© 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. 117517
    Prev
    Next
    INDEX
    Explanations

    conjunctions connecting diverse items

    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
     mainly
    -0.07
     principalmente
    -0.07
     Mainly
    -0.07
     you
    -0.06
     almost
    -0.06
     tối
    -0.06
     otherwise
    -0.06
     आमतौर
    -0.06
     essentiellement
    -0.06
     remains
    -0.06
    POSITIVE LOGITS
    而
    0.09
     while
    0.08
     而
    0.06
     mientras
    0.06
     당연
    0.06
     جبکہ
    0.06
     forman
    0.06
    而在
    0.06
    하면서
    0.06
    while
    0.06
    Activations Density 0.026%

    No Known Activations