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

    Neuronpedia

    INDEX
    Explanations

    _location

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on capitalized proper nouns and named entities (e.g. company, product, or person names) in the text.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     burge
    -0.07
     кот
    -0.06
     منطقة
    -0.06
     �
    -0.06
    getTime
    -0.06
    rosse
    -0.06
    dí
    -0.06
     Rhe
    -0.06
    Hit
    -0.06
     Rit
    -0.06
    POSITIVE LOGITS
    user
    0.08
     authorised
    0.07
    middlewares
    0.06
     outros
    0.06
     viv
    0.06
     notre
    0.06
    Hooks
    0.06
     endregion
    0.06
     turbines
    0.06
    ують
    0.06
    Act Density 1.897%

    No Known Activations