© 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. 66777
    Prev
    Next
    INDEX
    Explanations

    poor followed by nouns like kid, kitty, thing, dog

    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
    -1.26
    だめ
    -1.00
     skool
    -0.98
     устраи
    -0.95
    不思議な
    -0.95
     wichtiger
    -0.93
     partnership
    -0.92
    cL
    -0.91
     löytyy
    -0.91
    Snail
    -0.91
    POSITIVE LOGITS
    因为
    1.30
    这些
    1.08
    在
    1.06
    虽然
    1.04
    因為
    1.03
    随着
    1.03
    Poor
    1.03
    Первый
    1.03
    ที่
    1.02
    如果
    1.00
    Activations Density 0.021%

    No Known Activations