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

    Neuronpedia

    INDEX
    Explanations

    the beginning markers or indicators of new sections in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     
    -0.69
     (
    -0.66
     I
    -0.58
     what
    -0.55
    (
    -0.53
     someone
    -0.52
     pretty
    -0.51
     may
    -0.51
    :
    -0.50
     *
    -0.50
    POSITIVE LOGITS
    IsContent
    1.07
     estekak
    1.05
    ########.
    1.04
    脚注の使い方
    1.03
     betweenstory
    1.01
    StructEnd
    1.00
     nahilalakip
    0.99
    parsedMessage
    0.98
    __":
    0.95
    Бахар
    0.94
    Act Density 0.075%

    No Known Activations