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

    Neuronpedia

    INDEX
    Explanations

    questions and interrogative phrases

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

    question about advice

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

    spots dialogue/turn markers—quoted speech, speaker labels (e.g., "A:", "BD:"), and punctuation that marks questions/answers or turn boundaries.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     supplémentaire
    -0.35
    ku
    -0.34
    bleven
    -0.34
     financeiro
    -0.33
    ValueStyle
    -0.32
    śred
    -0.32
     prévenir
    -0.32
     épais
    -0.31
     Libertad
    -0.31
     nadzieję
    -0.30
    POSITIVE LOGITS
     AssemblyCulture
    0.73
     betweenstory
    0.73
    ########.
    0.69
    MessageOf
    0.68
    +#+
    0.65
    ⤹
    0.65
    0.64
    <?
    0.63
    transQ
    0.62
     invokingState
    0.61
    Act Density 0.010%

    No Known Activations