© 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. 12098
    Prev
    Next
    INDEX
    Explanations

    do so

    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
    Not in Any Lists

    No Comments

    Negative Logits
     unf
    -0.09
     Pix
    -0.08
    è¿ĺåĮħæĭ¬
    -0.08
    è¿ĺä¼ļ
    -0.08
    ì¶ľ
    -0.08
    Pix
    -0.07
    Äĥm
    -0.07
    æĶ¶èİ·
    -0.07
    ä»İå°ı
    -0.07
    ä»ĸ对
    -0.07
    POSITIVE LOGITS
    è¿Ļæł·åģļ
    0.23
    è¿Ļä¹Īåģļ
    0.22
     doing
    0.17
     Doing
    0.16
    doing
    0.14
     done
    0.14
    åģļçļĦ
    0.14
    Doing
    0.14
    æŃ¤ä¸¾
    0.13
    æĢİä¹Īåģļ
    0.13
    Activations Density 0.138%

    No Known Activations