© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 111643
    Prev
    Next
    INDEX
    Explanations

    various technical and natural language snippets

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

    XML closing tags.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    The neuron fires on XML/HTML‐style closing tags (i.e. tokens like “</message>,” “</translation>,” “</context>,” etc.).

    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
    炁
    -1.70
     ImportError
    -1.70
     zeer
    -1.65
     municípios
    -1.63
    🕵
    -1.63
    
    -1.63
    🔮
    -1.62
    ynchronously
    -1.57
     estamp
    -1.57
     for
    -1.56
    POSITIVE LOGITS
                
    2.08
    Many
    1.82
           
    1.81
    Our
    1.80
     Если
    1.77
     Sometimes
    1.69
             
    1.69
         
    1.66
    In
    1.66
     There
    1.66
    Activations Density 0.002%

    No Known Activations