© 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. 57670
    Prev
    Next
    INDEX
    Explanations

    too + excessive/insufficient quantity/degree

    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¸Ģå®ļç¨ĭ度
    -0.11
     ìķĦ주
    -0.10
    ç¥Ŀ
    -0.10
    è¾ĥå¤ļ
    -0.10
     very
    -0.10
    æľºéģĩ
    -0.10
    æĽ´å¥½åľ°
    -0.09
    ượt
    -0.09
    ãĤĪãĤĬ
    -0.09
    éĿŀ常
    -0.09
    POSITIVE LOGITS
    好åIJĹ
    0.11
    æł·çļĦ
    0.10
    zeitig
    0.10
    IN
    0.09
    åij¨å¹´
    0.09
    é«ĺçļĦ
    0.09
    麻çĥ¦
    0.09
    åįķåįķ
    0.09
    ifi
    0.09
    thing
    0.09
    Activations Density 0.016%

    No Known Activations