© 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. 9439
    Prev
    Next
    INDEX
    Explanations

    enum type choices

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

    The neuron is highly activated by programming‐language definition keywords and constructs (e.g. “function,” “enum,” “public,” the definition‐site parenthesis, etc.), i.e. tokens that introduce or declare code entities.

    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
    staden
    -1.12
    𝕙
    -0.93
    ğlık
    -0.90
    ]-\
    -0.89
     diamantes
    -0.89
    +```
    -0.89
     Kindern
    -0.86
    addListener
    -0.85
     inserire
    -0.85
     reduzir
    -0.85
    POSITIVE LOGITS
     enum
    1.25
     choices
    1.18
     выбора
    1.02
     Enum
    1.02
     ahora
    1.00
     Choices
    0.99
    Enum
    0.99
     we
    0.94
     枚
    0.94
    枚
    0.92
    Activations Density 0.020%

    No Known Activations