© 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. 118764
    Prev
    Next
    INDEX
    Explanations

    Dani Pedrosa and Daniil Kvyat

    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
    </h3>
    -2.81
    菝
    -2.69
    ቷ
    -2.64
    蹀
    -2.64
    omg
    -2.44
    〬
    -2.44
     normalen
    -2.42
    ܆
    -2.39
     mirando
    -2.38
    ⎔
    -2.38
    POSITIVE LOGITS
     was
    3.59
    e
    3.28
    胧
    2.94
    a
    2.89
     场
    2.78
    .
    2.58
    窳
    2.56
     in
    2.53
     Some
    2.42
    ا
    2.42
    Activations Density 0.007%

    No Known Activations