© 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. 39495
    Prev
    Next
    INDEX
    Explanations

    existing X entities

    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
     présenter
    -2.48
    秾
    -2.45
    蠖
    -2.31
     résister
    -2.22
     konkrete
    -2.20
    行く
    -2.17
     etik
    -2.16
     sekunder
    -2.16
    2
    -2.16
        
    -2.14
    POSITIVE LOGITS
     thoughtful
    2.88
    閙
    2.78
    後記
    2.66
    尛
    2.56
     administrar
    2.48
     Particularly
    2.47
    墬
    2.45
    与此同时
    2.44
    曏
    2.42
    的吗
    2.41
    Activations Density 0.008%

    No Known Activations