© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 19801
    Prev
    Next
    INDEX
    Explanations

    built-in features

    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
    à¹Ĥà¸Ķยสาร
    -0.07
    amarin
    -0.07
    resta
    -0.07
    fleet
    -0.07
     Äijẳng
    -0.06
     dél
    -0.06
    едÑĥ
    -0.06
    filmer
    -0.06
    (JFrame
    -0.06
    ASIL
    -0.06
    POSITIVE LOGITS
     re
    0.05
     Re
    0.05
    çļĦ综åIJĪ
    0.05
     ped
    0.04
    -re
    0.04
     I
    0.04
     J
    0.04
     Gram
    0.04
     E
    0.04
    ×ķ
    0.04
    Activations Density 0.008%

    No Known Activations