© 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. 54605
    Prev
    Next
    INDEX
    Explanations

    domain descriptors

    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
    à°³
    -0.08
    waves
    -0.08
     Ñģобой
    -0.08
    umps
    -0.08
     compliments
    -0.07
    èĬ±åĽŃ
    -0.07
    ÑĪки
    -0.07
     Som
    -0.07
    TRA
    -0.07
    ections
    -0.07
    POSITIVE LOGITS
    çŃīæĸ¹éĿ¢çļĦ
    0.10
    大æĸ¹
    0.09
     Fried
    0.09
    çļĦ梦æĥ³
    0.09
    æĸ¹éĿ¢çļĦ
    0.09
    è¿Ļ座
    0.09
    åĬ£
    0.09
    æįIJçĮ®
    0.09
    æįIJåĬ©
    0.09
    å¼ķèµ·çļĦ
    0.09
    Activations Density 0.250%

    No Known Activations