© 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. 12434
    Prev
    Next
    INDEX
    Explanations

    transferring or passing to

    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
    é¢Ĭ
    -0.09
     unp
    -0.09
    é£ĺ
    -0.09
    moid
    -0.08
    ToDo
    -0.08
    éļ³
    -0.08
    ä¸ĭæīĭ
    -0.08
    为ä¹ĭ
    -0.08
    borg
    -0.08
     гол
    -0.08
    POSITIVE LOGITS
    æĦŁæŁĵ
    0.13
    ä¼łå¯¼
    0.12
    ä¼łæŁĵ
    0.12
     infect
    0.11
    ä¼łéĢĴ
    0.11
    ä¼łè¾¾
    0.10
    onto
    0.09
     transferring
    0.09
     infectious
    0.09
    转移åĪ°
    0.09
    Activations Density 0.062%

    No Known Activations