© 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. 54273
    Prev
    Next
    INDEX
    Explanations

    increasing quantities

    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
     differing
    -1.08
     entwickelte
    -1.02
    RAH
    -1.00
    excellents
    -0.98
     więks
    -0.98
     prosegu
    -0.98
     rác
    -0.95
     upgraded
    -0.95
     diminished
    -0.94
     biztos
    -0.94
    POSITIVE LOGITS
     by
    2.47
     to
    1.97
     beyond
    1.48
     enough
    1.45
     while
    1.37
     significantly
    1.33
     further
    1.33
     considerably
    1.32
     substantially
    1.18
     accordingly
    1.17
    Activations Density 0.235%

    No Known Activations