© 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. 6305
    Prev
    Next
    INDEX
    Explanations

    midnight swim, midnight blues

    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
     ÃŃch
    -0.11
    ðŁı»
    -0.11
    lessly
    -0.10
    ington
    -0.09
    ounty
    -0.09
    itional
    -0.09
    stead
    -0.09
    ascript
    -0.08
    ../
    -0.08
    entine
    -0.08
    POSITIVE LOGITS
    èĪ¬
    0.11
    èĪ¬çļĦ
    0.10
    wake
    0.09
    -blue
    0.09
    eer
    0.09
    avl
    0.09
    ä¸įæĸ¹ä¾¿
    0.09
    è¿Ļ两个
    0.09
    iating
    0.08
    iator
    0.08
    Activations Density 0.003%

    No Known Activations