© 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. 57504
    Prev
    Next
    INDEX
    Explanations

    new announcements or studies

    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
     ëĭ¹ìĭľ
    -0.10
    å½ĵæŶçļĦ
    -0.10
    管çIJĨå±Ģ
    -0.09
    oose
    -0.09
    麽
    -0.09
    aat
    -0.09
    åĢĴ
    -0.08
    ä»ĸæĽ¾
    -0.08
    æŁ
    -0.08
    alink
    -0.08
    POSITIVE LOGITS
    è¿ijæĹ¥
    0.12
    æŃ£åľ¨
    0.12
    ç»Īäºİ
    0.11
    æݨåĩºäºĨ
    0.10
    ä¸įè¿ľ
    0.10
    æĿ¥åĪ°äºĨ
    0.10
    ä¸Ģ个æĸ°çļĦ
    0.10
    æŃ£åľ¨è¿Ľè¡Į
    0.10
    缮åīįæŃ£åľ¨
    0.10
    æݨåĩºçļĦ
    0.10
    Activations Density 0.405%

    No Known Activations