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

    Neuronpedia

    INDEX
    Explanations

    references to cakes, desserts, and celebratory occasions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    nesota
    -0.84
    ually
    -0.81
    OSE
    -0.68
     Lomb
    -0.68
    iveness
    -0.68
    selves
    -0.67
    ostics
    -0.67
    ued
    -0.65
    GV
    -0.65
    audi
    -0.64
    POSITIVE LOGITS
     batter
    0.95
     cake
    0.91
    cake
    0.91
    cakes
    0.90
    meal
    0.90
     dough
    0.88
     cakes
    0.86
    pillar
    0.86
     recipe
    0.83
    cup
    0.81
    Act Density 0.055%

    No Known Activations