© 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. 34-GEMMASCOPE-RES-131K
    7. 111001
    Prev
    Next
    INDEX
    Explanations

    float, double, decimal types

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

    This neuron fires on programming‐language syntax and identifiers—e.g. braces, parentheses, keywords, and CamelCase method or class names—i.e. it detects source code tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     サッカー
    -0.82
     Moriarty
    -0.81
     disinfection
    -0.79
     until
    -0.77
    Ẓ
    -0.77
    öffentlich
    -0.77
    minlength
    -0.74
    เมือง
    -0.74
    unikasi
    -0.73
     समय
    -0.72
    POSITIVE LOGITS
    double
    3.78
     double
    3.69
     float
    3.27
    float
    3.05
     Double
    2.81
    Double
    2.67
     doubles
    2.45
     DOUBLE
    2.41
    DOUBLE
    2.31
     Float
    2.25
    Activations Density 0.084%

    No Known Activations