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

    Neuronpedia

    INDEX
    Explanations

    numbers and scientific terms related to charges, legislation, or emergencies

    oai_token-act-pair · gemini-2.0-flash

    police/scientific interactions

    np_max-act · gemini-2.0-flash

    The neuron strongly activates on rare, domain-specific technical or proper-name tokens—e.g. specialized scientific jargon and unusual named entities.

    oai_token-act-pair · o4-miniTriggered by @drejcpesjakpesjak
    New Auto-Interp
    Negative Logits
    StoryboardSegue
    -0.57
    الحياه
    -0.56
     onBind
    -0.49
    LookAnd
    -0.46
     فريبيس
    -0.45
     bere
    -0.44
    TypedDataSet
    -0.43
    AnchorTagHelper
    -0.43
    KommentareTeilen
    -0.43
    himovic
    -0.42
    POSITIVE LOGITS
     للاسماء
    0.86
    GOTREF
    0.68
    createStatement
    0.66
     ISNI
    0.65
    defn
    0.61
    AsUp
    0.61
     réaliste
    0.60
     Efq
    0.59
    ingtones
    0.59
     suono
    0.59
    Act Density 1.446%

    No Known Activations