© 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. 13524
    Prev
    Next
    INDEX
    Explanations

    always, maybe, could, would, just

    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
     what
    -2.02
     other
    -1.54
     different
    -1.49
     various
    -1.40
     numerous
    -1.38
     these
    -1.37
     stated
    -1.36
     Кроме
    -1.35
     this
    -1.34
     verschil
    -1.34
    POSITIVE LOGITS
     "'");
    1.44
    ১
    1.30
    </em>
    1.29
     ล
    1.29
     ง
    1.27
     че
    1.26
     "")
    1.26
    ра
    1.26
    SerializeField
    1.25
    ори
    1.25
    Activations Density 0.006%

    No Known Activations