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

    Neuronpedia

    INDEX
    Explanations

    the word "subject"

    oai_token-act-pair · gpt-3.5-turbo

    phrases related to subjects of documentaries or media coverage

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    frog
    -0.67
    fml
    -0.66
    pt
    -0.64
     condu
    -0.62
    MX
    -0.62
    ibr
    -0.62
    ept
    -0.62
    gob
    -0.60
    cup
    -0.60
    ptions
    -0.60
    POSITIVE LOGITS
    olls
    0.71
    international
    0.71
     ridicule
    0.65
     neutrality
    0.61
    æĺ¯
    0.60
     geopolitical
    0.60
     international
    0.59
     litigation
    0.59
    ãĥĩ
    0.58
     sear
    0.58
    Act Density 0.096%

    No Known Activations