© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 116157
    Prev
    Next
    INDEX
    Explanations

    light beam

    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
     jedes
    -0.06
     મારા
    -0.06
     конструкция
    -0.06
     제가
    -0.06
    構造
    -0.06
    上的
    -0.06
     surprisingly
    -0.06
    engineering
    -0.06
    France
    -0.06
     私
    -0.06
    POSITIVE LOGITS
     Gazette
    0.06
     would
    0.06
     during
    0.06
    ഴും
    0.06
     først
    0.06
     trei
    0.05
     деву
    0.05
    Haz
    0.05
     ሶ
    0.05
    lasti
    0.05
    Activations Density 0.005%

    No Known Activations