© 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. 5976
    Prev
    Next
    INDEX
    Explanations

    draft and beta versions

    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
    Not in Any Lists

    No Comments

    Negative Logits
    ä¸Ģ棵
    -0.09
    tie
    -0.08
    adt
    -0.08
    èł²
    -0.08
    æķ´æ´ģ
    -0.08
    éļľ
    -0.08
    åħ±åIJĮåĬªåĬĽ
    -0.08
    æİĴåĪĹ
    -0.08
     hare
    -0.08
    ç±Ģ
    -0.08
    POSITIVE LOGITS
    äºĭåıij
    0.10
    ypress
    0.10
    çĿ¹
    0.09
    æĺİå¹´
    0.09
    ç²Ĺç³Ļ
    0.09
    ä¸įä¹ħ
    0.09
    spath
    0.09
    ä»įå°Ĩ
    0.09
    èĤ¤
    0.09
    éķ¿æķĪæľºåĪ¶
    0.09
    Activations Density 0.077%

    No Known Activations