© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 107345
    Prev
    Next
    INDEX
    Explanations

    United States or USA

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

    The neuron fires on tokens belonging to formal metadata—especially official titles, affiliations, section headings, addresses, and their accompanying numbers (e.g. “Vice President,” “of,” “the,” geographic or institutional names, and ordinal or street numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Europe
    -1.09
     Europe
    -0.92
     Britons
    -0.87
    日本の
    -0.85
     اروپا
    -0.83
    台灣
    -0.83
     Europeans
    -0.82
    Australian
    -0.82
     domes
    -0.81
    Ireland
    -0.80
    POSITIVE LOGITS
     United
    1.79
     USA
    1.59
    United
    1.36
     UNITED
    1.34
    united
    1.23
    USA
    1.16
     united
    1.13
    UNITED
    1.07
     США
    0.96
    usa
    0.91
    Activations Density 0.101%

    No Known Activations