© 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. 10-GEMMASCOPE-RES-131K
    7. 19620
    Prev
    Next
    INDEX
    Explanations

    5, n, 3, p, Y, low, alpha, dot, non

    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
     not
    -3.20
     for
    -3.19
     with
    -3.17
     which
    -3.16
     from
    -3.16
     and
    -3.13
     what
    -3.06
     just
    -2.95
     in
    -2.83
     who
    -2.83
    POSITIVE LOGITS
     refroid
    2.69
     rafraî
    2.61
     verhind
    2.61
     problèmes
    2.58
     myndig
    2.52
     corriger
    2.50
     soucis
    2.48
     niedersachsen
    2.44
     véhic
    2.42
     régal
    2.42
    Activations Density 0.280%

    No Known Activations