© 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-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7150
    Prev
    Next
    INDEX
    Explanations

    SQL and programming code structures

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

    The neuron activates on code‐ or technology‐related keywords and identifiers (e.g. “List,” “SQL,” “JavaScript,” “Service,” “static,” “.htaccess”).

    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
    ぉ
    -1.98
    老闆
    -1.93
    הגב
    -1.84
    鷽
    -1.84
     protege
    -1.83
     picioare
    -1.81
     Vedi
    -1.80
    ього
    -1.80
    찔
    -1.77
     Protects
    -1.76
    POSITIVE LOGITS
     This
    2.09
     这
    2.06
    stwie
    1.98
    症
    1.98
    still
    1.98
    на
    1.92
     COULD
    1.89
    
    1.89
    DETAILED
    1.87
     этих
    1.84
    Activations Density 0.036%

    No Known Activations