© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 84473
    Prev
    Next
    INDEX
    Explanations

    programming language keywords

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

    The neuron is strongly activated by programming‐language declaration and import keywords and their accompanying syntax tokens (e.g. “declare,” “import” sub‐keywords like “absolute,” “division,” “print_function,” parentheses, semicolons, etc.).

    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
     fazenda
    -1.07
     feira
    -1.03
     Mujeres
    -0.98
     what
    -0.96
     because
    -0.96
    алы
    -0.94
     anticipate
    -0.93
     campesinos
    -0.91
     appropriately
    -0.91
     Reyes
    -0.91
    POSITIVE LOGITS
    ladolid
    1.20
    lorus
    1.15
    Questa
    1.13
     bestanden
    1.08
    udsman
    1.07
     était
    1.06
    1.05
     obligaciones
    1.05
    みが
    1.03
    こともあり
    1.03
    Activations Density 0.012%

    No Known Activations