© 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
    3. Gemma-3-12B
    5. 24-GEMMASCOPE-2-RES-16K
    7. 6726
    Prev
    Next
    INDEX
    Explanations

    U.S

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

    The neuron strongly activates on the U . S . abbreviation (tokens like “U,” “S,” and the surrounding periods), i.e. mentions of “U.S.”.

    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_24_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
     regiments
    0.99
     lipoproteins
    0.97
     waiters
    0.96
     outliers
    0.95
     deň
    0.95
    т
    0.95
     Eminem
    0.95
    0
    0.95
     Birla
    0.94
     magnitudes
    0.93
    POSITIVE LOGITS
     امریکا
    1.13
    પણે
    1.09
    ètres
    1.05
    了很多
    1.02
     آمریکا
    1.02
     متحده
    0.98
    ième
    0.95
    ີນ
    0.95
     अमेरिका
    0.94
    yon
    0.94
    Activations Density 0.056%

    No Known Activations