© 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. 26145
    Prev
    Next
    INDEX
    Explanations

    and Louder, Everyone, andonment, Free, Strangling, Legislative

    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
    è´µ
    -0.09
    æĮ¤
    -0.09
     поÑĤ
    -0.09
    åıijå±ķæ½ľåĬĽ
    -0.08
    afi
    -0.08
    ç¨ļ
    -0.08
    æıIJéĢŁ
    -0.08
    ouston
    -0.08
    è§Ĥä¼Ĺ
    -0.08
    éħį
    -0.08
    POSITIVE LOGITS
    è¿Ļ个è¯į
    0.12
    åIJįè¯į
    0.11
    è¯įæĿ¡
    0.10
    çĻ¾ç§ij
    0.09
     terms
    0.09
    æľ¯è¯Ń
    0.09
    è¯įè¯Ń
    0.09
    讹
    0.09
     term
    0.09
    USES
    0.09
    Activations Density 0.753%

    No Known Activations