© 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. 34571
    Prev
    Next
    INDEX
    Explanations

    useful information

    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¸Ģ个éĹ®é¢ĺ
    -0.10
     expectations
    -0.10
    gars
    -0.10
    cade
    -0.09
    å½±åĵį
    -0.09
    èĨı
    -0.09
     Patel
    -0.09
    è¦ģæ±Ĥ
    -0.09
    ç¯Ŀ
    -0.09
    人工
    -0.09
    POSITIVE LOGITS
     ÃŃch
    0.13
    /use
    0.10
    两级
    0.10
    /help
    0.10
    ä¼łæĿ¥
    0.09
    7
    0.09
     helpful
    0.08
    8
    0.08
    裨
    0.08
    广大
    0.08
    Activations Density 0.096%

    No Known Activations