© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 49361
    Prev
    Next
    INDEX
    Explanations

    Russian or Cyrillic fragments

    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
     introduction
    -0.09
    æĽ´å¤ļ
    -0.09
    -major
    -0.09
    ä¸Ģåı£
    -0.09
    å¼ķè¿Ľ
    -0.08
    åĩĽ
    -0.08
    æĺ¯ä»İ
    -0.08
     Innoc
    -0.08
    аÑĤÑĥÑĢ
    -0.08
    ophage
    -0.08
    POSITIVE LOGITS
    æıŃéľ²
    0.10
    æĥ³èµ·æĿ¥
    0.09
    代
    0.09
    æľīåIJį
    0.09
    dn
    0.09
    å¨ģèĥģ
    0.09
    æłĹ
    0.08
    å®īå±ħ
    0.08
    主管
    0.08
    é¢Ĩ
    0.08
    Activations Density 0.190%

    No Known Activations