© 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. 17253
    Prev
    Next
    INDEX
    Explanations

    sentence start

    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
     ever
    -0.06
    -ever
    -0.06
    为ä¸Ń
    -0.06
    å¼Ģ设æľī
    -0.06
     nul
    -0.05
    اتا
    -0.05
     frustr
    -0.05
    ocal
    -0.05
    è¯ĿæĿ¥
    -0.05
     borne
    -0.05
    POSITIVE LOGITS
    trieb
    0.07
    inned
    0.06
    umos
    0.05
    imme
    0.05
    retty
    0.05
    åħģ
    0.05
    æ°¸åĴĮ
    0.05
     <![
    0.05
    åŃIJåĴĮ
    0.05
    òn
    0.05
    Activations Density 0.069%

    No Known Activations