© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 83605
    Prev
    Next
    INDEX
    Explanations

    code namespaces and libraries

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

    The neuron is detecting code tokens that are .NET-style identifiers—especially CamelCase namespace and class names in using/namespace declarations.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     after
    -0.96
    ITÉ
    -0.93
     المعلومات
    -0.92
     before
    -0.92
    ANNED
    -0.89
     and
    -0.89
    獰
    -0.89
    الثة
    -0.86
     without
    -0.85
    ZZA
    -0.85
    POSITIVE LOGITS
    ._
    1.18
     cependant
    1.16
    ")]
    1.15
    TagHelpers
    1.12
    AspNetCore
    1.09
    _
    1.09
    sharp
    1.06
    ___
    1.06
    WebControls
    1.06
     insbesondere
    1.05
    Activations Density 0.042%

    No Known Activations