© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 20174
    Prev
    Next
    INDEX
    Explanations

    best ever

    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
    çļĦä¸ĢäºĽ
    -0.10
    æĥħåĨµè¿Ľè¡Į
    -0.09
    çļĦä¸įåIJĮ
    -0.09
    è¿ĺæľīä¸ĢäºĽ
    -0.08
    ãĤ¹ãĥĿ
    -0.08
    è§Ī
    -0.08
    sters
    -0.08
    Spoiler
    -0.08
    è´£ç¼ĸ
    -0.08
     Mastery
    -0.08
    POSITIVE LOGITS
     ever
    0.13
    ä¹ĭä¸Ģ
    0.13
     EVER
    0.12
    Ever
    0.11
    ä¸ĸçķĮä¸Ĭ
    0.11
     imaginable
    0.10
     Ever
    0.10
    ever
    0.10
    åħ¶æ¬¡æĺ¯
    0.10
     Ive
    0.09
    Activations Density 0.148%

    No Known Activations