© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-4-31B
    5. 30-RES-MATRYOSHKA-131K
    7. 118909
    Prev
    Next
    INDEX
    Explanations

    then, after, and then

    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
     had
    -0.07
    Having
    -0.07
     having
    -0.06
    !!!!!
    -0.06
    Specific
    -0.05
     ≥
    -0.05
    !
    -0.05
     BC
    -0.05
     predominantly
    -0.05
     possibili
    -0.05
    POSITIVE LOGITS
     kemudian
    0.08
     তারপর
    0.07
     then
    0.07
    然后再
    0.07
     ل
    0.07
    efore
    0.07
    然後
    0.07
     затем
    0.07
     говорить
    0.07
    然后
    0.06
    Activations Density 0.013%

    No Known Activations