© 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. 62967
    Prev
    Next
    INDEX
    Explanations

    code or legal formatting

    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
    è·³è·ĥ
    -0.09
     bakım
    -0.09
    SHOT
    -0.09
    æ¼ı
    -0.09
    azu
    -0.09
    æ¶īæ¡Ī
    -0.08
    失è°ĥ
    -0.08
    åľ¨æ¸¸æĪıä¸Ń
    -0.08
     кÑĢÑĥ
    -0.08
    .att
    -0.08
    POSITIVE LOGITS
    åIJįçīĩ
    0.13
    åĿĬ
    0.10
    çŃ¹çłģ
    0.09
    ä¸ĬéŨ
    0.09
    ç»ıèIJ¥èĮĥåĽ´
    0.09
    ä½ľåĿĬ
    0.09
    é£İæ³¢
    0.08
    åIJ¯çĶ¨
    0.08
    ede
    0.08
    éĻįèĩ³
    0.08
    Activations Density 0.003%

    No Known Activations