© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 8430
    Prev
    Next
    INDEX
    Explanations

    std:: and constexpr keywords

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

    The neuron fires on source‐code metadata lines (e.g. “Definition at line … of file …”, line-number prefixes and C/C++ doc comments) rather than the document’s natural language content.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    árny
    0.64
     sinc
    0.59
    giant
    0.59
     plaint
    0.58
    malloc
    0.58
     لوبی
    0.57
    <0x0F>
    0.57
     الطريقه
    0.56
    elder
    0.55
     intoxication
    0.55
    POSITIVE LOGITS
     std
    1.35
     constexpr
    1.27
    std
    1.19
    constexpr
    1.08
     BOOST
    0.95
     nullptr
    0.93
     Stands
    0.82
    nullptr
    0.80
     boost
    0.79
     [&
    0.78
    Activations Density 0.012%

    No Known Activations