© 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. 14253
    Prev
    Next
    INDEX
    Explanations

    programming code and questions

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

    The neuron fires on code “using” or import statements that reference UI-related namespaces (e.g. System.Windows.Input, Xaml.Input, Data, Navigation, Shapes).

    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
     from
    -1.21
     then
    -1.02
    On
    -1.02
    Each
    -1.00
    with
    -0.98
     earlier
    -0.96
     on
    -0.93
    ϕ
    -0.91
     each
    -0.91
    Then
    -0.90
    POSITIVE LOGITS
    Appellee
    1.00
     allerlei
    1.00
    你还
    0.98
    colori
    0.98
     impure
    0.96
     Nepali
    0.96
    强者
    0.95
    Nepal
    0.94
     dü
    0.93
    perhaps
    0.93
    Activations Density 0.001%

    No Known Activations