© 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. 50385
    Prev
    Next
    INDEX
    Explanations

    code snippets and markup

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

    This neuron responds to programming‐style identifiers (especially snake_case names like page_manager_tasks, build/sbt, and menu_manage), i.e. code tokens rather than natural language words.

    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
     tráfico
    -0.78
    نح
    -0.77
     εφαρ
    -0.76
    Estilo
    -0.76
    usion
    -0.75
    ⽣
    -0.74
    AMAZING
    -0.73
     آمده
    -0.73
    Índice
    -0.73
     APPELLANT
    -0.73
    POSITIVE LOGITS
    Flash
    0.92
    isie
    0.85
    ureusement
    0.85
    笑着
    0.85
    πο
    0.83
    saida
    0.83
    urie
    0.82
    Jeremy
    0.81
     dopo
    0.79
    вић
    0.77
    Activations Density 0.005%

    No Known Activations