© 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. 1978
    Prev
    Next
    INDEX
    Explanations

    public sector, national security

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

    The neuron responds to formal domain–specific content words and proper names—i.e. technical jargon, organization names, place names, and similar high-register tokens.

    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
    诡
    0.80
     backstory
    0.79
     ужа
    0.75
     passcode
    0.73
     deced
    0.70
    腥
    0.70
    搞笑
    0.69
     밝혔
    0.68
     मुस्लिमों
    0.68
     uptick
    0.66
    POSITIVE LOGITS
     patrim
    0.95
     sectoral
    0.95
     massif
    0.85
     interloc
    0.78
     panorama
    0.77
     magist
    0.76
     transversal
    0.76
     XNUMX
    0.75
     sector
    0.75
     cinemat
    0.73
    Activations Density 0.523%

    No Known Activations