© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3.5-4B
    5. 15-RES-MATRYOSHKA-65K
    7. 22499
    Prev
    Next
    INDEX
    Explanations

    new + common class names

    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
    Negative Logits
    еÑĢе
    -0.07
    ئÛĮ
    -0.06
    Bợi
    -0.06
    edan
    -0.06
    ãĥ¼ãĤ¸ãĥ¼
    -0.05
    yen
    -0.05
     erog
    -0.05
    eger
    -0.05
    acon
    -0.05
    ilion
    -0.05
    POSITIVE LOGITS
    à¸Ķำ
    0.06
    æ²īæ²ī
    0.05
     dalamnya
    0.05
    оÑĤÑĮ
    0.05
    onth
    0.05
    usted
    0.05
    tır
    0.05
    ALLY
    0.05
    åıĭ们
    0.05
    Selective
    0.05
    Activations Density 0.013%

    No Known Activations