© 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. 62338
    Prev
    Next
    INDEX
    Explanations

    beginning of clauses

    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
    æݨä»ĭ
    -0.10
    ä¸ĢæĬ¹
    -0.09
    æĬĬä»ĸ
    -0.09
    çľĭäºĨä¸Ģ
    -0.09
    çľĭäºĨçľĭ
    -0.09
    ãĤĴãģĶ
    -0.08
    转åıij
    -0.08
    ä»ĸå·²ç»ı
    -0.08
    毫ä¸įçĬ¹è±«
    -0.08
    umper
    -0.08
    POSITIVE LOGITS
    åºľ
    0.09
    æĪĺæĸĹ
    0.08
    çķı
    0.08
    èĢķ
    0.08
    ownt
    0.08
    游æĪıä¸Ń
    0.08
    lya
    0.08
    æł¹æį®ä¸įåIJĮ
    0.08
    eward
    0.07
    aken
    0.07
    Activations Density 0.059%

    No Known Activations