© 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. 89721
    Prev
    Next
    INDEX
    Explanations

    modelmodel. model

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

    The neuron fires strongly on programming‐style tokens—punctuation (like “.”, “:”, “{”) and capitalized code identifiers (e.g. TableView, Delegate, Model)—i.e. it identifies code snippets.

    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
     férfi
    -0.70
     bőr
    -0.68
     Einzel
    -0.67
    ロナ
    -0.66
     intégr
    -0.66
    ട
    -0.66
    жат
    -0.66
    Einzelnachweise
    -0.65
     världen
    -0.65
    aş
    -0.64
    POSITIVE LOGITS
     model
    1.18
     MODEL
    1.12
     модель
    1.09
     모델
    1.09
    model
    1.01
    MODEL
    0.99
    setModel
    0.94
    UserRole
    0.94
     MODELS
    0.93
     models
    0.92
    Activations Density 0.006%

    No Known Activations