© 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. 34-GEMMASCOPE-RES-131K
    7. 33643
    Prev
    Next
    INDEX
    Explanations

    convergence and learning outcomes

    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
     אוי
    -0.93
    isters
    -0.91
     faſt
    -0.89
     bourg
    -0.89
     linke
    -0.87
     suple
    -0.87
    ándo
    -0.85
     departement
    -0.85
     minimalista
    -0.85
     inacces
    -0.85
    POSITIVE LOGITS
     large
    1.33
     convergence
    1.27
     small
    1.23
     Convergence
    1.16
     converge
    1.13
     regions
    1.13
    Convergence
    1.12
     badly
    1.11
     near
    1.08
     hardly
    1.07
    Activations Density 0.059%

    No Known Activations