© 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. 128568
    Prev
    Next
    INDEX
    Explanations

    programming class definitions

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

    This neuron detects code namespace or module path tokens—e.g. backslashes in PHP namespaces and framework or library names like “React,” “yii,” or “Magento” in import/extend statements.

    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
     then
    -2.45
     is
    -1.87
     will
    -1.52
    了不少
    -1.42
     what
    -1.41
     some
    -1.40
     as
    -1.39
     would
    -1.39
    
    -1.36
     by
    -1.34
    POSITIVE LOGITS
     naszych
    1.72
     слегка
    1.67
     renk
    1.66
    ㍉
    1.52
     döner
    1.52
     nasze
    1.48
     parlement
    1.48
    蹀
    1.46
     taal
    1.44
    읗
    1.44
    Activations Density 0.023%

    No Known Activations