© 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. 3075
    Prev
    Next
    INDEX
    Explanations

    strong

    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
    ahas
    -0.09
    åIJĪçIJĨæĢ§
    -0.09
    æ¯ķ
    -0.09
    大æĸ¹
    -0.09
    çķ¢
    -0.08
    æĶ¹éĿ©åıijå±ķ
    -0.08
    çįIJ
    -0.08
    ktop
    -0.08
    åIJĪæł¼
    -0.08
    oins
    -0.08
    POSITIVE LOGITS
    -strong
    0.15
    strong
    0.15
     strong
    0.14
    ,strong
    0.13
    Strong
    0.13
    å¾Ī强
    0.13
    å¼·ãģĦ
    0.13
    æľīåĬĽ
    0.12
     stronger
    0.12
     strongest
    0.12
    Activations Density 0.111%

    No Known Activations