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

    Neuronpedia

    INDEX
    Explanations

    phrases related to undercover operations or illicit activities

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Negative Logits
    ufact
    -0.99
    thur
    -0.66
    iasco
    -0.66
    ÄŁ
    -0.66
    ité
    -0.64
    tsky
    -0.64
     Scotia
    -0.64
     Seym
    -0.64
    kees
    -0.64
     Uz
    -0.62
    POSITIVE LOGITS
    rays
    1.18
    ray
    1.18
     sting
    1.14
    Ray
    0.95
    lers
    0.92
     Sting
    0.89
    bean
    0.83
    pots
    0.82
    ega
    0.80
    iest
    0.77
    Act Density 6.493%

    No Known Activations