© 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. 119649
    Prev
    Next
    INDEX
    Explanations

    code blocks and structure

    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
     لت
    -0.07
     생각을
    -0.07
     அறி
    -0.07
     المقص
    -0.06
     оба
    -0.06
     salah
    -0.06
     ули
    -0.06
    ண்
    -0.06
     на
    -0.06
     simulated
    -0.06
    POSITIVE LOGITS
     repeatedly
    0.06
     asking
    0.06
    Writable
    0.06
    通常
    0.06
     rescheduling
    0.06
    artwork
    0.06
     हाइट
    0.05
    ⁢
    0.05
     acknowledging
    0.05
     seeking
    0.05
    Activations Density 0.020%

    No Known Activations