© 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. 53722
    Prev
    Next
    INDEX
    Explanations

    code, script, or metadata delimiters

    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
    仨
    -0.10
    lest
    -0.09
    éĥ½åĸľæ¬¢
    -0.09
    çī¹çĤ¹æĺ¯
    -0.09
    é¹Ī
    -0.08
    æĻı
    -0.08
    骥
    -0.08
    éĹ®é¢ĺæĺ¯
    -0.08
    á»ıi
    -0.08
    æѤåīį
    -0.08
    POSITIVE LOGITS
    /is
    0.11
    æľ¬èº«çļĦ
    0.10
    æłª
    0.09
    /how
    0.09
    ataka
    0.09
    æľ¬èº«
    0.08
    Isl
    0.08
    çľ¼åħī
    0.08
     ÑģобÑģÑĤвенно
    0.08
    âĸ¹
    0.08
    Activations Density 0.254%

    No Known Activations