© 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. 18112
    Prev
    Next
    INDEX
    Explanations

    create or make

    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
     оÑĤкÑĢ
    -0.10
    umu
    -0.10
    endale
    -0.10
    éĶ¹
    -0.09
    çļĦæĥħæĦŁ
    -0.09
    à¸Ķำà¹Ģà¸Ļ
    -0.09
    äºĶå®ĺ
    -0.09
    éĤ£åıª
    -0.09
    讫
    -0.09
    æĥħæĦŁ
    -0.08
    POSITIVE LOGITS
    ä»İä¸Ń
    0.12
     dara
    0.12
    ä¸Ģåī¯
    0.09
    èĭŁ
    0.09
    anna
    0.09
    æļ´
    0.09
    ly
    0.09
    çĤ¯
    0.09
    éĴĪ对æĢ§
    0.08
    符
    0.08
    Activations Density 0.284%

    No Known Activations