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

    Neuronpedia

    INDEX
    Explanations

    This neuron appears to be activated by markers of news reporting or journalistic writing, particularly: 1. Dates and time references (e.g. "last Saturday", "in 2011", "today") 2. Names of news organizations or reporters (e.g. "

    oai_token-act-pair · claude-3-5-sonnet-20240620Triggered by @myinzi123

    instances of political violence or unrest

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    Follows a named news source

    np_acts-logits-general · gemini-2.0-flash

    news reports containing numbers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     TestBed
    -0.49
     Normdatei
    -0.46
    WithIOException
    -0.43
    DockStyle
    -0.43
    vedra
    -0.42
     tasche
    -0.42
    LAZY
    -0.42
     fourrure
    -0.42
    xodo
    -0.41
     logements
    -0.41
    POSITIVE LOGITS
     disambiguazione
    0.48
    pexpr
    0.47
     Multin
    0.46
     geopolitical
    0.44
     Reuters
    0.41
    RenderAtEndOf
    0.40
     multinational
    0.40
     Asia
    0.40
     politico
    0.39
     countries
    0.39
    Act Density 0.451%

    No Known Activations