© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 9147
    Prev
    Next
    INDEX
    Explanations

    JSON data structures

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

    The neuron fires on JSON‐style structural characters (quotes, braces, brackets, etc.), i.e. it detects JSON syntax elements.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    );
    -2.02
     "";
    -1.73
     –
    -1.73
    )");
    -1.65
    垆
    -1.63
    			
    -1.58
    </i>
    -1.56
    ↵
    -1.55
    і
    -1.55
     száz
    -1.54
    POSITIVE LOGITS
     farba
    1.66
    УДК
    1.65
     ilumina
    1.64
     Fuerzas
    1.64
    ៚
    1.63
     there
    1.63
    —,
    1.54
    縂
    1.54
     plena
    1.45
    heute
    1.45
    Activations Density 0.029%

    No Known Activations