© 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. 16368
    Prev
    Next
    INDEX
    Explanations

    programming terms

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

    The neuron activates on code tokens naming style, layout, or configuration properties (e.g. CSS/JS properties like “transition,” “opacity,” “Auto Layout,” “transform,” “regX,” etc.).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    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
     biografias
    -0.88
    verten
    -0.85
    
    -0.83
    footnote
    -0.83
     Allociné
    -0.79
    */].
    -0.79
    リアル
    -0.79
     innehåller
    -0.78
     technologically
    -0.78
     Heilige
    -0.78
    POSITIVE LOGITS
     its
    1.05
    ulum
    0.82
    oarece
    0.81
     الكبير
    0.79
     macro
    0.77
    ECE
    0.75
    intérêt
    0.75
    個月
    0.75
     of
    0.74
     stabilité
    0.73
    Activations Density 0.140%

    No Known Activations