© 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. 40091
    Prev
    Next
    INDEX
    Explanations

    custom + descriptor

    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
    å¾Īæĺ¯
    -0.09
    çIJī
    -0.09
    robat
    -0.08
    oning
    -0.08
     оÑĩеÑĢед
    -0.08
    çıł
    -0.08
    ovy
    -0.08
    éĥ½å¾Ī
    -0.08
    veled
    -0.08
    otted
    -0.08
    POSITIVE LOGITS
    äºĭå®ľ
    0.10
    è©
    0.09
    arily
    0.09
     Minds
    0.09
    çķª
    0.09
    èµī
    0.08
    ç¨İè´¹
    0.08
    zed
    0.08
    -made
    0.08
    èĢIJçĶ¨
    0.08
    Activations Density 0.027%

    No Known Activations