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

    Neuronpedia

    INDEX
    Explanations

    references to uniqueness or novel attributes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    </em>
    -0.65
     censura
    -0.63
    <em>
    -0.62
    стма
    -0.62
    ber
    -0.57
     Mors
    -0.56
    so
    -0.56
    Để
    -0.55
    Bbb
    -0.55
     scold
    -0.55
    POSITIVE LOGITS
     UNIQUE
    1.49
     unique
    1.43
     Unique
    1.42
    unique
    1.40
    UNIQUE
    1.39
    UniqueId
    1.39
     uniqueness
    1.38
     uniques
    1.38
     uniqu
    1.36
    Unique
    1.34
    Act Density 0.066%

    No Known Activations