© 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. 120629
    Prev
    Next
    INDEX
    Explanations

    names with appended initials or titles

    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
     للم
    -0.07
     enthusiasm
    -0.07
     gives
    -0.07
     says
    -0.06
     encourages
    -0.06
     الرسمي
    -0.06
     avoids
    -0.06
     gave
    -0.06
     followed
    -0.06
     doesn
    -0.06
    POSITIVE LOGITS
     "
    0.06
     fully
    0.06
    CC
    0.06
    इसी
    0.05
    AC
    0.05
    ที่มี
    0.05
     obstante
    0.05
     সালফেট
    0.05
     semuanya
    0.05
     thì
    0.05
    Activations Density 0.015%

    No Known Activations