© 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. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 24005
    Prev
    Next
    INDEX
    Explanations

    self- prefixes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    adamkarvonen/qwen3-32b-saes/saes_Qwen_Qwen3-32B_batch_top_k/resid_post_layer_32
    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
    èľĢ
    -0.12
    çİ°è¡Į
    -0.10
    èIJ±
    -0.10
    _dom
    -0.09
    ç½ijç«Ļåľ°åĽ¾
    -0.09
    Dom
    -0.09
    æ»ĵ
    -0.09
    ä¸Ģå¤ľ
    -0.09
    ÖĢ
    -0.09
    leh
    -0.09
    POSITIVE LOGITS
    verständlich
    0.13
    åĬ¨çĶŁæĪIJ
    0.12
    ständ
    0.10
    (Self
    0.10
    self
    0.10
    èĩªå¾ĭ
    0.10
    ä½łèĩªå·±
    0.10
    ridge
    0.09
     بÙĨÙģس
    0.09
    /self
    0.09
    Activations Density 0.128%

    No Known Activations