© 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. 67858
    Prev
    Next
    INDEX
    Explanations

    articles before nouns

    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
     would
    -1.81
     when
    -1.73
     cuando
    -1.72
     такого
    -1.48
     Поэтому
    -1.46
     very
    -1.38
     as
    -1.34
    ,
    -1.31
     However
    -1.31
     actually
    -1.31
    POSITIVE LOGITS
    !
    1.51
     tomadas
    1.45
     saludables
    1.37
     seguras
    1.37
     domés
    1.32
     pady
    1.30
     atento
    1.30
     первый
    1.29
     establecidas
    1.26
     marcadas
    1.25
    Activations Density 0.045%

    No Known Activations