© 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. 10-GEMMASCOPE-RES-131K
    7. 104262
    Prev
    Next
    INDEX
    Explanations

    expected behavior

    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
     brinda
    -1.40
     molto
    -1.34
     interak
    -1.34
    eleven
    -1.31
     demostr
    -1.27
     absolut
    -1.27
    
    -1.27
     comprome
    -1.27
    Weaver
    -1.26
    がない
    -1.25
    POSITIVE LOGITS
     when
    2.42
     with
    2.22
     or
    1.91
     during
    1.85
     there
    1.83
     When
    1.76
    when
    1.74
     Probably
    1.66
     Perhaps
    1.66
     Despite
    1.66
    Activations Density 0.660%

    No Known Activations