© 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. 27861
    Prev
    Next
    INDEX
    Explanations

    polygonal shapes

    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
    åIJįçīĮ
    -0.09
    人æ°ijéĵ¶è¡Į
    -0.09
     Wand
    -0.09
    /rc
    -0.09
    åºĶ对
    -0.09
    ffield
    -0.08
    arious
    -0.08
    uously
    -0.08
    ign
    -0.08
     Pill
    -0.08
    POSITIVE LOGITS
    åĦ¿
    0.11
    åħĴ
    0.09
     meshes
    0.09
    äºĨ大éĩıçļĦ
    0.09
    æĹłæķ°
    0.09
    stin
    0.08
    æ¼ıæ°´
    0.08
     kuk
    0.08
    å°ıå·§
    0.08
     Din
    0.08
    Activations Density 0.134%

    No Known Activations