© 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. 22-GEMMASCOPE-RES-131K
    7. 26531
    Prev
    Next
    INDEX
    Explanations

    roll eyes

    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_22/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
     provide
    -1.85
     if
    -1.83
     take
    -1.75
     give
    -1.75
     of
    -1.73
     Making
    -1.73
     for
    -1.72
     Using
    -1.66
     comes
    -1.66
     Before
    -1.64
    POSITIVE LOGITS
    </i>
    2.14
    </em>
    2.05
    wonderful
    1.66
     haupts
    1.62
    
    1.57
    臜
    1.57
    鱨
    1.56
    籺
    1.56
    耵
    1.55
     magnificent
    1.55
    Activations Density 0.000%

    No Known Activations