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

    Neuronpedia

    INDEX
    Explanations

    names or words with a recurring pattern "uz"

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    brates
    -0.75
    ADS
    -0.75
     Predator
    -0.74
     Yard
    -0.70
    riott
    -0.69
    brate
    -0.69
     plaque
    -0.68
    Interstitial
    -0.68
    itures
    -0.67
     solicitation
    -0.67
    POSITIVE LOGITS
    ombie
    1.12
    hao
    1.11
    hou
    1.06
    arro
    1.05
    uz
    0.97
    vu
    0.95
    illas
    0.92
    insk
    0.91
    alez
    0.90
    illa
    0.89
    Act Density 7.502%

    No Known Activations