© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 14601
    Prev
    Next
    INDEX
    Explanations

    android xml and gradle paths

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

    The neuron is triggered by non-natural-language code/configuration tokens—things like uppercase acronyms (e.g. APK), file- and tool-names (apktool, jadx), XML/namespace attributes (xmlns), module/file path segments, version numbers, and other programming‐related symbols.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
    ınıza
    -1.54
     mascarilla
    -1.54
    gemaakt
    -1.52
     vendido
    -1.45
     nélk
    -1.45
     rocas
    -1.39
     taxas
    -1.39
    şil
    -1.34
     Heilige
    -1.30
    
    -1.30
    POSITIVE LOGITS
    still
    1.45
     OGSÅ
    1.38
     aún
    1.38
    iress
    1.36
     alcune
    1.35
    邢
    1.34
     vengono
    1.33
     debería
    1.29
     his
    1.28
     иногда
    1.27
    Activations Density 0.010%

    No Known Activations