© 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. 62647
    Prev
    Next
    INDEX
    Explanations

    proper names and interjections

    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
    warts
    -0.09
    èĪľ
    -0.09
    party
    -0.08
    intage
    -0.08
    OLA
    -0.08
    çķĶ
    -0.08
    æĽ¿
    -0.08
    ãĥŃãĥ¼
    -0.08
    éļĶå£ģ
    -0.08
    _stderr
    -0.08
    POSITIVE LOGITS
    ä¼ĺèī¯
    0.11
    æ¸Ŀ
    0.10
    aller
    0.10
    essor
    0.09
    æ½¼
    0.09
    ä¸įå®Į
    0.09
    culator
    0.08
     Alonso
    0.08
    å«£
    0.08
    &p
    0.08
    Activations Density 0.033%

    No Known Activations