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

    Neuronpedia

    INDEX
    Explanations

    phrases related to illegal or criminal activities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot
    New Auto-Interp
    Negative Logits
    oleon
    -0.86
    etics
    -0.82
    utra
    -0.82
    ĸļ
    -0.78
    oras
    -0.78
    ois
    -0.76
    bourg
    -0.75
    hetti
    -0.75
    chel
    -0.75
    ophers
    -0.73
    POSITIVE LOGITS
     downloading
    0.94
    ities
    0.91
     dumping
    0.88
     manslaughter
    0.87
     substances
    0.83
     Parenthood
    0.83
     alien
    0.81
     bystand
    0.80
     pregnancies
    0.79
     conduct
    0.79
    Act Density 1.070%

    No Known Activations