© 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. Qwen3.5-9B Base
    5. 15-QWENSCOPE-RES-64K
    7. 25342
    Prev
    Next
    INDEX
    Explanations

    contaminants and impurities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     informat
    -0.07
     longest
    -0.07
    iosa
    -0.07
    SSI
    -0.07
    _Selected
    -0.07
    oose
    -0.07
    çĭ¬åįł
    -0.07
    åį¡ç½Ĺ
    -0.07
    mente
    -0.07
    íıIJ
    -0.06
    POSITIVE LOGITS
    æĺ¯å¯¼èĩ´
    0.09
     unwitting
    0.09
     sources
    0.09
     presence
    0.08
     accidentally
    0.08
    ä¸Ģä¸Ŀ
    0.08
    æĿ¥æºIJ
    0.08
     contamination
    0.08
     unintention
    0.08
     inadvertently
    0.08
    Activations Density 0.037%

    No Known Activations