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

    Neuronpedia

    INDEX
    Explanations

    quoted text within quotes or directly mentioning quotes

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "quote" and related references to quotations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ichick
    -0.83
    gart
    -0.74
    ¯¯
    -0.73
    ikes
    -0.69
    vik
    -0.68
    ggles
    -0.68
    icz
    -0.67
    estate
    -0.67
    fare
    -0.66
    tails
    -0.64
    POSITIVE LOGITS
     quotes
    0.97
     quote
    0.93
     excerpts
    0.88
     scripture
    0.88
     verse
    0.88
     attributed
    0.87
     Verse
    0.87
     above
    0.85
     quotation
    0.85
     excerpt
    0.83
    Act Density 0.062%

    No Known Activations