© 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
    2. Gemma-2-27B
    3. 10-GEMMASCOPE-RES-131K
    4. 7274
    Prev
    Next
    INDEX
    Explanations

    sprawling landscape/entity

    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
    畯
    -2.28
     conclude
    -2.23
    which
    -2.20
    轳
    -2.08
     există
    -2.08
     gebruik
    -2.05
     prevenir
    -2.05
     and
    -2.05
     folosi
    -2.05
    踭
    -2.03
    POSITIVE LOGITS
    橚
    2.70
        
    2.36
     ditemui
    2.34
    be
    2.25
     immaculate
    2.23
     rufen
    2.23
     for
    2.19
     inquisitive
    2.14
    i
    2.08
    .”
    2.05
    Activations Density 0.006%

    No Known Activations