© 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. 40214
    Prev
    Next
    INDEX
    Explanations

    its own

    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
    èĢģ人家
    -0.10
    åĩłåı¥
    -0.10
    è¦ģéĹ»
    -0.10
    åħ¶å®ĥéĹ®é¢ĺ
    -0.10
    éĤ£è¾¹
    -0.10
    çļĦåį°è±¡
    -0.10
    çļĦ主é¢ĺ
    -0.10
    èĩªçIJĨ
    -0.09
    çħ§æł·
    -0.09
    家åĸ»æĪ·æĻĵ
    -0.09
    POSITIVE LOGITS
    nia
    0.09
    embre
    0.09
    idences
    0.09
    onder
    0.08
    arl
    0.08
    ridge
    0.08
    zia
    0.08
     Cinder
    0.08
    chal
    0.08
    /original
    0.08
    Activations Density 0.021%

    No Known Activations