© 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. 116982
    Prev
    Next
    INDEX
    Explanations

    program paths

    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
     donde
    -0.08
    ).
    -0.08
     where
    -0.08
     whose
    -0.07
     cuya
    -0.07
     cuyos
    -0.07
    where
    -0.07
    whose
    -0.07
     dimana
    -0.06
     utilizing
    -0.06
    POSITIVE LOGITS
     contradicts
    0.06
     to
    0.06
    attup
    0.06
     إلى
    0.06
     Suzy
    0.06
     اصل
    0.06
     ਤਾਂ
    0.06
    zelf
    0.05
    checkBox
    0.05
     this
    0.05
    Activations Density 0.001%

    No Known Activations