© 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. 49398
    Prev
    Next
    INDEX
    Explanations

    ultra-low, high, wide

    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
    讥
    -0.12
    FilterWhere
    -0.10
    /cms
    -0.09
    avax
    -0.09
    å¾Īä¸įéĶĻ
    -0.09
    edImage
    -0.09
    resse
    -0.09
    åĩłå¼ł
    -0.09
    å¾Īæ¸ħæ¥ļ
    -0.09
    éĻIJãĤĬ
    -0.09
    POSITIVE LOGITS
     \$
    0.09
    ernel
    0.08
    æŃ¤åŁºç¡Ģä¸Ĭ
    0.08
    ausal
    0.08
    umper
    0.08
    ecer
    0.08
    à¸ļรร
    0.08
    sheet
    0.08
    æŀģé«ĺ
    0.08
    æįŀ
    0.08
    Activations Density 0.078%

    No Known Activations