© 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. 116705
    Prev
    Next
    INDEX
    Explanations

    for loop setup

    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
     когда
    -0.07
     कभी
    -0.07
     lorsqu
    -0.07
    一部分
    -0.07
     quando
    -0.07
    时不
    -0.07
    需要的
    -0.07
    部分的
    -0.06
    부분
    -0.06
     cuando
    -0.06
    POSITIVE LOGITS
    WITH
    0.07
    Probit
    0.06
     @"
    0.06
    だと
    0.06
    цией
    0.06
    発表
    0.05
    與
    0.05
     típico
    0.05
     নিষিদ্ধ
    0.05
     usual
    0.05
    Activations Density 0.006%

    No Known Activations