© 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. 55619
    Prev
    Next
    INDEX
    Explanations

    changes and new developments

    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¸įéĹ´æĸŃ
    -0.10
    _indent
    -0.10
    ä¹ŁçŁ¥éģĵ
    -0.09
    ç£ħ
    -0.09
    ä¹Łä¸įæķ¢
    -0.09
    å¾ħéģĩ
    -0.09
    椽
    -0.09
    对å¾ħ
    -0.08
    积æŀģä½ľçĶ¨
    -0.08
    ä¸İä¸ŃåĽ½
    -0.08
    POSITIVE LOGITS
    æĶ¹åıĺ
    0.14
    åıĺå¾ĹæĽ´
    0.12
    æĶ¹åıĺäºĨ
    0.11
    æĶ¹è®Ĭ
    0.11
     change
    0.11
    åıĺåĮĸ
    0.11
    æĽ´æĶ¹
    0.10
    fäll
    0.10
    æĸ°çļĦ
    0.10
    æİł
    0.09
    Activations Density 0.529%

    No Known Activations