© 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. 18627
    Prev
    Next
    INDEX
    Explanations

    comparisons of efficiency and cost

    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
    éĢĤ度
    -0.09
    åıĺå½¢
    -0.09
     yếu
    -0.09
    è¿ĩåī©
    -0.08
    å¾Īæ·±
    -0.08
    cheng
    -0.08
    çĶ¨éĩı
    -0.08
    身ä¸Ĭ
    -0.08
    à¸Ńà¸ĩ
    -0.08
    ike
    -0.08
    POSITIVE LOGITS
    è¿Ļç§įæĸ¹å¼ı
    0.11
    æĽ´å®¹æĺĵ
    0.11
     easier
    0.10
    æĽ´æľī
    0.10
    æĽ´èĥ½
    0.09
    æľ«ç«¯
    0.09
    æĽ´éĢĤåIJĪ
    0.09
    åĴĮ社ä¼ļ
    0.09
     McM
    0.09
    çļĦæĸ¹å¼ı
    0.09
    Activations Density 0.073%

    No Known Activations