© 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. 120929
    Prev
    Next
    INDEX
    Explanations

    code syntax elements like (); or " )

    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
     以及
    -0.07
    获取
    -0.07
     আনুশকা
    -0.07
    因為
    -0.07
    涉及
    -0.06
     गईं
    -0.06
     বৎ
    -0.06
    因为
    -0.06
     totiž
    -0.06
    म्मू
    -0.06
    POSITIVE LOGITS
     one
    0.08
     or
    0.06
     should
    0.06
    だったので
    0.06
     was
    0.06
     data
    0.06
     has
    0.06
     expos
    0.06
    one
    0.05
     evidence
    0.05
    Activations Density 0.007%

    No Known Activations