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

    Neuronpedia

    INDEX
    Explanations

    the word "divergent" or variations of the word

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    EEK
    -0.75
     Introduced
    -0.66
    è¦ļéĨĴ
    -0.65
    ammy
    -0.62
    ucky
    -0.61
    oggle
    -0.61
    HAEL
    -0.60
    AA
    -0.60
    CHA
    -0.60
    ellen
    -0.60
    POSITIVE LOGITS
    gent
    1.32
    ging
    1.16
    gencies
    1.10
    ministic
    1.09
    gently
    1.03
    ged
    1.00
    gence
    0.99
    tic
    0.99
    gency
    0.97
    ming
    0.93
    Act Density 0.023%

    No Known Activations