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

    Neuronpedia

    INDEX
    Explanations

    symbols, dates, and punctuation marks in structured text

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

    Ill.Dec. citation format

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    <unused74>
    -0.66
    <unused8>
    -0.66
    <unused43>
    -0.66
    <unused52>
    -0.66
    <unused42>
    -0.66
    [@BOS@]
    -0.66
    <unused16>
    -0.66
    <unused23>
    -0.66
    <unused14>
    -0.66
    <unused3>
    -0.66
    POSITIVE LOGITS
     Zivil
    0.29
     igualdad
    0.28
     potencia
    0.28
    Take
    0.28
    #
    0.28
     totalidad
    0.27
     <<<<<<<<<<<<<<
    0.27
    take
    0.26
    setContentType
    0.26
    Top
    0.26
    Act Density 0.001%

    No Known Activations