© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19784
    Prev
    Next
    INDEX
    Explanations

    say nothing, research paper, full version

    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
    æĽ¿
    -0.06
    anner
    -0.05
    alle
    -0.05
    еÑģа
    -0.05
    çŀ¬æģ¯
    -0.05
    ä¸į该
    -0.05
    atti
    -0.05
    ÙĨاÙĨ
    -0.05
    aka
    -0.05
    Ro
    -0.05
    POSITIVE LOGITS
     sout
    0.06
    går
    0.06
    engeb
    0.06
    æ°´æĺ¯
    0.06
     iny
    0.06
    ç»Ļå®ĥ
    0.06
    itam
    0.05
    à¸Ńาย
    0.05
     Macy
    0.05
    ศ
    0.05
    Activations Density 0.017%

    No Known Activations