© 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. 12404
    Prev
    Next
    INDEX
    Explanations

    what a or what better

    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
    èݽ
    -0.10
     ÐĴедÑĮ
    -0.10
    _PROC
    -0.10
    æ¯ķç«Łæĺ¯
    -0.09
    clamp
    -0.09
    æ¯ķç«Ł
    -0.09
    åıijå±ķåİĨç¨ĭ
    -0.09
    owan
    -0.09
    Eventually
    -0.09
    æĪ³
    -0.09
    POSITIVE LOGITS
    åIJĦé¡¹å·¥ä½ľ
    0.10
    åĢºæĿĥ
    0.08
     ка
    0.08
     prescribe
    0.08
    åįļ主
    0.08
    为客æĪ·
    0.08
     stron
    0.08
    ieten
    0.08
    åĪļåĪļ
    0.08
    åĢį
    0.07
    Activations Density 0.131%

    No Known Activations