© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 14183
    Prev
    Next
    INDEX
    Explanations

    technical conference proceedings

    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
     and
    -1.02
     because
    -1.02
     which
    -1.02
     to
    -0.98
     or
    -0.94
     of
    -0.94
     who
    -0.93
     too
    -0.92
     they
    -0.89
    we
    -0.88
    POSITIVE LOGITS
     honneur
    1.20
     vœ
    1.13
    鎳
    1.07
     aéri
    1.05
     medlems
    1.05
     tomber
    1.02
     kän
    1.01
    ᡝ
    1.00
    Sprache
    1.00
    ㌘
    1.00
    Activations Density 0.001%

    No Known Activations