© 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. 117457
    Prev
    Next
    INDEX
    Explanations

    hexadecimal strings and letters

    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
    都
    -0.06
     arbets
    -0.06
     Alli
    -0.05
     all
    -0.05
    大部分
    -0.05
    ޫ
    -0.05
    প্রায়
    -0.05
    すべての
    -0.05
     सगळे
    -0.05
    ahkan
    -0.05
    POSITIVE LOGITS
     with
    0.07
     including
    0.07
     modulo
    0.07
     dianggap
    0.07
     computationally
    0.07
     pierws
    0.06
     speaking
    0.06
     используя
    0.06
     wówczas
    0.06
     spoken
    0.06
    Activations Density 0.003%

    No Known Activations