© 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
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128468
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on metadata‐declaration keywords and markers—e.g. the use or definition of annotations/attributes (like “annotations”, “@”, “interface” in annotation contexts).

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

    annotation attribute reflection

    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
    нцип
    -1.02
    および
    -0.94
    病情
    -0.93
    งิน
    -0.92
    ข้อมูล
    -0.92
    crypt
    -0.88
     Grèce
    -0.88
    canf
    -0.88
    changers
    -0.87
    Bat
    -0.86
    POSITIVE LOGITS
     ubezpiec
    0.96
    서울
    0.96
    formulario
    0.96
    actrice
    0.93
     such
    0.93
    Choisissez
    0.92
     зае
    0.92
    outil
    0.91
     foram
    0.91
    ࣪
    0.91
    Activations Density 0.004%

    No Known Activations