© 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. 6156
    Prev
    Next
    INDEX
    Explanations

    good idea

    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
     ÑĢаÑģÑģмаÑĤÑĢива
    -0.10
    пиÑģÑĭва
    -0.10
    кладÑĭва
    -0.09
     Trip
    -0.09
    etal
    -0.08
     Freedom
    -0.08
    Trip
    -0.08
    trip
    -0.08
    çīĪ
    -0.08
    æİĽ
    -0.08
    POSITIVE LOGITS
    主æĦı
    0.23
     idea
    0.19
    çļĦæĥ³æ³ķ
    0.19
     Idea
    0.18
     ideas
    0.18
    念头
    0.17
    idea
    0.17
    æĥ³æ³ķ
    0.16
     Ideas
    0.15
     ÙģÙĥرة
    0.14
    Activations Density 0.202%

    No Known Activations