© 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. 48509
    Prev
    Next
    INDEX
    Explanations

    do not share email

    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
    禳
    -0.10
     interchange
    -0.08
    _gc
    -0.08
     m
    -0.08
    ERE
    -0.08
    æij©
    -0.07
     mo
    -0.07
    ,
    -0.07
    é»ĺ
    -0.07
     set
    -0.07
    POSITIVE LOGITS
    ä»İæľª
    0.11
    æ°¸è¿ľä¸įä¼ļ
    0.11
    åıªæĥ³
    0.10
    ä¸įä¼ļåĨį
    0.10
    inem
    0.10
    ç»Ŀ对ä¸įä¼ļ
    0.10
    çªģåĩ»
    0.10
    åıªä¼ļ
    0.09
    ulo
    0.09
    ä»İæĿ¥æ²¡æľī
    0.09
    Activations Density 0.050%

    No Known Activations