© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 129571
    Prev
    Next
    INDEX
    Explanations

    The neuron flags code‐structuring tokens in Android/Java source—especially class declarations and layout/resource identifiers (e.g. “class,” “layout,” “activity,” numeric resource IDs).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    android activity main

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    子さん
    -1.07
    subsubsection
    -1.02
    Карьера
    -1.01
    より
    -1.00
     also
    -1.00
     olyan
    -0.97
     from
    -0.92
     nuevamente
    -0.92
    سياسي
    -0.91
     именно
    -0.91
    POSITIVE LOGITS
    forRoot
    1.34
     first
    1.07
    கம்
    1.04
     bross
    1.02
    خول
    1.01
    falz
    1.01
    torio
    1.01
    娛
    1.00
    鱂
    1.00
     estekak
    0.98
    Activations Density 0.018%

    No Known Activations