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

    Neuronpedia

    INDEX
    Explanations

    words related to physical or metaphorical holes

    oai_token-act-pair · gpt-3.5-turbo

    references to 'holes' and 'assholes' in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ERG
    -0.86
    Ô
    -0.83
    ione
    -0.76
    ency
    -0.76
    ivity
    -0.74
    ulet
    -0.71
    ENCY
    -0.71
    issance
    -0.69
    anche
    -0.69
    aic
    -0.67
    POSITIVE LOGITS
    hole
    1.40
    holes
    1.09
     Hole
    1.01
     hole
    0.91
    ocene
    0.88
    hog
    0.85
    izons
    0.79
     holes
    0.78
    crew
    0.72
    lord
    0.71
    Act Density 0.007%

    No Known Activations