© 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. 27671
    Prev
    Next
    INDEX
    Explanations

    months ago, now releasing, treatment and

    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
    Colour
    -0.10
    伦æķ¦
    -0.10
     Colour
    -0.09
     Heights
    -0.09
    åĽ´æ£ĭ
    -0.08
     ÙģÙī
    -0.08
    èĬĿ
    -0.08
    èĢĹ
    -0.08
    colo
    -0.08
    iev
    -0.08
    POSITIVE LOGITS
    æĪIJ为ä¸ŃåĽ½
    0.09
    åĦ¿
    0.09
    驾
    0.09
    $return
    0.09
     TOR
    0.08
     Ñĥдален
    0.08
    åħĴ
    0.08
    æĪIJå°±æĦŁ
    0.08
     sph
    0.08
    ä¸ĢèĤ¡
    0.08
    Activations Density 1.540%

    No Known Activations