© 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. 32440
    Prev
    Next
    INDEX
    Explanations

    connected, linked, related

    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
    éĢ¸
    -0.09
     eBook
    -0.08
    ROWS
    -0.08
    ä»Ĭ
    -0.08
     Dix
    -0.08
    ITS
    -0.08
    åıĬåħ¶
    -0.08
    çIJĨæĢ§
    -0.08
    rooms
    -0.08
    åĨįåĬłä¸Ĭ
    -0.08
    POSITIVE LOGITS
    ç´§å¯Ĩ
    0.17
     ÑģвÑıзан
    0.16
     ìĹ°ê²°
    0.15
    缸è¿ŀ
    0.15
     connected
    0.14
    éĵ¾æİ¥
    0.14
    å¯ĨåĪĩ
    0.14
     linked
    0.13
    è¿ŀæİ¥
    0.13
    ãģ¤ãģª
    0.13
    Activations Density 0.208%

    No Known Activations