© 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. 62702
    Prev
    Next
    INDEX
    Explanations

    quilting and stitching

    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
    è¿Ļæł·æīįèĥ½
    -0.09
    æĭĹ
    -0.09
    ä¼´
    -0.09
    NDER
    -0.08
    offline
    -0.08
    èĽĭ
    -0.08
    éĢ¢
    -0.08
    _CLASSES
    -0.08
    ewire
    -0.08
    vertising
    -0.08
    POSITIVE LOGITS
    quil
    0.10
     quilt
    0.09
     Musk
    0.09
    allee
    0.08
    æĹĮ
    0.08
    å¹Ľ
    0.08
    umbs
    0.08
     quil
    0.08
    ç¼
    0.08
    gend
    0.08
    Activations Density 0.028%

    No Known Activations