© 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. 5599
    Prev
    Next
    INDEX
    Explanations

    improve performance and outcomes

    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
    xcd
    -0.08
    å¾ģ
    -0.08
    ведениÑı
    -0.08
    éĹ®éĹ®
    -0.08
    entar
    -0.08
    лÑĥÑĪ
    -0.08
     flock
    -0.08
    -mf
    -0.08
     Relief
    -0.07
    ä¸į足
    -0.07
    POSITIVE LOGITS
    éĺ»æĮ¡
    0.14
    éģ®
    0.13
    æķ£çĥŃ
    0.12
    ç¾İè§Ĥ
    0.11
    æĶ¯æĴij
    0.11
    积èĵĦ
    0.10
    æĮ¡ä½ı
    0.10
    å¼Ģå¥ĸ
    0.09
    éļĶçĥŃ
    0.09
     Shade
    0.09
    Activations Density 0.353%

    No Known Activations