© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 22871
    Prev
    Next
    INDEX
    Explanations

    serial numbers or identifiers

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

    This neuron activates on C# source‐code tokens—especially PascalCase identifiers (like class, namespace, and method names) and their punctuation (dots, braces, semicolons).

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
     to
    -1.43
     in
    -1.40
    /*",
    -1.34
    cester
    -1.33
    体が
    -1.29
     zvlá
    -1.28
    quirrel
    -1.25
     Militär
    -1.23
    もありますが
    -1.23
    ucson
    -1.20
    POSITIVE LOGITS
     demselben
    1.45
    ませんでした
    1.37
     betreffenden
    1.31
    ｘ
    1.27
     entschieden
    1.27
     dagli
    1.27
     parlando
    1.27
    ൅
    1.26
     Bewußt
    1.26
     bedeutend
    1.25
    Activations Density 0.016%

    No Known Activations