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

    Neuronpedia

    INDEX
    Explanations

    greetings and introductions

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    auer
    -0.66
     fail
    -0.66
    eele
    -0.66
     clauses
    -0.66
     neglect
    -0.62
     omit
    -0.62
    akespe
    -0.61
     dehuman
    -0.61
    creen
    -0.61
     underest
    -0.61
    POSITIVE LOGITS
     everyone
    0.93
     dear
    0.89
     folks
    0.88
     Everyone
    0.88
     ladies
    0.87
     fellow
    0.86
    ya
    0.85
     everybody
    0.84
    reetings
    0.83
     guys
    0.80
    Act Density 0.094%

    No Known Activations