© 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. 50615
    Prev
    Next
    INDEX
    Explanations

    recent study 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
    hay
    -0.09
     ÑģобÑģÑĤвенно
    -0.09
    дей
    -0.09
    æĺ¯ä¸Ģ件
    -0.08
    han
    -0.08
    -too
    -0.08
    çĤ¹å¤ļ
    -0.08
    DDL
    -0.08
    ENSE
    -0.08
    责任ç¼ĸè¾ij
    -0.08
    POSITIVE LOGITS
    çłĶ究表æĺİ
    0.26
    è°ĥæŁ¥æĺ¾ç¤º
    0.19
     studies
    0.17
     study
    0.15
    çłĶ究
    0.14
     recent
    0.14
    æľĢè¿ij
    0.13
    æĬ¥åijĬæĺ¾ç¤º
    0.12
    recent
    0.12
     Studies
    0.12
    Activations Density 0.273%

    No Known Activations