© 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. 21991
    Prev
    Next
    INDEX
    Explanations

    **System/Code Keywords**

    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
    gara
    -0.07
    ogie
    -0.06
    uito
    -0.06
    表达æĸ¹å¼ı
    -0.06
    ách
    -0.06
    odr
    -0.06
     concer
    -0.06
    åĴ¨
    -0.06
    ambar
    -0.06
    cker
    -0.06
    POSITIVE LOGITS
     sut
    0.05
    UpperCase
    0.05
    per
    0.05
    egis
    0.05
    start
    0.05
     íıī
    0.05
    æ¯ı件
    0.04
     rumor
    0.04
    å¾Ģ
    0.04
     juste
    0.04
    Activations Density 0.005%

    No Known Activations