© 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. 127980
    Prev
    Next
    INDEX
    Explanations

    concepts followed by specifics

    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
     if
    -2.27
     who
    -2.02
     because
    -2.00
     as
    -1.75
     now
    -1.59
     not
    -1.58
     so
    -1.57
     to
    -1.52
     which
    -1.52
     or
    -1.50
    POSITIVE LOGITS
    ulence
    1.47
     ४
    1.45
     célèbre
    1.45
    最初は
    1.42
     orgánica
    1.40
    沨
    1.40
    ༑
    1.39
     самых
    1.38
    色んな
    1.38
    栝
    1.38
    Activations Density 0.706%

    No Known Activations