© 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. 26108
    Prev
    Next
    INDEX
    Explanations

    adhesives and tape

    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
    åĩĽ
    -0.11
    çļĦæł¹æľ¬
    -0.09
    èµĺ
    -0.09
    ä»»
    -0.09
    æ°´éĩı
    -0.09
     BT
    -0.08
    ãģķãģĦ
    -0.08
    ä¸ĭåİ»
    -0.08
     Harris
    -0.08
    BT
    -0.08
    POSITIVE LOGITS
     tape
    0.11
     tapes
    0.11
     Tape
    0.10
     taped
    0.10
    tape
    0.09
    天çĦ¶
    0.09
    æĿ¡
    0.09
    ester
    0.09
    éģ®
    0.08
    椽
    0.08
    Activations Density 0.118%

    No Known Activations