© 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. 21408
    Prev
    Next
    INDEX
    Explanations

    became, though, amid, learned, off, had

    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
    terapia
    -0.06
    kki
    -0.06
    æĢªä¸įå¾Ĺ
    -0.06
    åIJĪæ³ķçļĦ
    -0.06
    /shop
    -0.06
    صÙī
    -0.06
    ergus
    -0.06
     بدÙĦاÙĭ
    -0.06
    agged
    -0.06
    åIJĪçIJĨåľ°
    -0.06
    POSITIVE LOGITS
     Wolver
    0.05
    å®ŀè·µ
    0.05
    大éĥ¨åĪĨ
    0.05
    em
    0.05
    æĿ¿
    0.05
    atta
    0.05
    表
    0.05
     Pase
    0.05
    浪
    0.05
    åĮĸæĪIJ
    0.05
    Activations Density 0.010%

    No Known Activations