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

    Neuronpedia

    INDEX
    Explanations

    Factual reporting

    np_max-act · gemini-2.0-flash

    This neuron activates on tokens associated with movie titles or film‐review contexts (e.g. title fragments and review words like “engrossing,” “The movie,” and quoted film names).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    тал
    -0.07
    .em
    -0.07
    $config
    -0.06
    .expect
    -0.06
     auctions
    -0.06
     fled
    -0.06
    Station
    -0.06
     futures
    -0.06
     Editors
    -0.06
    Fear
    -0.06
    POSITIVE LOGITS
    toString
    0.07
     pacientes
    0.07
    .TextBox
    0.07
     PodsDummy
    0.06
     problème
    0.06
    Spoiler
    0.06
     softened
    0.06
     smack
    0.06
    _pack
    0.06
     аналог
    0.06
    Act Density 0.193%

    No Known Activations