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

    Neuronpedia

    INDEX
    Explanations

    comparative statements indicating something is quantitatively superior or inferior

    oai_token-act-pair · gpt-3.5-turbo

    phrases indicating a significant distance or disparity

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ortium
    -0.62
     Polo
    -0.59
     Turns
    -0.58
    piracy
    -0.58
     Dice
    -0.57
    ulus
    -0.56
    opoulos
    -0.55
     Revolutionary
    -0.55
     Correction
    -0.55
    ictions
    -0.55
    POSITIVE LOGITS
    med
    1.13
    fetched
    1.07
     outwe
    1.07
     outnumbered
    0.98
     superior
    0.94
     fewer
    0.91
     outweigh
    0.91
     worse
    0.89
    thing
    0.89
     preferable
    0.89
    Act Density 0.036%

    No Known Activations