© 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. 22-GEMMASCOPE-RES-131K
    7. 54314
    Prev
    Next
    INDEX
    Explanations

    android code documentation

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

    This neuron fires strongly on punctuation tokens that mark code structure—particularly angle brackets (“<”) (as in XML/HTML or C# generics/comments) and dots (“.”) in qualified names.

    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
    produksi
    -1.28
     its
    -1.10
    tentu
    -1.02
    高齢
    -0.95
     regardless
    -0.95
     красивые
    -0.91
    бр
    -0.90
    iteľ
    -0.88
    empio
    -0.88
     even
    -0.87
    POSITIVE LOGITS
    riceps
    1.33
    LookAnd
    1.18
     volledige
    1.09
     natuurlijke
    1.04
    icoot
    1.01
     belangrijke
    1.00
     voedsel
    1.00
    holic
    1.00
    Endnotes
    1.00
     حركة
    1.00
    Activations Density 0.001%

    No Known Activations