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

    Neuronpedia

    INDEX
    Explanations

    quotations within text

    oai_token-act-pair · gpt-3.5-turbo

    quotation marks and dialogue

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     eleph
    -0.85
     quir
    -0.83
     avatar
    -0.78
     occas
    -0.77
     pumped
    -0.77
     squid
    -0.76
     stra
    -0.76
     plaster
    -0.75
     subur
    -0.75
     booked
    -0.75
    POSITIVE LOGITS
    Let
    1.59
    Tonight
    1.58
    Look
    1.57
    If
    1.56
    Everybody
    1.55
    Clearly
    1.55
    I
    1.54
    Enough
    1.54
    Nobody
    1.54
    Absolutely
    1.53
    Act Density 0.137%

    No Known Activations