© 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. 18091
    Prev
    Next
    INDEX
    Explanations

    `<li>` followed by `name`, `role`, `inline`, or `key`

    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
     partially
    -0.06
     repr
    -0.05
    Ñīен
    -0.05
    ×ij×ķר
    -0.05
    èĸĦçļĦ
    -0.05
    岸
    -0.05
    æĦı
    -0.05
    åĪ©çļĦ
    -0.05
    é£
    -0.05
    éĹª
    -0.05
    POSITIVE LOGITS
    зÑĢе
    0.06
    ist
    0.06
    ãģ«è¡Įãģ£ãģ¦ãģįãģ¾ãģĹãģŁ
    0.06
    Ñĩее
    0.06
    akter
    0.06
    è´¹çŃī
    0.06
    ducted
    0.06
    olds
    0.06
    ver
    0.06
    verk
    0.06
    Activations Density 0.002%

    No Known Activations