© 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. 73677
    Prev
    Next
    INDEX
    Explanations

    Vietnamese, Cambodia, Hanoi

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

    The neuron activates on polite help‐requests—phrases where the writer asks the reader to “please,” “instruct me,” “show me,” “help me,” 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_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
     Tiểu
    -1.19
     Thần
    -1.11
     Tử
    -1.08
     Địa
    -1.02
     Dân
    -0.99
     Trọng
    -0.93
     Đình
    -0.89
     Empfindung
    -0.87
     Thị
    -0.85
     Chủ
    -0.85
    POSITIVE LOGITS
     Kho
    0.91
     Kh
    0.91
     Khmer
    0.91
     ensalada
    0.91
     Vietnamese
    0.90
     Kha
    0.89
     bá
    0.88
     Vietnam
    0.87
     kä
    0.87
     pelea
    0.86
    Activations Density 0.057%

    No Known Activations