© 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. 120003
    Prev
    Next
    INDEX
    Explanations

    version comparison

    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
     its
    -0.08
    它的
    -0.08
     namely
    -0.07
     iaitu
    -0.07
    有些
    -0.07
     notably
    -0.07
     yani
    -0.07
     LIABLE
    -0.07
    namely
    -0.06
    看似
    -0.06
    POSITIVE LOGITS
     donned
    0.06
     وس
    0.06
    responsibility
    0.06
     solves
    0.06
    expertise
    0.06
     decisions
    0.06
     இருக்கும்
    0.06
     решения
    0.06
     प्रॉफिट
    0.05
     takes
    0.05
    Activations Density 0.001%

    No Known Activations