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

    Neuronpedia

    INDEX
    Explanations

    instances where the document mentions quantities or amounts of something

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "more" indicating comparisons or quantities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    xtap
    -0.91
    ĺħ
    -0.82
    Guard
    -0.73
    Fram
    -0.70
    ÃŁ
    -0.68
    atan
    -0.65
    Gra
    -0.64
    OCK
    -0.63
    DragonMagazine
    -0.63
    Enlarge
    -0.63
    POSITIVE LOGITS
     than
    1.41
     Than
    1.04
     ado
    0.98
    than
    0.94
     stringent
    0.91
     importantly
    0.87
     sophisticated
    0.80
     extensive
    0.79
     frequent
    0.79
     detailed
    0.74
    Act Density 0.137%

    No Known Activations