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

    Neuronpedia

    INDEX
    Explanations

    news-related and rumor-related terms and phrases

    oai_token-act-pair · gpt-3.5-turbo

    phrases that indicate the occurrence of news reports or announcements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    urally
    -0.82
    uits
    -0.81
    ĸļ
    -0.78
    unte
    -0.78
    uit
    -0.77
    emale
    -0.75
    untarily
    -0.74
    olon
    -0.73
    cientious
    -0.73
    erto
    -0.73
    POSITIVE LOGITS
     Inquis
    1.03
     rumor
    1.02
    Rum
    1.01
    BBC
    0.97
     gossip
    0.97
     rumours
    0.96
    Sources
    0.94
     rumors
    0.93
     headlines
    0.92
     TMZ
    0.92
    Act Density 0.795%

    No Known Activations