© 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. 10-GEMMASCOPE-RES-131K
    7. 75388
    Prev
    Next
    INDEX
    Explanations

    system, levels, service, agency, file

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

    This neuron fires most strongly on multiword technical or domain‐specific compound terms (e.g. “software development,” “streaming service,” “testosterone levels,” “magnetic field,” etc.).

    oai_token-act-pair · o4-miniTriggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    ),
    -1.52
     )
    -1.50
    ?
    -1.45
    ).
    -1.43
    隄
    -1.34
     eased
    -1.33
    
    -1.32
     ).
    -1.30
     new
    -1.30
    (!__
    -1.25
    POSITIVE LOGITS
     fontos
    1.42
    zeichneten
    1.37
     drinken
    1.30
     КА
    1.29
     tärke
    1.29
     there
    1.27
     functie
    1.23
     waarschijnlijk
    1.23
    matian
    1.22
     combinatie
    1.21
    Activations Density 0.587%

    No Known Activations