© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 27944
    Prev
    Next
    INDEX
    Explanations

    class definitions and imports

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

    This neuron detects opening curly‐brace tokens “{” that start code blocks (e.g. class or method definitions).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     boneco
    -1.01
     文化
    -0.98
     urodz
    -0.94
     otoño
    -0.92
     anunciou
    -0.91
    唖
    -0.91
    cM
    -0.89
    alcool
    -0.88
     Après
    -0.85
     државе
    -0.85
    POSITIVE LOGITS
    utilizzo
    1.03
    ynchronously
    1.02
    bidities
    1.01
    hamed
    1.00
    Lalu
    0.99
     пун
    0.98
    stays
    0.98
     Repub
    0.97
    geth
    0.96
    ynchronous
    0.94
    Activations Density 0.004%

    No Known Activations