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

    Neuronpedia

    INDEX
    Explanations

    comparative phrases indicating a preference or difference

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     uſed
    -1.31
     purpoſe
    -1.27
     ſtate
    -1.26
     pleaſure
    -1.23
     poffe
    -1.22
     fevere
    -1.15
     houſe
    -1.15
     Efq
    -1.14
     raiſ
    -1.14
     myſelf
    -1.14
    POSITIVE LOGITS
    而非
    0.99
    而不是
    0.98
     not
    0.91
     than
    0.81
     rather
    0.79
     anstatt
    0.79
     وليس
    0.75
    //</
    0.75
     instead
    0.74
    ChildScrollView
    0.73
    Act Density 0.074%

    No Known Activations