© 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. 12872
    Prev
    Next
    INDEX
    Explanations

    "No." followed by numbers

    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
    -2.50
     all
    -2.38
     other
    -2.19
     because
    -2.16
     both
    -2.14
     these
    -2.08
     new
    -2.02
     with
    -2.02
     should
    -2.02
     mostly
    -1.98
    POSITIVE LOGITS
     régal
    1.81
     mélang
    1.80
     inond
    1.75
     éton
    1.74
    を作って
    1.74
    だけではなく
    1.73
     verdo
    1.73
    ようになって
    1.71
    の位置
    1.63
     appliquer
    1.61
    Activations Density 0.046%

    No Known Activations