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

    Neuronpedia

    INDEX
    Explanations

    Places (cities and states)

    np_max-act · gemini-2.0-flash

    address information formatted in JSON.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on geographic place names (cities, states/provinces, directional qualifiers like “West”) in the address strings.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ınızda
    -0.07
    Chooser
    -0.07
    μί
    -0.07
    .AddDays
    -0.07
    TableRow
    -0.06
     Scores
    -0.06
    DOCUMENT
    -0.06
    Kill
    -0.06
     Fore
    -0.06
    EEE
    -0.06
    POSITIVE LOGITS
    τις
    0.07
     vis
    0.07
    иб
    0.06
    =torch
    0.06
     slag
    0.06
    aca
    0.06
    _sched
    0.06
    ocr
    0.06
    	BIT
    0.06
    _RESOLUTION
    0.06
    Act Density 0.017%

    No Known Activations