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

    Neuronpedia

    INDEX
    Explanations

    words related to news and journalism sources

    oai_token-act-pair · gpt-3.5-turbo

    names of various news publications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    sed
    -0.77
    etime
    -0.73
     retard
    -0.69
    itars
    -0.66
    vae
    -0.64
    ¯
    -0.64
    lessly
    -0.63
    NetMessage
    -0.62
    gone
    -0.62
    ĸļ
    -0.62
    POSITIVE LOGITS
     Newspaper
    1.06
     Herald
    1.02
     Tribune
    1.00
     Editorial
    1.00
     Newsp
    0.99
     editorial
    0.94
     Observer
    0.94
    Journal
    0.86
     Wire
    0.85
     Literary
    0.85
    Act Density 0.074%

    No Known Activations