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

    Neuronpedia

    INDEX
    Explanations

    foreign words containing accents

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    orial
    -0.74
    ually
    -0.69
    achine
    -0.66
    uality
    -0.65
    pread
    -0.65
    etsk
    -0.64
    omorph
    -0.63
    ationally
    -0.63
    uitous
    -0.63
    idences
    -0.60
    POSITIVE LOGITS
    rien
    1.05
    s
    0.88
    ÃįÃį
    0.82
    sid
    0.81
    sa
    0.79
    atl
    0.79
    rio
    0.78
    ng
    0.78
    dra
    0.76
    sand
    0.76
    Act Density 0.041%

    No Known Activations