© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B-IT
    5. 10-GEMMASCOPE-2-TRANSCODER-262K
    7. 51797
    Prev
    Next
    INDEX
    Explanations

    clean lines, curves, quality, perfection

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-it/transcoder_all/layer_10_width_262k_l0_small_affine
    Prompts (Dashboard)
    238,145 prompts, 512 tokens each
    Dataset (Dashboard)
    lmsys + oasst1
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     جوړونکی
    0.75
    浧
    0.71
    <unused471>
    0.67
     څرنګوالی
    0.66
    <unused505>
    0.64
    <unused984>
    0.63
    ensureEqual
    0.63
    <unused2043>
    0.63
    +](=
    0.62
     İlç
    0.62
    POSITIVE LOGITS
     we
    0.82
    We
    0.73
    In
    0.73
    ää
    0.69
    we
    0.68
    The
    0.67
    You
    0.64
     was
    0.64
    When
    0.64
    ным
    0.64
    Activations Density 0.089%

    No Known Activations