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

    Neuronpedia

    INDEX
    Explanations

    news-related terms in sentences

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "reported," indicating references to news or announcements

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    isable
    -0.79
    creen
    -0.71
    venge
    -0.69
    ophone
    -0.67
    youtube
    -0.64
    gencies
    -0.64
     gauge
    -0.64
    ophers
    -0.63
    uv
    -0.63
    ococ
    -0.63
    POSITIVE LOGITS
     quoting
    0.84
     extensively
    0.74
     yesterday
    0.74
     ARTICLE
    0.70
     excerpts
    0.69
    biz
    0.68
     citing
    0.67
     reporter
    0.67
     headlined
    0.67
     reporters
    0.67
    Act Density 0.048%

    No Known Activations