© 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. 41-GEMMASCOPE-2-RES-262K
    7. 25380
    Prev
    Next
    INDEX
    Explanations

    SageMaker

    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_all/layer_41_width_262k_l0_big
    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
    ured
    0.40
     CLI
    0.40
     prefixes
    0.39
    istered
    0.39
    THERS
    0.39
    дации
    0.38
    CLI
    0.38
    毗
    0.38
    ileged
    0.36
     confound
    0.36
    POSITIVE LOGITS
    Sage
    0.64
    brush
    0.60
    Sag
    0.55
     Sage
    0.55
     sag
    0.52
     sage
    0.49
     Sag
    0.49
     SAG
    0.48
    sag
    0.47
     sagitt
    0.46
    Activations Density 0.001%

    No Known Activations