© 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. 58661
    Prev
    Next
    INDEX
    Explanations

    drug approval

    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
    ç§ģ人
    -0.09
    ä¸¤å¼ł
    -0.08
    rips
    -0.08
    à°³
    -0.08
    çĴŁ
    -0.08
    æ¹ĥ
    -0.08
    irtual
    -0.08
    åIJ¯åıij
    -0.08
    å¢ĥ
    -0.08
     ste
    -0.08
    POSITIVE LOGITS
    æī¹åĩĨ
    0.25
    审æī¹
    0.17
    æŃ£å¼ı
    0.17
    æī¹å¤į
    0.17
    认åı¯
    0.16
     approve
    0.15
     approval
    0.15
    æł¸åĩĨ
    0.14
    åIJĮæĦı
    0.13
    审议éĢļè¿ĩ
    0.13
    Activations Density 0.116%

    No Known Activations