© 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. 75220
    Prev
    Next
    INDEX
    Explanations

    So for example if

    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
    降临
    -1.09
    躊
    -1.02
    ídas
    -1.02
     seldom
    -0.99
     краю
    -0.95
    Facile
    -0.94
    bagai
    -0.93
     Seems
    -0.93
    早めに
    -0.93
    %">
    -0.91
    POSITIVE LOGITS
     bijvoorbeeld
    1.70
     if
    1.47
    例えば
    1.41
     например
    1.39
     مث
    1.32
    たとえば
    1.26
     esimerkiksi
    1.24
     beispielsweise
    1.16
     example
    1.09
     once
    1.07
    Activations Density 0.053%

    No Known Activations