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

    Neuronpedia

    INDEX
    Explanations

    names or locations containing the specific string "ju"

    oai_token-act-pair · gpt-3.5-turbo

    the frequent occurrence of specific syllables or sounds in words

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    tons
    -0.85
     Spectrum
    -0.76
     Regulations
    -0.68
     Orbit
    -0.66
    riad
    -0.66
     Horizons
    -0.66
     Advent
    -0.64
    space
    -0.61
    graduate
    -0.60
     Passenger
    -0.60
    POSITIVE LOGITS
    xtap
    1.37
    ju
    1.14
    ulkan
    1.02
    arez
    0.92
    ppa
    0.92
    velength
    0.91
    ice
    0.91
    ipers
    0.90
    amph
    0.88
    peror
    0.88
    Act Density 0.012%

    No Known Activations