© 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. 33999
    Prev
    Next
    INDEX
    Explanations

    seeking guidance or solutions

    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
    å¾Ĺ天
    -0.10
    ansa
    -0.09
    à¸Īำ
    -0.09
    æĮł
    -0.09
    è¡Ķ
    -0.08
     본
    -0.08
    éĻĢ
    -0.08
    æľĢ大çļĦ
    -0.08
    .inline
    -0.08
     Spit
    -0.08
    POSITIVE LOGITS
    æħ·æħ¨
    0.10
    åħijæį¢
    0.10
    æ³ī
    0.09
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    0.09
    æĸ°çļĦä¸Ģ
    0.09
     ÙģÙĬÙĩا
    0.09
    驯
    0.08
    åħ¬åĬŀ
    0.08
    åģļçļĦäºĭ
    0.08
    ä¼ĹæīĢ
    0.08
    Activations Density 0.126%

    No Known Activations