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

    Neuronpedia

    INDEX
    Explanations

    phrases related to political activities and processes

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
     accomp
    -0.50
     Lomb
    -0.46
     measuring
    -0.45
     nodd
    -0.45
     sprink
    -0.45
     differe
    -0.43
     ank
    -0.42
     mot
    -0.42
     tagging
    -0.42
     advant
    -0.42
    POSITIVE LOGITS
     for
    1.15
    for
    0.87
    For
    0.78
     FOR
    0.69
     For
    0.69
    ioned
    0.65
    fort
    0.64
    irection
    0.62
    untarily
    0.60
     to
    0.60
    Act Density 0.525%

    No Known Activations