© 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. 22354
    Prev
    Next
    INDEX
    Explanations

    to be verb followed by action/state

    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
    erten
    -0.06
    ä¸įä½ıçļĦ
    -0.06
    .Sm
    -0.06
     '),
    -0.06
    äºĨä¸Ģè·³
    -0.06
    ĮĢ
    -0.06
    .Cho
    -0.06
    æĿŃ
    -0.06
    icult
    -0.06
     *))
    -0.06
    POSITIVE LOGITS
    te
    0.05
    orat
    0.05
    o
    0.05
    оби
    0.05
    /
    0.05
     freaking
    0.05
    ork
    0.05
    åľ¨è¿Ļ个
    0.05
    à§Į
    0.05
     /
    0.05
    Activations Density 0.014%

    No Known Activations