© 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. 940
    Prev
    Next
    INDEX
    Explanations

    Ragnar, Edgar, Wagner, Madden

    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
     rÃłng
    -0.16
    ington
    -0.11
    sWith
    -0.10
    ness
    -0.10
    gear
    -0.09
     же
    -0.09
    ingly
    -0.09
    owing
    -0.09
     dÃłng
    -0.09
    åĭĥåĭĥ
    -0.09
    POSITIVE LOGITS
    ê¼´
    0.09
    пÑĢоÑģ
    0.09
    acho
    0.09
    çIJĨä¼ļ
    0.09
    urally
    0.09
    丼æŀĹ
    0.08
    à¹Į
    0.08
    /OR
    0.08
    istr
    0.08
    æĸ°å¾ģç¨ĭ
    0.08
    Activations Density 0.002%

    No Known Activations