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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "knows" or related forms in context

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    knows how or why

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    存于互联网档案馆
    -0.52
    лючение
    -0.45
     stället
    -0.43
    ofday
    -0.42
    Conclusão
    -0.41
    йлер
    -0.41
     christlichen
    -0.40
    //
    -0.40
    LikeLike
    -0.39
     vilja
    -0.39
    POSITIVE LOGITS
    knows
    0.96
     Knows
    0.94
     knows
    0.90
    Knows
    0.75
    airs
    0.65
    cares
    0.65
    sees
    0.65
    rawDesc
    0.62
    angs
    0.62
    cels
    0.61
    Act Density 0.008%

    No Known Activations