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

    Neuronpedia

    INDEX
    Explanations

    instances where a sudden change or realization occurs

    oai_token-act-pair · gpt-3.5-turbo

    instances of sudden changes or unexpected events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    atu
    -0.82
    fman
    -0.77
    tein
    -0.75
    rf
    -0.74
    artisan
    -0.72
    iox
    -0.72
    76561
    -0.71
    agle
    -0.66
    conn
    -0.66
    icket
    -0.65
    POSITIVE LOGITS
    aneously
    1.04
     vanish
    0.93
     burst
    0.89
     suddenly
    0.89
     combust
    0.88
     vanished
    0.87
     disappeared
    0.86
     disappear
    0.85
     awakened
    0.81
     vanishing
    0.80
    Act Density 0.047%

    No Known Activations