© 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. 17481
    Prev
    Next
    INDEX
    Explanations

    after the completion or end

    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
    ^[
    -0.06
    ary
    -0.06
    iare
    -0.06
    mek
    -0.06
    åİĮæģ¶
    -0.06
    бло
    -0.05
    onio
    -0.05
    arel
    -0.05
    pto
    -0.05
     grouped
    -0.05
    POSITIVE LOGITS
    estin
    0.07
    oldt
    0.06
    ingan
    0.06
    ago
    0.06
     dalje
    0.06
    ä¸ĭ线
    0.06
    clang
    0.06
    对äºİä¸Ģ个
    0.06
    Beyond
    0.06
     priÄį
    0.06
    Activations Density 0.034%

    No Known Activations