© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 67086
    Prev
    Next
    INDEX
    Explanations

    Tucker and Hugo names

    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
     are
    -2.52
    料
    -2.39
     diversos
    -2.33
     москов
    -2.27
    }$
    -2.27
     hermosos
    -2.25
     isoliert
    -2.23
    🧌
    -2.17
    深的
    -2.17
     marvelous
    -2.17
    POSITIVE LOGITS
    酲
    2.58
    When
    2.58
     When
    2.55
    Nearly
    2.44
    欵
    2.42
    CurtirCurtir
    2.39
    </td>
    2.36
    我能
    2.34
    0
    2.30
    Before
    2.22
    Activations Density 0.003%

    No Known Activations