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

    Neuronpedia

    INDEX
    Explanations

    initial document boundary markers and beginning‐of‐sequence cues

    oai_token-act-pair · o3-miniTriggered by @yemengyus1

    instances of punctuation or quotation marks in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Shakspeare
    -1.70
     myſelf
    -1.70
     Efq
    -1.64
     Monfieur
    -1.60
     itſelf
    -1.51
     Theſe
    -1.48
     Houſe
    -1.45
     ―――――
    -1.44
     Majefty
    -1.43
     photolibrary
    -1.43
    POSITIVE LOGITS
     the
    1.23
     a
    1.06
     in
    0.97
     and
    0.93
     an
    0.92
    ,
    0.90
     on
    0.89
     as
    0.88
     it
    0.86
    -
    0.86
    Act Density 0.672%

    No Known Activations