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

    Neuronpedia

    INDEX
    Explanations

    thriller

    np_max-act · gemini-2.0-flash

    This neuron responds to mentions of movie genre labels—especially “thriller” (and related terms like “mystery,” “action thriller,” “financial thriller,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     دانشجوی
    -0.07
     coco
    -0.07
     railways
    -0.07
     skateboard
    -0.06
     Gale
    -0.06
     praying
    -0.06
     Railway
    -0.06
    坐
    -0.06
    umno
    -0.06
     propagated
    -0.06
    POSITIVE LOGITS
    нов
    0.07
     densely
    0.06
    isas
    0.06
     WIDTH
    0.06
    ัณฑ
    0.06
     hr
    0.06
    ながら
    0.06
     QS
    0.06
     phosphory
    0.06
     jih
    0.06
    Act Density 0.045%

    No Known Activations