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

    Neuronpedia

    INDEX
    Explanations

    the word "therefore" as a signal of conclusion or inference in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ing
    -0.71
    š
    -0.60
    man
    -0.60
    hæ
    -0.58
    hol
    -0.56
    gan
    -0.55
    /'.$
    -0.55
    dag
    -0.54
     SH
    -0.53
    dup
    -0.52
    POSITIVE LOGITS
     therefore
    3.42
     Therefore
    2.97
    Therefore
    2.92
    therefore
    2.84
     therefor
    2.53
     hence
    2.21
    Поэтому
    2.18
     donc
    2.14
     portanto
    2.13
     thus
    2.10
    Act Density 0.128%

    No Known Activations