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

    Neuronpedia

    INDEX
    Explanations

    names of people or places with accented characters

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     Ashes
    -0.65
     Classics
    -0.63
     candles
    -0.62
     wedge
    -0.60
     Dunk
    -0.59
     aggrav
    -0.59
     drawer
    -0.59
     kernels
    -0.59
    ++++
    -0.58
     Athena
    -0.57
    POSITIVE LOGITS
    rio
    1.34
    ñ
    1.02
    vez
    1.02
    ndum
    0.99
    ctica
    0.98
    rez
    0.95
    ez
    0.94
    ï
    0.92
    lez
    0.90
    zar
    0.90
    Act Density 0.023%

    No Known Activations