© 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. 6260
    Prev
    Next
    INDEX
    Explanations

    president or head of government

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

    The neuron activates on named entities—especially personal names and titles (e.g. “Prime Minister,” individual personal names, and other proper nouns).

    oai_token-act-pair · o4-miniTriggered 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
    妹妹
    0.64
     cello
    0.52
    姐姐
    0.51
     Muscle
    0.50
    Sister
    0.50
    纱
    0.49
    ungeon
    0.49
    orod
    0.49
     Leuven
    0.49
     Sister
    0.49
    POSITIVE LOGITS
     president
    1.42
     राष्ट्रपति
    1.41
     prezident
    1.39
     President
    1.38
     대통령
    1.38
    总统
    1.37
    President
    1.34
    राष्ट्रपति
    1.34
     presidente
    1.34
     президен
    1.33
    Activations Density 0.401%

    No Known Activations