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

    Neuronpedia

    INDEX
    Explanations

    phrases associated with specific characters or situations in a text

    oai_token-act-pair · gpt-3.5-turbo

    symbols or characters representing some form of emotional or dramatic emphasis

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Audrey
    -0.83
     Miss
    -0.82
     Debor
    -0.79
     Billy
    -0.78
     Harvey
    -0.76
     Hil
    -0.76
     minim
    -0.76
     unbeliev
    -0.75
     IPM
    -0.74
     Donna
    -0.74
    POSITIVE LOGITS
    ĺ
    2.18
    ĺħ
    1.38
    Ĺ
    1.35
    ĸ
    1.32
    IJ
    1.26
    Ķ
    1.23
    ¥
    1.21
    ķ
    1.19
    ¦
    1.17
    ħ
    1.15
    Act Density 0.075%

    No Known Activations