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

    Neuronpedia

    INDEX
    Explanations

    contractions of "do not" and "were not" as well as phrases indicating hesitation and skepticism

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
     Pierce
    -0.62
     Silk
    -0.61
     Moonlight
    -0.60
    ħĭ
    -0.58
     Shogun
    -0.57
     Showdown
    -0.57
     contrasted
    -0.57
     Rampage
    -0.56
    ©¶æ
    -0.55
     Saturn
    -0.55
    POSITIVE LOGITS
    existent
    1.09
    .","
    0.97
    hin
    0.89
     versa
    0.86
    existence
    0.84
    .?
    0.83
    MSN
    0.82
     thereof
    0.78
    epad
    0.77
     altogether
    0.76
    Act Density 0.724%

    No Known Activations