© 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. 67813
    Prev
    Next
    INDEX
    Explanations

    established concepts and practices

    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
     REALLY
    -1.98
    婠
    -1.93
     nieuwe
    -1.85
     strikingly
    -1.84
     marvelous
    -1.84
    崤
    -1.81
    祧
    -1.81
    푘
    -1.80
    菒
    -1.80
    驁
    -1.80
    POSITIVE LOGITS
    During
    1.98
     blatantly
    1.83
    ?”
    1.83
    /*
    1.80
    Even
    1.79
     навіть
    1.77
     unrelenting
    1.74
    瀋
    1.73
    .”
    1.72
     plethora
    1.68
    Activations Density 0.019%

    No Known Activations