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

    Neuronpedia

    INDEX
    Explanations

    dates in the format of numbers followed by a dash and another number

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of years in the format of "18XX" or "19XX"

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     beat
    -0.69
     bundled
    -0.64
     stomp
    -0.64
     laugh
    -0.62
     pasta
    -0.62
     tsunami
    -0.62
     guilt
    -0.60
     faster
    -0.60
     bang
    -0.60
     crashing
    -0.59
    POSITIVE LOGITS
    18
    3.24
    17
    2.43
    19
    2.36
    16
    2.19
    14
    2.10
    1900
    1.95
    22
    1.95
    15
    1.94
    28
    1.91
    12
    1.89
    Act Density 0.021%

    No Known Activations