© 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. 120931
    Prev
    Next
    INDEX
    Explanations

    code structure and common tokens

    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
     muchos
    -0.07
     presión
    -0.07
     muitos
    -0.07
     constraints
    -0.06
     पिछली
    -0.06
     العديد
    -0.06
     circonstances
    -0.06
     기존
    -0.06
     প্রাস
    -0.06
     viele
    -0.06
    POSITIVE LOGITS
    【
    0.06
     after
    0.06
    ,
    0.06
     Während
    0.06
     via
    0.06
     ናቸው
    0.06
     primary
    0.05
    Während
    0.05
     في
    0.05
     durante
    0.05
    Activations Density 0.009%

    No Known Activations