© 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. 47590
    Prev
    Next
    INDEX
    Explanations

    before

    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
    belt
    -0.09
    竺
    -0.09
    çļĦç»ıåİĨ
    -0.09
    åį°è±¡
    -0.08
    ç»®
    -0.08
    _OLD
    -0.08
    ç»ıåİĨäºĨ
    -0.08
     пеÑĢевод
    -0.08
    å®Ľå¦Ĥ
    -0.08
    çļĦåı¯èĥ½æĢ§
    -0.08
    POSITIVE LOGITS
    ught
    0.10
    è¿Ļä¸ĢåĪĩ
    0.09
    aki
    0.09
    åĩºçİ°
    0.09
    é£İåIJ¹
    0.09
    ä¸ĢåĪ»
    0.09
     sake
    0.09
    ä¸Ģåľº
    0.08
    hand
    0.08
     Marina
    0.08
    Activations Density 0.048%

    No Known Activations