© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 50-GEMMASCOPE-2-TRANSCODER-262K
    7. 197569
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_50_width_262k_l0_small_affine
    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
     WALK
    0.43
     받을
    0.38
     অর্ধ
    0.38
    pathTo
    0.37
     இடை
    0.37
     సామ
    0.37
    chromo
    0.36
    坯
    0.36
     elastomer
    0.36
     Elliot
    0.36
    POSITIVE LOGITS
    тун
    0.39
    ‌ی
    0.38
    зан
    0.38
     perks
    0.38
    남
    0.37
    stances
    0.36
    egan
    0.35
    قول
    0.34
     ș
    0.34
     ی
    0.34
    Activations Density 0.000%

    No Known Activations