© 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. 47404
    Prev
    Next
    INDEX
    Explanations

    writing code or text

    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
    æľīæĿ¡ä»¶
    -0.08
    å¤ļæł·
    -0.08
    æľµ
    -0.08
     Wonderland
    -0.08
     Alien
    -0.08
    ritch
    -0.08
    ç»Ļ人
    -0.08
    æ´¾
    -0.08
    ä½IJ
    -0.07
    åĨ³
    -0.07
    POSITIVE LOGITS
    å°±æĥ³
    0.13
    èį¤
    0.10
    çĩĥçĥ§
    0.10
    æĪĺæĸĹ
    0.10
    è¡·
    0.10
    çĸµ
    0.09
    èĮ¬
    0.09
    åIJĿ
    0.09
    ä¸įä¸ĭ
    0.09
    è´Łè´£
    0.09
    Activations Density 0.197%

    No Known Activations