© 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. 33903
    Prev
    Next
    INDEX
    Explanations

    unavoidable, impossible, unremovable, too much

    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
    纵深
    -0.08
    æĿ¾
    -0.08
     bere
    -0.08
     Cust
    -0.08
     Reserved
    -0.08
    æī¾åĽŀ
    -0.08
    ä¸Ĭå²Ĺ
    -0.08
    å®ģæĦ¿
    -0.08
    é¢ĦçķĻ
    -0.08
    顺
    -0.08
    POSITIVE LOGITS
    æĹłæ³ķ
    0.14
    ä¸įåı¯ä»¥
    0.13
    ä¸įå¾Ĺä¸į
    0.13
    没æľīåĬŀæ³ķ
    0.13
     inev
    0.13
     нелÑĮзÑı
    0.12
    ä¸įåı¯æĪĸ
    0.12
    没æ³ķ
    0.12
    ä¸įèĥ½
    0.12
    éĢĥèĦ±
    0.12
    Activations Density 0.104%

    No Known Activations