© 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. 128861
    Prev
    Next
    INDEX
    Explanations

    capitalized words indicating start of phrase

    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
     themed
    -0.06
     प्रति
    -0.05
     incarceration
    -0.05
     an
    -0.05
     a
    -0.05
     પ્રતિ
    -0.05
     స్థితి
    -0.05
    ppled
    -0.05
     stunning
    -0.05
    ঘরে
    -0.05
    POSITIVE LOGITS
     prilikom
    0.07
     প্রথম
    0.06
     hlavní
    0.06
    虽然
    0.06
     iako
    0.06
     fleks
    0.06
     অন্যরা
    0.06
    While
    0.06
     primeira
    0.06
     aerodynamics
    0.06
    Activations Density 0.001%

    No Known Activations