© 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-IT
    5. 31-GEMMASCOPE-2-RES-262K
    7. 78656
    Prev
    Next
    INDEX
    Explanations

    repetitive text, loops, questions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/resid_post/layer_31_width_262k_l0_medium
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ੀ
    0.49
    聲
    0.47
     ecosystem
    0.44
    !:
    0.44
    音
    0.44
    UNA
    0.43
    感謝
    0.42
     атмосфер
    0.42
     endeavor
    0.41
    !?
    0.41
    POSITIVE LOGITS
     %><%=
    0.46
    begin
    0.46
     marbles
    0.45
    给他
    0.45
     રીતે
    0.43
     permiten
    0.43
    addAlignment
    0.43
    jasama
    0.43
     Penang
    0.43
     seguire
    0.42
    Activations Density 0.003%

    No Known Activations