© 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. 30050
    Prev
    Next
    INDEX
    Explanations

    part or stage completion

    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
    æ¾Ī
    -0.08
     biggest
    -0.08
    èĭ±æĸĩ
    -0.07
    enza
    -0.07
    妥
    -0.07
    èĽ©
    -0.07
    ä¸ĸçķĮåIJĦåĽ½
    -0.07
    æĦŁæħ¨
    -0.07
    -dashboard
    -0.07
     |_|
    -0.07
    POSITIVE LOGITS
    éļıæŶ
    0.11
    éļ¨æĻĤ
    0.09
     kê
    0.09
    å¤¸å¼ł
    0.09
    å¤ĩçĶ¨
    0.08
    æĪĺåĽ½
    0.08
    ä¹Łä¸įåı¯èĥ½
    0.08
    éģĵåħ·
    0.08
    å¾Īåı¯èĥ½
    0.08
    追
    0.08
    Activations Density 0.204%

    No Known Activations