© 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. 119556
    Prev
    Next
    INDEX
    Explanations

    start of sentence after punctuation

    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
     følgende
    -0.07
     সবকিছু
    -0.07
     несколько
    -0.07
     그냥
    -0.07
     просто
    -0.07
     немного
    -0.07
     یک
    -0.07
     كلها
    -0.07
     eine
    -0.07
     wszystko
    -0.07
    POSITIVE LOGITS
     the
    0.07
    ...!
    0.06
     ষে
    0.06
     Based
    0.06
     seats
    0.06
     based
    0.06
    建成
    0.05
    роят
    0.05
    based
    0.05
    take
    0.05
    Activations Density 0.009%

    No Known Activations