sure
    0.23
     counting
    0.21
     much
    0.21
     least
    0.20
    Act Density 0.067%

    No Known Activations