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

    Neuronpedia

    INDEX
    Explanations

    phrases related to actions or situations such as administration or projects

    oai_token-act-pair · gpt-3.5-turbo

    negations and references to significant quantities or degrees

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     �
    -0.83
     ãĢĮ
    -0.75
     Dres
    -0.72
     Berks
    -0.66
     Phelps
    -0.66
     Goff
    -0.65
     Frey
    -0.64
    schild
    -0.64
     Pry
    -0.63
    ription
    -0.62
    POSITIVE LOGITS
    "
    1.40
    ",
    1.29
    "!
    1.27
    "âĢ¦
    1.25
    "?
    1.24
    usterity
    1.24
    ";
    1.23
    "—
    1.23
    %"
    1.23
    ".
    1.21
    Act Density 0.227%

    No Known Activations