© 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. 130211
    Prev
    Next
    INDEX
    Explanations

    This neuron fires most strongly on code documentation markers—especially XML doc comment tags like `<summary>`/`</summary>`, triple‐slash comments (`///`), and attribute brackets (`[...]`) around class or method declarations.

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

    abstract class definitions

    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
    mitsubishi
    -0.84
     ferons
    -0.83
     flera
    -0.82
     situate
    -0.81
    ߞ
    -0.81
     Большая
    -0.79
     Такие
    -0.78
    értelmű
    -0.77
    subtle
    -0.77
     Теперь
    -0.77
    POSITIVE LOGITS
    class
    1.27
     class
    1.23
     an
    1.10
    ($__
    0.92
     classe
    0.92
     TType
    0.91
     度
    0.88
    ijão
    0.88
     _$
    0.88
     relay
    0.85
    Activations Density 0.023%

    No Known Activations