© 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. 130860
    Prev
    Next
    INDEX
    Explanations

    The neuron detects mentions of specific languages—especially in the context of listing or praising someone’s language skills.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    english proficiency and learning

    np_acts-logits-general · gemini-2.5-flash-lite
    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
     strane
    -1.26
    Hyderabad
    -1.03
    пку
    -1.02
     already
    -1.02
    病例
    -1.01
     koruyucu
    -1.00
     parlak
    -1.00
     lemn
    -0.99
     tenda
    -0.99
     dekoratif
    -0.99
    POSITIVE LOGITS
     communication
    1.54
     communicative
    1.19
     communicating
    1.14
    communication
    1.10
    |};
    1.08
    thebibliography
    1.07
     communic
    1.05
     communicate
    1.04
     lorsqu
    0.99
     conversation
    0.99
    Activations Density 0.010%

    No Known Activations