© 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. 53455
    Prev
    Next
    INDEX
    Explanations

    excited about

    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
    athom
    -0.09
     Matthews
    -0.09
    åŃľ
    -0.09
    åIJµæŀ¶
    -0.09
    æĭĽæĬķæłĩ
    -0.09
    _DOM
    -0.08
     dra
    -0.08
    éĹ®æĪij
    -0.08
    .bias
    -0.08
    åı¸é©¬
    -0.08
    POSITIVE LOGITS
     rep
    0.10
    æĬĬèĩªå·±çļĦ
    0.10
    idl
    0.08
    ä¹Łæĺ¯éĿŀ常
    0.08
    ç²¾
    0.08
    éĿŀåĩ¡
    0.08
    æľªæĿ¥
    0.08
    èİ·å¾ĹæĽ´å¤ļ
    0.08
    ValuePair
    0.08
    ä¹Łå¾Ī
    0.08
    Activations Density 0.004%

    No Known Activations