© 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. 120212
    Prev
    Next
    INDEX
    Explanations

    non-English words

    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
     sooner
    -0.06
     wholesale
    -0.06
    dui
    -0.06
     intrusive
    -0.06
    Meanwhile
    -0.06
    𝒂
    -0.06
     Profes
    -0.06
    优先
    -0.06
     निवेदन
    -0.06
    मेट्री
    -0.06
    POSITIVE LOGITS
     perchè
    0.06
     tăng
    0.06
     inac
    0.06
     соответствии
    0.06
     maravilh
    0.06
     exatamente
    0.06
     защото
    0.05
     สามารถ
    0.05
     porque
    0.05
     omdat
    0.05
    Activations Density 0.001%

    No Known Activations