© 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. 82352
    Prev
    Next
    INDEX
    Explanations

    interface, internationalization, programming

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

    The neuron fires specifically on standalone numeric tokens—sequences of digits—wherever they appear in the text.

    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
     excru
    -0.73
     menjalankan
    -0.73
    軼
    -0.71
     disintegration
    -0.70
     Iggy
    -0.69
    entric
    -0.69
     Spouse
    -0.68
     Luv
    -0.68
    Vesti
    -0.68
    avascript
    -0.68
    POSITIVE LOGITS
     interface
    6.72
    interface
    5.81
     interfaces
    5.41
     Interface
    5.09
    Interface
    4.88
    interfaces
    4.38
    接口
    4.38
     Interfaces
    4.28
     интерфей
    4.25
     INTERFACE
    4.19
    Activations Density 0.215%

    No Known Activations