© 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. 27676
    Prev
    Next
    INDEX
    Explanations

    print editions and versions

    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
    éļıæŶéļı
    -0.09
    é£İåIJ¹
    -0.09
    åĨĻçĿĢ
    -0.09
    åŀł
    -0.08
    缩
    -0.08
    æijĦåĥı头
    -0.08
    ê¾¼
    -0.08
    ä¸īæĺ¯
    -0.08
    æ¬ł
    -0.08
    饥
    -0.08
    POSITIVE LOGITS
    æ´ĭæ´ĭ
    0.09
    aspers
    0.08
    ÑĨÑĥ
    0.08
     Malk
    0.08
    å¹´åIJİ
    0.08
    æ¥
    0.08
    å¤ļç§įå½¢å¼ı
    0.08
    æ¼Ĥ
    0.08
    æī§åĭ¤
    0.08
    /cop
    0.08
    Activations Density 0.002%

    No Known Activations