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

    Neuronpedia

    INDEX
    Explanations

    structured phrases indicating a recurring event or action

    oai_token-act-pair · gpt-3.5-turbo

    the phrase "Every 1" or similar repeated structures

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    dict
    -0.73
    hess
    -0.67
    olas
    -0.64
     Pru
    -0.60
    illions
    -0.59
     Jah
    -0.59
    ainment
    -0.58
     Metatron
    -0.55
     pron
    -0.55
     stripes
    -0.54
    POSITIVE LOGITS
    THING
    1.42
    where
    1.28
     conceivable
    1.04
    body
    1.02
    single
    0.97
    day
    0.95
    WHERE
    0.91
    Ĥª
    0.88
     single
    0.87
    things
    0.86
    Act Density 0.039%

    No Known Activations