© 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. 39819
    Prev
    Next
    INDEX
    Explanations

    Clare, declare, present invention

    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
    æķĿ
    -0.10
    æĬĬæĮģ
    -0.09
    çĤ¹åįĬ
    -0.09
    lds
    -0.09
    éķĬ
    -0.08
    éĺģ
    -0.08
    让人们
    -0.08
    ç¼ĸåĪ¶
    -0.08
    ECH
    -0.08
    褶
    -0.08
    POSITIVE LOGITS
    edly
    0.10
    erior
    0.09
    ohl
    0.09
     Husband
    0.09
    åĻª
    0.09
    cene
    0.09
     Reno
    0.09
    oria
    0.09
    olian
    0.09
    istani
    0.08
    Activations Density 0.001%

    No Known Activations