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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Qwen3-32B
    5. 32-RESID-BATCHTOPK-65K
    7. 52022
    Prev
    Next
    INDEX
    Explanations

    Bain, ismet

    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
    æłª
    -0.10
    $class
    -0.09
    -unit
    -0.09
     наиболее
    -0.09
    gie
    -0.09
    ãģĭãĤĬ
    -0.09
     ê¶
    -0.09
    ATAB
    -0.08
    人æĿĥ
    -0.08
    ús
    -0.08
    POSITIVE LOGITS
    åĬŁå¤«
    0.10
    eson
    0.09
    认
    0.09
    æľīä¸ĢäºĽ
    0.09
    æİĸ
    0.09
    é«Ģ
    0.08
    èµį
    0.08
    иÑĦ
    0.08
     unde
    0.08
    èģĶåIJĪ
    0.08
    Activations Density 0.001%

    No Known Activations