© 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. 66196
    Prev
    Next
    INDEX
    Explanations

    closing curly braces and brackets

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

    This neuron detects occurrences of double curly‐brace template markers (e.g. “}}”) in templating or markup syntax.

    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
    很大的
    -2.52
    "
    -2.52
     The
    -2.39
     famed
    -2.30
     us
    -2.23
     Extensive
    -2.20
     emphas
    -2.17
     erhi
    -2.16
    gigabyte
    -2.16
     freaking
    -2.14
    POSITIVE LOGITS
    栳
    3.05
    僂
    2.61
    梘
    2.56
    涐
    2.53
    柃
    2.47
     anaknya
    2.33
    }}
    2.33
    所以我
    2.33
     })
    
    2.31
     offizielle
    2.31
    Activations Density 0.010%

    No Known Activations