© 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. 128960
    Prev
    Next
    INDEX
    Explanations

    experimental conditions and procedures

    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
     kills
    -0.07
     concetto
    -0.07
     believe
    -0.06
    OCO
    -0.06
     overtake
    -0.06
     progett
    -0.06
     यज्ञ
    -0.06
    BufferedReader
    -0.06
    lava
    -0.06
    Learning
    -0.06
    POSITIVE LOGITS
     hanya
    0.06
     също
    0.06
     فقط
    0.06
     بهذه
    0.06
     combinations
    0.06
    只能
    0.06
     wyłącznie
    0.06
    combinations
    0.05
    このような
    0.05
     procede
    0.05
    Activations Density 0.002%

    No Known Activations