© 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. 38750
    Prev
    Next
    INDEX
    Explanations

    matchup or rematch

    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
    ãģ»ãģĨ
    -0.09
    åįģåŃĹ
    -0.08
     contour
    -0.08
    ä¸ĢåĪ»
    -0.08
    åłĤ
    -0.08
    ãĤĢ
    -0.08
    راب
    -0.08
    /he
    -0.08
    oration
    -0.08
    gratis
    -0.08
    POSITIVE LOGITS
    sWith
    0.10
    åħ¨ä½ĵåijĺå·¥
    0.10
    pecific
    0.09
    éĺ´éĺ³
    0.09
    éªĦ
    0.09
    avy
    0.08
    åģĥ
    0.08
    ระหว
    0.08
     Slee
    0.08
    -specific
    0.08
    Activations Density 0.006%

    No Known Activations