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

    Neuronpedia

    INDEX
    Explanations

    the number '2' occurring in percentages

    oai_token-act-pair · gpt-3.5-turbo

    numerical data, particularly percentages

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    urat
    -0.61
    etch
    -0.60
     Tate
    -0.60
    imaru
    -0.56
    sed
    -0.56
    laws
    -0.55
    atown
    -0.55
    akeru
    -0.55
    channelAvailability
    -0.54
    avascript
    -0.54
    POSITIVE LOGITS
    ½
    1.19
     ½
    0.95
    %-
    0.91
    ·
    0.91
    %
    0.91
     percent
    0.89
    270
    0.89
     percentage
    0.89
    120
    0.88
    76
    0.87
    Act Density 0.089%

    No Known Activations