© 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. 47702
    Prev
    Next
    INDEX
    Explanations

    sequence phrase completion

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

    attributes and their impact

    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
    جس
    0.45
     petits
    0.45
     इस्लाम
    0.45
     slapping
    0.44
    広
    0.43
     conspir
    0.43
     سبيل
    0.43
     Бүген
    0.43
     cardinal
    0.43
    を残
    0.43
    POSITIVE LOGITS
    h
    0.59
    Factory
    0.54
     Factory
    0.50
    FACTORY
    0.50
    v
    0.49
    dh
    0.49
    ut
    0.48
    ㅎ
    0.48
     onsite
    0.48
    Remote
    0.47
    Activations Density 0.000%

    No Known Activations