© 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. 119681
    Prev
    Next
    INDEX
    Explanations

    certain words followed by 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
    ὥ
    -0.07
     cực
    -0.07
    unicate
    -0.06
    abspath
    -0.06
    花费
    -0.06
    蔫
    -0.06
    <0x97>
    -0.06
     از
    -0.06
     বাদশা
    -0.06
     একদম
    -0.05
    POSITIVE LOGITS
     её
    0.07
     an
    0.06
     ее
    0.06
     heres
    0.06
     nelle
    0.06
     ch
    0.06
     গড়ে
    0.06
     new
    0.06
    Ancest
    0.06
    Hep
    0.05
    Activations Density 0.003%

    No Known Activations