Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 8116
    Prev
    Next
    INDEX
    Explanations

    using directive

    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_34/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
     from
    -1.02
    employeeId
    -0.91
    PageModule
    -0.90
     then
    -0.89
    zipCode
    -0.85
    容易
    -0.85
    ichthys
    -0.84
    交付
    -0.81
    bandcamp
    -0.79
    databinding
    -0.79
    POSITIVE LOGITS
    using
    1.98
     using
    1.70
     use
    1.35
     USING
    1.34
     использовании
    1.33
     utilisant
    1.31
     usando
    1.25
     استفاده
    1.23
     menggunakan
    1.20
     sử
    1.17
    Activations Density 0.015%

    No Known Activations