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

    Neuronpedia

    INDEX
    Explanations

    punctuation marks and various quotation styles

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ^(@)
    -1.37
     -"
    -1.21
     Jefus
    -1.19
     Efq
    -1.19
     doubtnut
    -1.15
     ...'
    -1.13
     chofe
    -1.10
     Chrift
    -1.08
     pleaf
    -1.08
     '"
    -1.08
    POSITIVE LOGITS
     “
    1.86
    ’
    1.83
    ”
    1.71
     ‘
    1.64
    “
    1.60
    ’,
    1.57
    .’
    1.52
    .”
    1.49
    ,”
    1.49
    ’.
    1.48
    Act Density 0.707%

    No Known Activations