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

    Neuronpedia

    INDEX
    Explanations

    references to muffins and temperature measurement, especially the word "therm" with varying intensities

    oai_token-act-pair · gpt-3.5-turbo

    references to muffins and temperature measurements in cooking

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Defenders
    -0.77
     Proposition
    -0.75
    =-=-=-=-=-=-=-=-
    -0.74
    =-=-=-=-
    -0.74
     Corinthians
    -0.68
     Pwr
    -0.63
     Origin
    -0.62
    ostics
    -0.60
     Binding
    -0.60
     Supplement
    -0.59
    POSITIVE LOGITS
     muff
    1.05
    arella
    1.01
    onse
    0.98
    cakes
    0.94
    aton
    0.88
    rito
    0.86
    ule
    0.85
    ueller
    0.83
    ascript
    0.83
    alore
    0.82
    Act Density 0.010%

    No Known Activations