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

    Neuronpedia

    INDEX
    Explanations

    phrases related to high quality or performance

    oai_token-act-pair · gpt-3.5-turbo

    descriptions of high quality or high-performance items

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ulpt
    -0.71
    uary
    -0.70
    edIn
    -0.69
     Reign
    -0.68
    ADRA
    -0.67
    pty
    -0.67
     Carib
    -0.66
    ¾
    -0.66
    ovie
    -0.66
     Canaver
    -0.65
    POSITIVE LOGITS
    density
    1.11
     density
    1.07
     voltage
    1.06
     dimensional
    1.04
     quality
    1.02
    speed
    1.00
     impedance
    1.00
     throughput
    1.00
     fidelity
    0.97
     intensity
    0.95
    Act Density 0.089%

    No Known Activations