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

    Neuronpedia

    INDEX
    Explanations

    phrases with a specific symbol followed by characters, potentially indicating a specific type of coding or markup language

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of a specific token or character string, likely indicating a formatting or encoding issue in the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    glers
    -0.79
    orescence
    -0.78
    urat
    -0.77
    oresc
    -0.77
    abwe
    -0.76
    creen
    -0.70
     bluff
    -0.68
    lihood
    -0.67
    kered
    -0.67
    itud
    -0.67
    POSITIVE LOGITS
    0.92
    0.89
    Arsenal
    0.85
    Ö¼
    0.85
    everyone
    0.84
    Australia
    0.83
    âĢ¢âĢ¢
    0.82
    Kings
    0.81
    Leaks
    0.81
    Canada
    0.80
    Act Density 0.010%

    No Known Activations