© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 184231
    Prev
    Next
    INDEX
    Explanations

    changes improve

    np_acts-logits-general · gemini-2.5-flash-lite

    internalizing patterns

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     게
    0.53
     tinted
    0.53
     strainer
    0.47
     विजे
    0.47
     보여
    0.47
     う
    0.47
     fires
    0.46
     moisturizer
    0.46
     fiddle
    0.46
     ملی
    0.46
    POSITIVE LOGITS
    -
    0.52
    Supports
    0.46
    NAMEN
    0.44
    路
    0.44
    最大的
    0.44
    支持
    0.44
    身高
    0.43
    约
    0.43
    的支持
    0.43
    善
    0.43
    Activations Density 0.001%

    No Known Activations