© 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. 12255
    Prev
    Next
    INDEX
    Explanations

    last step or finishing

    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
    Negative Logits
    ä¸įæŃ¢
    -0.10
    篱
    -0.09
    еÑĢж
    -0.09
    çĥĪ士
    -0.09
    uppe
    -0.09
    аÑĨиÑı
    -0.09
    æľīè¿Ļæł·çļĦ
    -0.09
    çĨ¬
    -0.08
    åŃĿ
    -0.08
    è§ĦåĪĴ建设
    -0.08
    POSITIVE LOGITS
    çݯèĬĤ
    0.11
     final
    0.10
    å«ģ
    0.10
    ÐĽÐŀ
    0.09
    aker
    0.09
    immers
    0.09
    æľĢåIJİ
    0.09
    final
    0.09
    æ¥Ŀ
    0.09
    åĴĮå®ĮåĸĦ
    0.09
    Activations Density 0.178%

    No Known Activations