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

    Neuronpedia

    INDEX
    Explanations

    dates written using the format of beginning of the numbers followed by a space and the ending of the numbers

    oai_token-act-pair · gpt-3.5-turbo

    specific numbers or dates, particularly those related to historical events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ical
    -0.80
    orter
    -0.78
    iques
    -0.77
    ickers
    -0.75
    itional
    -0.74
    ete
    -0.72
    ovie
    -0.71
    iguous
    -0.71
    ional
    -0.71
    ically
    -0.70
    POSITIVE LOGITS
    650
    1.05
    mph
    0.87
    91
    0.87
    th
    0.85
    58
    0.84
    92
    0.83
    85
    0.83
    71
    0.82
    94
    0.82
    87
    0.82
    Act Density 0.037%

    No Known Activations