© 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. 16855
    Prev
    Next
    INDEX
    Explanations

    implementing, sharing, supporting, focusing, working, importance

    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
    uffs
    -0.09
     Pathfinder
    -0.08
    çĪĨåĩº
    -0.08
    ÙħÙĬد
    -0.08
     golden
    -0.08
    tre
    -0.08
    alie
    -0.08
    éķŃ
    -0.07
    IndexOf
    -0.07
    .stamp
    -0.07
    POSITIVE LOGITS
    éĹ»
    0.11
    é¾Ł
    0.10
    é½IJ
    0.10
    è²Ĥ
    0.10
    asset
    0.09
    æģĥ
    0.09
    æĬī
    0.09
    é½Ĭ
    0.09
    inator
    0.08
    AndWait
    0.08
    Activations Density 0.187%

    No Known Activations