© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 22178
    Prev
    Next
    INDEX
    Explanations

    attributed information source

    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
    ä¸İåħ¶ä»ĸ
    -0.11
    å°ĨåĨĽ
    -0.11
    满äºĨ
    -0.09
    åŃ½
    -0.09
    é»ijé¾Ļ
    -0.08
    çĦ¯
    -0.08
    åħļçļĦé¢Ĩ导
    -0.08
     bound
    -0.08
    æĹłç¼ĺ
    -0.08
     Rao
    -0.08
    POSITIVE LOGITS
    æ¶Īæģ¯ç§°
    0.11
    人士
    0.11
    tep
    0.10
    RIA
    0.09
    åªĴ
    0.09
    æĤī
    0.09
    éĿłè¿ij
    0.09
     пÑĢиÑĩ
    0.09
    åĽ´ç»ķ
    0.09
    æ§İ
    0.09
    Activations Density 0.012%

    No Known Activations