© 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. 5539
    Prev
    Next
    INDEX
    Explanations

    edit, leave, or recyclables

    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
    éĢģå¾Ģ
    -0.09
    åIJĮä¸Ģ个
    -0.09
    åĽ¢ç»ĵ
    -0.08
    çķĻå®Ī
    -0.08
    èµĸ以çĶŁåŃĺ
    -0.08
    تÙĪاصÙĦ
    -0.08
    ç§ijæĻ®
    -0.08
    çĵľ
    -0.08
    ustry
    -0.08
    mpl
    -0.08
    POSITIVE LOGITS
    è®°
    0.10
    èĸ®
    0.09
    ç²®
    0.09
    ä»Ĩ
    0.09
    å¤ĸåĩº
    0.08
    æīĭèħķ
    0.08
     recording
    0.08
    tog
    0.08
    éĨĮ
    0.08
    è¿ĽæĿ¥
    0.08
    Activations Density 0.098%

    No Known Activations