© 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. 21178
    Prev
    Next
    INDEX
    Explanations

    slight variation or difference

    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
    åŃIJä¾Ľ
    -0.05
    samp
    -0.05
     disruptive
    -0.05
    jang
    -0.05
    à¸Ńà¸ģ
    -0.05
    -ab
    -0.05
     aug
    -0.05
    acular
    -0.05
    upa
    -0.05
    奢
    -0.05
    POSITIVE LOGITS
     slightly
    0.10
    çķ¥å¾®
    0.10
     slight
    0.09
    åģıå·®
    0.08
    çķ¥
    0.08
    ç¨įå¾®
    0.08
    çķ¥æľī
    0.08
     subtly
    0.08
    ç¨įç¨į
    0.07
    å¾®å¦Ļ
    0.07
    Activations Density 0.189%

    No Known Activations