© 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. 104291
    Prev
    Next
    INDEX
    Explanations

    #define and code declarations

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

    The neuron fires on C/C++ preprocessor directive tokens—especially “#define” and related macro‐definition keywords.

    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
     presença
    -0.97
     Giappone
    -0.95
     katze
    -0.91
    parken
    -0.91
    neous
    -0.90
     possibilidade
    -0.89
    upaten
    -0.88
     ouvir
    -0.87
    そんな
    -0.86
     tailles
    -0.86
    POSITIVE LOGITS
     at
    1.30
     if
    1.03
     for
    1.02
     by
    1.02
     from
    0.91
     because
    0.91
     innebär
    0.89
     XXX
    0.89
    NONE
    0.88
     WHO
    0.88
    Activations Density 0.081%

    No Known Activations