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

    Neuronpedia

    INDEX
    Explanations

    title/headline

    np_max-act · gemini-2.0-flash

    This neuron activates on text that discusses or evaluates the article’s headline—especially in the context of clickbait or sensationalism.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8

    headlines that are misleading or sensational compared to the actual content of the article.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     фото
    -0.07
    Dam
    -0.07
     nod
    -0.06
    DOUBLE
    -0.06
    datal
    -0.06
     官
    -0.06
    กำหนด
    -0.06
     intents
    -0.06
     durum
    -0.06
    dict
    -0.06
    POSITIVE LOGITS
    -che
    0.08
    shire
    0.06
    	type
    0.06
     Responsive
    0.06
     disse
    0.06
    _tickets
    0.06
    _recursive
    0.06
     nose
    0.06
    .CenterScreen
    0.05
    (bodyParser
    0.05
    Act Density 0.007%

    No Known Activations