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

    Neuronpedia

    INDEX
    Explanations

    instances where the directive to "consider" is used, often followed by a topic or proposition

    oai_token-act-pair · gpt-3.5-turbo

    phrases that prompt the reader to consider or analyze specific ideas or examples

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    "}],"
    -0.76
    "},"
    -0.72
    ially
    -0.70
    eal
    -0.64
    Contents
    -0.63
    etermined
    -0.62
    icipated
    -0.62
    dar
    -0.62
    ][/
    -0.60
    IAL
    -0.60
    POSITIVE LOGITS
    able
    0.77
     yourselves
    0.76
    tainment
    0.73
     Exhibit
    0.71
    amaz
    0.71
     examples
    0.71
    andum
    0.68
     Compare
    0.68
    asm
    0.67
     imagine
    0.65
    Act Density 0.095%

    No Known Activations