© 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. 673
    Prev
    Next
    INDEX
    Explanations

    representative sample or people

    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
     извеÑģÑĤ
    -0.12
    lege
    -0.10
    rego
    -0.10
    æ¯Į
    -0.09
    ä¸įæĩĪ
    -0.09
    .localization
    -0.09
    缮
    -0.09
    ustum
    -0.09
    ä¾Ŀ
    -0.09
    éĢĨ
    -0.09
    POSITIVE LOGITS
    代表æĢ§
    0.33
    代表
    0.31
     representative
    0.29
    代表çĿĢ
    0.24
    为代表
    0.23
     Representative
    0.23
     representatives
    0.21
    represent
    0.20
    Represent
    0.18
     represent
    0.18
    Activations Density 0.102%

    No Known Activations