© 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. 20877
    Prev
    Next
    INDEX
    Explanations

    1

    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
    çĹħæĥħåĪĨæŀIJ
    -0.05
     radical
    -0.05
    èĢħ为
    -0.05
     luk
    -0.05
    æĪª
    -0.05
    æij
    -0.05
    æĥħå½¢ä¹ĭä¸ĢçļĦ
    -0.05
     exact
    -0.05
    积
    -0.05
    è¼
    -0.05
    POSITIVE LOGITS
    æĹłæģĻ
    0.06
    áĭ«
    0.06
    ouv
    0.06
    azed
    0.06
     ÄijÃłi
    0.06
    ijo
    0.06
    ÐĵлавнÑĭе
    0.06
    ounc
    0.06
    æıIJä¾ĽæĽ´å¥½çļĦ
    0.06
    ingly
    0.06
    Activations Density 0.018%

    No Known Activations