© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 7324
    Prev
    Next
    INDEX
    Explanations

    prefix followed by specified word(s)

    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_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
    ’
    -3.00
    ፨
    -2.80
    儇
    -2.72
     帅气
    -2.72
    ꈊ
    -2.70
    er
    -2.67
    ский
    -2.56
    になりました
    -2.47
    zyć
    -2.45
     يمكنك
    -2.45
    POSITIVE LOGITS
    买了
    2.64
    ми
    2.64
     According
    2.44
     they
    2.36
     If
    2.36
     originating
    2.34
    有沒有
    2.27
    有什麼
    2.22
     When
    2.20
     gewähr
    2.20
    Activations Density 0.312%

    No Known Activations