© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Gemma-3-4B-IT
    3. 5-GEMMASCOPE-2-TRANSCODER-262K
    4. 3245
    Prev
    Next
    INDEX
    Explanations

    produces 12 coumarins

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-4b-it/transcoder_all/layer_5_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
    id
    2.09
    ுக
    2.04
    ف
    2.01
    it
    2.01
    og
    1.99
     forbid
    1.98
     পড়িয়া
    1.95
    forbidden
    1.95
    ט
    1.87
    itake
    1.83
    POSITIVE LOGITS
     luas
    1.83
    ीकरण
    1.72
    punkt
    1.69
     Bxa
    1.69
    logram
    1.67
     BeautifulSoup
    1.64
     Poj
    1.64
    ted
    1.62
    mselves
    1.62
     того
    1.61
    Activations Density 0.000%

    No Known Activations