© 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. 5373
    Prev
    Next
    INDEX
    Explanations

    diabetes risk, Windows desktop, pine tree, lesbian individuals, Ruby code

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

    The neuron fires on content‐bearing nouns—especially key topic or entity words (e.g. “cow,” “milk,” “human,” “motorcycle,” “blood”) that signal the subject matter of a sentence.

    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
    Mitsubishi
    0.75
    만
    0.68
    Barcode
    0.68
    는
    0.66
    Garage
    0.66
     only
    0.65
    ),
    0.65
    Router
    0.64
    は
    0.64
    Storage
    0.64
    POSITIVE LOGITS
     hostilities
    0.77
     mollus
    0.66
     individualism
    0.66
     fascism
    0.65
     uprisings
    0.64
     extremism
    0.64
    ग्रियों
    0.64
     negli
    0.63
     materialism
    0.63
     idealism
    0.63
    Activations Density 0.708%

    No Known Activations