© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 29434
    Prev
    Next
    INDEX
    Explanations

    connectors (and, or, but, matter)

    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
    Negative Logits
    ä¸ĢæĹı
    -0.10
     cpt
    -0.10
    åıįåĢĴ
    -0.10
    çĬ¶åħĥ
    -0.09
    èĢĮçİ°åľ¨
    -0.09
    è¹Ĥ
    -0.09
    åĽ°éĽ£
    -0.09
    æĸ°è¥¿
    -0.09
    èĥĦ
    -0.09
    ãģ¾ãģĹãģ¦
    -0.09
    POSITIVE LOGITS
    顺
    0.09
    ä¼´éļıçĿĢ
    0.08
    ä¼´
    0.08
     pit
    0.08
    仲
    0.08
     accompanying
    0.08
    éłĨ
    0.07
    顺便
    0.07
    æ¯ı个
    0.07
    bard
    0.07
    Activations Density 0.012%

    No Known Activations