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

    Neuronpedia

    INDEX
    Explanations

    numerical sequences in a specific format

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    omial
    -1.20
    lar
    -1.14
    emonium
    -1.14
    ates
    -1.06
    atis
    -1.06
    riors
    -1.00
    ĸļ
    -1.00
    abase
    -1.00
    querade
    -1.00
    rals
    -0.98
    POSITIVE LOGITS
    ×ķ
    1.08
    mington
    0.94
    fps
    0.94
    ¯¯
    0.92
    ivery
    0.90
    brate
    0.84
    phrine
    0.84
    brates
    0.84
    izabeth
    0.83
    dfx
    0.81
    Act Density 1.910%

    No Known Activations