© 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. 128004
    Prev
    Next
    INDEX
    Explanations

    documentation headers and code identifiers

    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
    Negative Logits
     ('
    -0.06
     that
    -0.06
    !
    -0.06
     McM
    -0.06
     they
    -0.06
     '
    -0.06
    ,(
    -0.06
    再
    -0.06
    ('
    -0.06
    '-
    -0.05
    POSITIVE LOGITS
     knowledgeable
    0.08
     könnte
    0.07
     müz
    0.07
     longstanding
    0.07
     không
    0.07
     könnten
    0.07
     لاعب
    0.07
    不僅
    0.07
     could
    0.07
     menjelaskan
    0.07
    Activations Density 0.005%

    No Known Activations