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

    Neuronpedia

    INDEX
    Explanations

    references to the word "Telegraph."

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    regn
    -0.67
    java
    -0.66
     Salvador
    -0.65
     Dillon
    -0.64
    rior
    -0.64
    laws
    -0.63
    GD
    -0.62
    atri
    -0.62
    lahoma
    -0.62
     Garn
    -0.61
    POSITIVE LOGITS
    legraph
    1.09
    ic
    0.82
     Tycoon
    0.81
    orld
    0.81
    orial
    0.79
    ically
    0.79
    aic
    0.78
    shire
    0.78
    iership
    0.76
    ical
    0.76
    Act Density 0.019%

    No Known Activations