© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 129939
    Prev
    Next
    INDEX
    Explanations

    numerical context or identifiers

    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
     generalizes
    -0.06
     [
    -0.06
     если
    -0.06
     imparting
    -0.06
    নির্ব
    -0.06
     enhancing
    -0.06
     אם
    -0.05
     entirely
    -0.05
     enhance
    -0.05
     enhances
    -0.05
    POSITIVE LOGITS
     constitué
    0.07
    其中包括
    0.07
     مشتمل
    0.06
    例如
    0.06
    包含
    0.06
     例如
    0.06
    ニャ
    0.06
    嵌
    0.06
     مك
    0.06
     composto
    0.06
    Activations Density 0.002%

    No Known Activations