© 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. 26095
    Prev
    Next
    INDEX
    Explanations

    behind or following

    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
    loud
    -0.09
    è¾£
    -0.09
    eka
    -0.09
    çĩĬ
    -0.09
    æĮ¹
    -0.08
    åķIJ
    -0.08
    è´µ
    -0.08
    çIJª
    -0.08
    èĹIJ
    -0.08
    erno
    -0.08
    POSITIVE LOGITS
    身åIJİ
    0.15
    è·Łè¿Ľ
    0.15
     behind
    0.14
    åIJİéĿ¢
    0.14
    Behind
    0.13
     Behind
    0.13
    åIJİéĿ¢çļĦ
    0.12
    è·ŁçĿĢ
    0.11
    å°¾
    0.11
    ç´§è·Ł
    0.11
    Activations Density 0.062%

    No Known Activations