© 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. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 21318
    Prev
    Next
    INDEX
    Explanations

    causes of

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    decoderesearch/qwen-3.5-saes/qwen-3.5-4b
    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
    伺
    -0.05
    moth
    -0.05
     un
    -0.05
     non
    -0.05
    ä¾ĿéĻĦ
    -0.05
     fuer
    -0.05
     co
    -0.05
    tera
    -0.05
     similarly
    -0.04
    =rand
    -0.04
    POSITIVE LOGITS
     Noticed
    0.06
    Opened
    0.06
    à¹Ģà¸Ľà¸µà¸¢
    0.06
     ÙħÛĮÙĦ
    0.06
    าà¸Ĺ
    0.06
    yanto
    0.06
    lessly
    0.06
    ÐŀÐĴ
    0.06
    åIJı
    0.06
    arg
    0.06
    Activations Density 0.199%

    No Known Activations