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

    Neuronpedia

    INDEX
    Explanations

    actions or tasks that should be prioritized or done before anything else

    oai_token-act-pair · gpt-3.5-turbo

    the word "first" and its variations to indicate priority or sequence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gould
    -0.65
    stead
    -0.63
     Sher
    -0.59
    âĶģ
    -0.59
    Quantity
    -0.59
     Doomsday
    -0.59
    pher
    -0.58
     Ell
    -0.57
     Tant
    -0.57
    IRC
    -0.55
    POSITIVE LOGITS
     responders
    1.15
     baseman
    1.14
     thing
    0.91
    hand
    0.88
     glance
    0.82
    aneously
    0.81
    ly
    0.80
     impression
    0.80
    ially
    0.79
     blush
    0.78
    Act Density 0.043%

    No Known Activations