© 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. 63308
    Prev
    Next
    INDEX
    Explanations

    years and ordinals

    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
     fare
    -0.09
    äºļ马
    -0.09
    积æ·Ģ
    -0.09
    ä¸ĢçĽ´åľ¨
    -0.09
    ina
    -0.08
    å¿ĥåºķ
    -0.08
    åѦç§ij
    -0.08
    çĥŁçģ«
    -0.08
    æĸ¹éĿ¢
    -0.08
     discipline
    -0.08
    POSITIVE LOGITS
    è¿Ļ个æĹ¶ä»£
    0.14
    çļĦæĹ¶ä»£
    0.14
    æĹ¶ä»£
    0.13
    æĸ°æĹ¶ä»£
    0.13
    çݰ代社ä¼ļ
    0.12
     moderne
    0.12
    æĻĤ代
    0.11
     اÙĦÙĤرÙĨ
    0.11
    å½ĵä»Ĭ
    0.11
    æĹ¶ä»£çļĦ
    0.11
    Activations Density 0.080%

    No Known Activations