© 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. 50266
    Prev
    Next
    INDEX
    Explanations

    mostly

    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
    Spi
    -0.09
    ãģ¹ãģį
    -0.09
    aries
    -0.08
    许å¤ļ人
    -0.08
    éĢ¼
    -0.08
    èĥ½åĬĽå¼º
    -0.08
    rosis
    -0.08
    å¾Ĺèµ·
    -0.08
    ometer
    -0.08
    ãģĹãģ¾ãģĦ
    -0.08
    POSITIVE LOGITS
    éĽĨä¸Ńåľ¨
    0.13
    æŶåĢĻ
    0.13
    åģľçķĻåľ¨
    0.10
    ä½ĵçİ°åľ¨
    0.10
    cular
    0.10
    amate
    0.10
    /all
    0.09
    ocked
    0.09
    ç¢Į
    0.09
    æ¿©
    0.09
    Activations Density 0.015%

    No Known Activations