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

    Neuronpedia

    INDEX
    Explanations

    locations and descriptive place words

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron activates strongly for place names and geographic/location words (proper nouns referring to cities, regions, or locations).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     NumPy
    0.24
     Questi
    0.22
     Kry
    0.22
    ாய்ச்ச
    0.22
    govern
    0.22
     Analysis
    0.22
     Maui
    0.22
    ially
    0.21
     Tacoma
    0.21
     instantiated
    0.21
    POSITIVE LOGITS
     picturesque
    0.33
     downtown
    0.32
     promenade
    0.32
     parku
    0.30
     fiume
    0.30
     stazione
    0.28
     promen
    0.28
     avenida
    0.27
     plaza
    0.27
    公園
    0.27
    Act Density 0.239%

    No Known Activations