© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 3306
    Prev
    Next
    INDEX
    Explanations

    remove or removal

    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
    rim
    -0.11
    erman
    -0.10
    umont
    -0.10
    æľ¬ç½ij
    -0.10
    çĦ¶æĺ¯
    -0.10
    ute
    -0.09
    acies
    -0.09
    mut
    -0.09
    /from
    -0.09
    rase
    -0.08
    POSITIVE LOGITS
    ä¸įäºĨ
    0.11
    äºĨ大éĩıçļĦ
    0.10
    enda
    0.09
    /comment
    0.09
    /add
    0.09
    /install
    0.09
    /Add
    0.09
     rÃłng
    0.09
    åĿļä¿¡
    0.09
    oral
    0.09
    Activations Density 0.061%

    No Known Activations