© 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-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 117607
    Prev
    Next
    INDEX
    Explanations

    end-of-sentence punctuation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/gemma-4-saes/gemma-4-31b
    Prompts (Dashboard)
    16,384 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     নিজেই
    -0.07
     you
    -0.07
     sebenarnya
    -0.07
     έχ
    -0.06
     přímo
    -0.06
     είναι
    -0.06
     চিন্তা
    -0.06
     even
    -0.06
     еще
    -0.06
     đấy
    -0.06
    POSITIVE LOGITS
     was
    0.07
     always
    0.06
     siempre
    0.06
    However
    0.06
     Window
    0.06
     ব্রহ্ম
    0.06
    一定要
    0.06
     ALWAYS
    0.06
     किल
    0.06
     demanded
    0.06
    Activations Density 0.011%

    No Known Activations