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

    Neuronpedia

    INDEX
    Explanations

    instances of quotation marks or apostrophes

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

    possessive or contraction endings

    np_acts-logits-general · gemini-2.5-flash-lite

    It detects apostrophe characters (the single-quote/curly quote) used in contractions and possessives.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Normdatei
    -1.02
     propOrder
    -0.81
    SharedDtor
    -0.80
     مشين
    -0.78
     kasarigan
    -0.75
     estekak
    -0.75
    featureID
    -0.74
     queſta
    -0.72
    niſſe
    -0.70
    ロウィン
    -0.69
    POSITIVE LOGITS
     the
    0.53
    The
    0.42
    ,
    0.38
    -
    0.35
     was
    0.34
     The
    0.34
     a
    0.31
    Of
    0.30
     (
    0.29
     is
    0.29
    Act Density 0.030%

    No Known Activations