© 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. 120139
    Prev
    Next
    INDEX
    Explanations

    phrase prefixes

    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
     такие
    -0.07
     those
    -0.07
     two
    -0.07
     precedente
    -0.07
     including
    -0.07
     সংগ্রামী
    -0.06
     heavier
    -0.06
     Always
    -0.06
     nicer
    -0.06
     modern
    -0.06
    POSITIVE LOGITS
    ,
    0.07
    要求
    0.07
    就是
    0.07
     então
    0.06
    assie
    0.06
     entonces
    0.06
     tổ
    0.06
    但
    0.06
     then
    0.06
     потребо
    0.05
    Activations Density 0.001%

    No Known Activations