© 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. 4733
    Prev
    Next
    INDEX
    Explanations

    breast cancer, bicycle helmet, tooth care, electoral campaigns, pig welfare, garage door

    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¸ĢéĿ¢
    -0.11
    arness
    -0.10
    æĺİæĹ¥
    -0.09
    çĭ»
    -0.09
    usz
    -0.09
     Теп
    -0.09
    涣
    -0.09
    struments
    -0.09
    ä¸ĩ家
    -0.08
    tones
    -0.08
    POSITIVE LOGITS
    -related
    0.10
    å¼ºåĽ½
    0.08
    ä¹ĭ家
    0.08
    -specific
    0.08
    è¡Ķ
    0.08
    uhl
    0.08
    缸åħ³çļĦ
    0.08
     ê´Ģ볨
    0.08
    ette
    0.08
     illusion
    0.08
    Activations Density 0.929%

    No Known Activations