© 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. 56531
    Prev
    Next
    INDEX
    Explanations

    don't understand why

    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
    ä¸įæŃ»
    -0.10
    ä¸įå°ij人
    -0.09
    æĿ¯
    -0.09
    åįĥæĸ¹çĻ¾è®¡
    -0.09
    gs
    -0.09
    大æ¦Ĥ
    -0.08
    æĹ©åľ¨
    -0.08
    ä¼ļç»Ļ
    -0.08
    æ¯ıå¹´éĥ½
    -0.08
    ParameterValue
    -0.08
    POSITIVE LOGITS
    æѤåĪ»
    0.09
    ucked
    0.09
    è¿Ļä¸ĢåĪ»
    0.08
    ળ
    0.08
    _FE
    0.08
    书éĿ¢
    0.08
    UILT
    0.08
     зап
    0.08
    /dist
    0.08
    è¿Ļ对
    0.08
    Activations Density 0.486%

    No Known Activations