© 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. 21480
    Prev
    Next
    INDEX
    Explanations

    interface actions

    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
    Å¡te
    -0.06
    izer
    -0.06
    urge
    -0.05
    tee
    -0.05
     steal
    -0.05
    nant
    -0.05
    èĦ¸ä¸ĬçļĦ
    -0.05
     Beck
    -0.05
    åĶĩ
    -0.05
    fol
    -0.05
    POSITIVE LOGITS
    çľł
    0.06
    !”.
    0.06
    à¸ļาย
    0.06
    ëijIJ
    0.05
    ofile
    0.05
    à¹Ģลย
    0.05
    .cg
    0.05
    !".
    0.05
    achen
    0.05
    æĸ¹ç¨ĭ
    0.05
    Activations Density 0.001%

    No Known Activations