© 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. 120174
    Prev
    Next
    INDEX
    Explanations

    your first

    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
     produces
    -0.06
     deference
    -0.06
     or
    -0.06
    XM
    -0.06
     момент
    -0.06
     weighty
    -0.06
    item
    -0.06
     στιγμή
    -0.05
     then
    -0.05
    吘
    -0.05
    POSITIVE LOGITS
    已經
    0.08
     bereits
    0.07
     đã
    0.06
     еще
    0.06
     پہلے
    0.06
     pastoral
    0.06
     فراموش
    0.06
     masih
    0.06
     সাধারন
    0.06
    Still
    0.06
    Activations Density 0.001%

    No Known Activations