© 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. 17453
    Prev
    Next
    INDEX
    Explanations

    temporal transition or conclusion

    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
    upe
    -0.06
    åѦèĢĮ
    -0.05
    ont
    -0.05
    kees
    -0.05
     never
    -0.05
     Trem
    -0.05
    ä¸įåıĺ
    -0.05
    quo
    -0.05
    never
    -0.05
    gue
    -0.05
    POSITIVE LOGITS
    edan
    0.07
    lah
    0.06
    ombre
    0.06
    аÑĤелÑĮно
    0.06
    ä½ľèĢħåľ¨
    0.06
    redi
    0.06
    ä½İ声
    0.06
    cesi
    0.05
    ensively
    0.05
    ovin
    0.05
    Activations Density 0.011%

    No Known Activations