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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 12735
    Prev
    Next
    INDEX
    Explanations

    countries to questions

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

    The neuron activates on full names of geopolitical entities (countries, states, regions, or governments).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    the phrase "Republic of" when referring to official country or state names.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    formal governmental or geopolitical entity names and titles, especially official “X of Y” jurisdictional constructions (states, republics, federations, parliaments, etc.)

    oai_token-act-pair · gpt-5Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Russland
    0.72
     США
    0.72
     Germany
    0.70
     Rusia
    0.68
     Russia
    0.68
     Россия
    0.68
     England
    0.67
     USA
    0.66
     Estados
    0.66
     Nebraska
    0.66
    POSITIVE LOGITS
    ь
    0.38
    튜
    0.38
    Mob
    0.37
    덫
    0.36
    Horizontal
    0.36
    з
    0.36
    дко
    0.36
    &
    0.36
    IEnumerable
    0.35
    cat
    0.35
    Activations Density 0.086%

    No Known Activations