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

    Neuronpedia

    INDEX
    Explanations

    a mix of characters, likely Cyrillic and Latin, with patterns of specific character sequences dispersed throughout the text

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    legate
    -0.66
    ulative
    -0.66
    pees
    -0.62
    rences
    -0.57
    olitan
    -0.57
    berman
    -0.56
    ulatory
    -0.54
    gency
    -0.54
    yrinth
    -0.53
    jriwal
    -0.53
    POSITIVE LOGITS
    heartedly
    0.61
    emic
    0.57
    е
    0.57
     Sabha
    0.56
    ONES
    0.55
    yy
    0.55
    III
    0.55
    åħ
    0.54
    chrom
    0.53
    td
    0.53
    Act Density 10.461%

    No Known Activations