© 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. 40-GEMMASCOPE-2-RES-262K
    7. 1680
    Prev
    Next
    INDEX
    Explanations

    E-commerce identifiers

    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_40_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
    重大
    0.33
    న్నో
    0.30
    ិច
    0.30
     treble
    0.30
     ಅನೇಕ
    0.29
    ्रीट
    0.29
    킥
    0.29
     marginalised
    0.28
     pilota
    0.28
     SLOT
    0.28
    POSITIVE LOGITS
    D
    0.33
    Emb
    0.31
    fort
    0.30
    com
    0.29
    f
    0.29
    b
    0.29
    News
    0.29
    ---------------
    0.28
    V
    0.28
    be
    0.28
    Activations Density 0.006%

    No Known Activations