© 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. 8652
    Prev
    Next
    INDEX
    Explanations

    independently, coordinated, approach

    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
    é«ĺ级
    -0.08
    æĬ¥åijĬæľŁåĨħ
    -0.08
    åĽ¾ä¸º
    -0.08
    积æŀģåıĤä¸İ
    -0.08
    ç¬ı
    -0.08
    _tm
    -0.08
    åİŁæłĩé¢ĺ
    -0.08
    ogl
    -0.08
    æĭ±
    -0.08
    å¿«æĬ¥
    -0.08
    POSITIVE LOGITS
    浪费
    0.11
    å½¼æѤ
    0.11
    Duplicate
    0.10
    åIJĦèĩªçļĦ
    0.10
     duplication
    0.10
     duplicated
    0.10
    åIJĦèĩª
    0.10
    ä¹±
    0.10
    诸侯
    0.10
    åĮĨåĮĨ
    0.09
    Activations Density 0.070%

    No Known Activations