© 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. 13007
    Prev
    Next
    INDEX
    Explanations

    suit, part, item, others, sal, redesign, Happiness, function

    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
     SPOILER
    -1.00
     verboten
    -0.90
     ingrat
    -0.88
    compared
    -0.88
     deus
    -0.87
    💢
    -0.87
    ようだ
    -0.86
     bahkan
    -0.86
     ymm
    -0.85
    fällt
    -0.85
    POSITIVE LOGITS
     simply
    1.56
     is
    1.51
     was
    1.50
     means
    1.37
     because
    1.37
     mentioned
    1.31
    是为了
    1.30
     refers
    1.25
     merely
    1.23
     I
    1.22
    Activations Density 0.043%

    No Known Activations